Report No. 453a-TA Appraisal of Kigoma Rural Development Project FILE COPY Tanzania July 15, 1974 Agriculture and Rural Development Department Not for Public Use Document of the International Bank for Reconstruction and Development International Development Association This report was prepared for official use only by the Bank Group. It may not be published, quoted or cited without Bank Group authorization. The Bank Group does not accept responsibility for the accuracy or completeness of the report. CURRENCY EQUIVALENTS US$1.00 Tanzania Shilling 7.1 4 (Tsh) Tshl.00 US$0.14 WEIGHTS AND MEASURES Metric System 1 hectare (ha) 10 000 m2 . 2.46 acres 1 kilometer (km) 0.62 miles 1 square kilometer (kn2) 0 0.39 sq. miles = 100 ha 1 kilogram (kg) 2.20 pounds 1 liter (1) 0.26 gallons 1,000 kg = 1 metric ton = 0.98 long ton ABBREVIATIONS KILIMO Ministry of Agriculture MAJI Ministry of Water and Power MAJI/K Regional Water Engineers - Kigoma NBC National Bank of Commerce MMC National Milling Corporation TANU Tanganyika African National Union TCA Tanzania Cotton Authority TRDB Tanzania Rural Development Bank Union Kigoma Regional Cooperative Union Government Fiscal Year July 1 - June 30 TANZANIA KIGOMA RURAL DEVELOPMENT PROJECT Table of Contents Page No. SUMMARY AND CONCLUSIONS ................... ..... i-iii I. INTRODUCTION ................................. 1 II. BACKGROUND ........ ............................. 2 A. The Economy of Tanzania . .. 2 B. The Agricultural Sector ................. .. 2 C. Agricultural Marketing and Pricing ........ 3 D. Agricultural Credit ..... .................. 4 III. THE PROJECT AREA ............................... 5 A. Economy of Kigoma Region .................. 5 B. Regional Government and Administration 6 C. Rural Infrastructure ..... ................. 8 D. Agriculture and Fisheries ....... .......... 10 IV. THE PROJECT ....... ............................. 11 A. Brief Description ........ ................. 11 B. Detailed Features ..... .................... 12 C. Project Costs ...... ....................... 16 D. Financing ................................. 18 E. Procurement ...... ......................... 18 F. Disbursement . . . 19 G. Accounts and Audit . . ................ 20 H. Ecological and External Effects ........ . 20 V. ORGANIZATION AND MANAGEMENT ...... .............. 20 Regional Government .......... ............. 20 Village Planning . .21 Agriculture ............................... 21 Regional Cooperative Union ............... . 22 Tanzania Rural Development Bank ............ 23 'fable of Contents (Continued) Pate No. VI. PRODUCTION, MARKETS, FARMER BENEFITS AND FINANCIAL RESULTS ...... ........................ 24 Production, Yields and Marketed Surplus 24 Markets and Prices . .25 Farmer Benefits ............ ............... 25 Financial Results ........... .............. 25 Government Revenues and Outlays .. .......... 26 VII. ECONOMIC ANALYSIS ...... ........................ 27 VIII. RECOMMENDATIONS. :8 ANNEXES 1. Cooperatives Chart 1 Organization Chart of Rural Development Division Table 1 Affiliated Primary Societies and Produce Handled Table 2 Union Produce Purchases - 1971-72 2. Tanzania Rural Development Bank 3. Agriculture in Kigoma Region Table 1 Mean Monthly Rainfall Records Table 2 Estimated Labor Needs (Mandays) and Audibility per 3.5 Acre Farm Table 3 Herd Projection for Boran Nucleus Herd 4. Government and Administration in Kigoma Region 5. Education 6. Rural Water Supply 7. Public Health 8. Fisheries Appendix A Resources and Enviroument of Lake Tanganyika Appendix B Rough Comparative Data on Lake Tanganyika Fishing Techniques 9. Recommended Qualifications and Responsibilities of Staff Financed under the Project 10. Regional Planning, Project Preparation and Evaluation Table of Contents (Continued) 11. Cost Estimates Table 1 Cost Estimates by Year Table 2 Water Supply and Health Facilities Table 3 Education Table 4 Feeder Roads and Tsetse Clearing Table 5 Estimated Credit Needs for Villages Productive Infrastructure Table 6 Loans and Equity Contribution to Kigoma Cooperative Union Table 7 Rural Training Centers Table 8 Trial and Training Center and Livestock Pilot Project Table 9 Technical Assistance Table 10 Project Preparation and Evaluation 12. Estimated Schedule of IDA Disbursements 13. Procedures for Selection of Villages Appendix A Proforma Village Site Feasibility Report (VSFR) 14. Trading and Lending Procedures Chart 1 Trading and Lending Cycles Table 1 Kigoma Cooperative Union - Annual Cash Flow Table 2 Purchase and Sale of Agricultural Inputs Table 3 Purchase and Sale of Marketable Surplus Table 4 Unit Costs of Wholesaling Agricultural Inputs Table 5 Unit Costs of Wholesaling Marketable Surplus Table 6 Kigoma Cooperative Union - Operating Costs Table 7 Kigoma Cooperative Union - Transport Cash Flow Table 7(a)Kigoma Cooperative Union - Buildup of Motor Vehicles Table 7(b)Kigoma Cooperative Union- Lorry Account Table 8 Capital Investment and Repayment Table 9 Kigoma Cooperative Union - On-lending Activities Table 10 Annual Cash Flow of Primary Society 15. Crop Budgets Table 1 Maize Acre Budgets Table 2 Bean Acre Budgets Table 3 Cotton Acre Budgets Table 4 Groundnuts Acre Budgets Table 5 350-Family Village Production Model at Full Development Table 6 350-Family Village Acreage and Production Buildup Table 7 350-Family Village Marketable Surplus Buildup Table 8 Buildup of Regional Marketable Surplus Table 9 Credit and Input Needs by Volume and Cost Table 10 Volume and Costs.of Inputs for Demonstrations 16. Economic Analysis Table of Contents (Continued) Cr:ARTS 8466 - Overall Government Organization MAPS 10941 - Project Area 10942 - Population Density and Major Ecological Zones SUMMARY AND CONCLUSIONS i. Tanzania places high priority on agriculture and rural development, in keeping with the needs of the 90% of its population who live in rural areas and mainly work in traditional agriculture. Government policies aim at gradually narrowing the sizable gap in rural and urban incomes. Particu- lar emphasis is being placed on the poorer regions. To facilitate rural de- velopment, Government in 1972 decentralized into 18 regional administrations the principal responsibility for planning and implementing development projects within each region. Directly productive and social investments and the need for developing collective and self-help activity through ujamaa villages are being stressed. ii. The project would aim, over a five-year period, to improve the production, incomes and living standards of some 250,000 people, about one- half of the rural population of Kigoma region, one of the poorest in Tanzania. A principal objective would be to double the annual per capita incomes of villagers under the project, from US$20 to US$40 (compared to rural per capita income of US$85 for Tanzania as a whole), through simple improvements in agri- cultural practices and the provision of inputs to about 135 newly-established and existing ujamaa villages. The project would also aim to provide a market- ing and credit system, extension-related agricultural research and social services. Mlaize, beans, cotton and groundnuts would be farmed in the villages on the basis of "block" farms in which each family would be responsible for a particular plot. iii. The project would consist of: (a) land-clearing of tsetse bush, simple feeder roads and improved social services (including water supplies, health services and schools) for villages; (b) loans to creditworthy village primary cooperative societies for crop inputs and productive infrastructure such as godowns, maize mills and simple fish-catching and processing equip- ment; (c) loans to the regional cooperative union (Union) for godowns and vehicles; (d) expanding the region's rural training centers, establishing a livestock pilot project and an agricultural trials and training center and providing vehicles, aerial photography and simple radio-telephone communica- tions within the region; (e) technical assistance by way of resident staff and provision for short-term consulting services; (f) Government proposes to prepare comprehensive rural development projects in other regions and provi- sion has been made for a project preparation team to prepare rural development projects in three other regions and to assist in the early phases of the evaluation of the Kigoma project. iv. The project would constitute a major part of the regional govern- ment's development activities and would be implemented through departments of the regional administration. The Regional Development Director (RDD) would be responsible for overall project planning and implementation. The regional ."vornment's annual plan for village development under the project would be submitted to IDA for discussion and agreement. The regional government's - ii - technicians would examine the viability of villages to be included under the project according to agreed-upon criteria. Government would also discuss with IDA its annual program for all village development in the region. The Tanzanian Rural Development Bank (TRDB), lending channel for a nmber of earlier IDA projects, would be the channel for lending to the Union and village primary societies. The Union would transport, warehouse and deliver production inputs and marketable surplus under the project. v. The proposed IDA credit of US$10.0 million would cover 75% of the project's total cost. Under parallel financing the United Nations Capital Development Fund (UNCDF) would provide a grant of US$1.5 million, to go towards the costs of the water supply and health components. The IDA credit and UNCDF grant would together cover the foreign exchange cost and two-thirds of the local cost, or 86% of total costs. Government would channel funds to TRDB at 4% annually under a subsidiary loan agreement satisfactoy to IDA. Interest rates would be in accordance with general government policy for agricultural credit; existing rates are 8-1/2 percent for short-term (seasonal) credit and 7-1/2 for medium- and long-term credit. These rates are currently under review. vi. Imported fertilizers, pesticides, sprayers, radio-telephone equip- ment, and aerial photography (US$1.4 million) would be purchased through international competitive bidding in accordance with Bank/IDA, guidelines. In evaluation of bids, local manufacturers would be allowed a preference margin of 15% or the existing rate of import duty whichever is lower. Orders of less than US$30,000 would be procured in accordance with Government pro- cedures, which are satisfactory. Vehicles would be required in small lots over the disbursement period and would therefore be purchased under local competitive bidding from local agents of well-represented foreign suppliers who are equipped with essential service facilities. Classrooms, houses, village water supply supplies, godowns, dispensaries, other building, tsetse clearing, and access roads, are of small scale, varied design and are geograph- ically scattered; they would be constructed by force account and local competi- tive bidding. vii. After 8 years, incremental marketable surplus at producer prices from villages under the project is projected at about T Sh 30 million (US$4.2 mil- lion). Gross annual direct foreign exchange earnings would be about US$4.8 million. viii. Allowing for use of self-help village labor for most land clearance and village housing and for certain other village structures, notably school buildings, the full capital cost of village settlement (including allowance for regional infrastructure expenditures) would be approximately US$75,000 generated as a result of increased agricultural production, would approximately double village and family income from agriculture over an 8-year period of village development. The economic rate of return to the project is estimated at 22%. - iii - ix. The project has greater than usual risks stemming from the economi- cally unproven nature of ujamaa villages and the need to create a completely new marketing and distribution network to service project villages. Tech- nical assistance would be provided under the project to strengthen village planning, agriculture, and the marketing and credit system, which should reduce the risk. This, and the project's potential for improving markedly the incomes of large numbers of rural poor in a disadvantaged region of the country at relatively low cost, make the risks worth taking. x. Six IDA credits totalling US$62.1 million have been previously approved for agricultural projects in Tanzania (one each in agricultural credit, tea, tobacco and cotton and two for livestock). A recently approved loan provided US$21 million for a cashewnut project. In addition, an educa- tion credit in 1971 provided US$3.3 million for agricultural training. Per- formance under these projects has been mixed. xi. The project would be suitable for an IDA credit of US$10 million to the Tanzanian Government. I. INTRODUCTION 1.01 Tanzania places high priority on agriculture and rural development in its development strategy, in keeping with the needs of over 90% of its population who live in rural areas and are mainly engaged in traditional peasant agriculture. Incomes of the rural majority have traditionally been much lower than those of the small urban sector and Government's long-term aim is to narrow this gap. Particular emphasis is being placed on the poorer regions and on the development of collective and self-help activity through ujamaa (in Swahili, literally "familyhood") villages (para. 2.05). In order to facilitate rural development, Government in 1972 decentralized principal responsibility for planning and implementing development projects to 18 regional administrations. 1.02 Since mid-1972 the newly-regionalized administration in Kigoma region has been organizing the development of the rural areas by encouraging the formation of ujamaa villages. The proposed project would provide direct- ly-productive and improved social infrastructure in selected new and existing villages. The project would also aim to strengthen the marketing and collec- tion system in the region through improved organization and management. The project would directly affect about 250,000 people, roughly one-half of the rural population of the region. 1.03 Six IDA credits totalling US$62.1 million have been approved thus far for agricultural projects in Tanzania. These include one for agricultural credit, two for livestock and one each for tobacco, tea and cotton. A recently approved loan provided US$21 million for a cashewnut project. In addition, an education credit in 1971 provided US$3.3 million for agricultural training. Performance under the projects has been mixed. Management and staffing has been a continuing problem. In reference to cotton loans extended under the now completed credit project, a need was found for improved supervision of cooperative society loan records and input distribution. The first livestock project has progressed satisfactorily and is nearing completion, and the second credit has only recently become effective. The development of the tobacco project has been disappointing and slower than expected and has encountered difficulties due to inadequate services (mainly water) in the new settlemwnt areas and the premature introduction of collective cultivation under the ujamaa system. There are recent signs that implementation of this project is now proceeding more satisfactorily. The smallholder tea project is also encountering problems with farmer recruitment, and the planting program is about two years behind schedule. Weak management and other difficulties have slowed progress, and it is expected that the project will be rephased. The cashewnut loan was only recently signed and is not yet effective. 1.04 The loan application was prepared by the Government assisted by the IBRD Resident Mission East Africa (RMEA). This report is based on the .indiags of an appraisal mission to Tanzania in October/November 1973 con- sisting of Messrs. D. Martinusen, S. MacLeod, D. Pickering and D. Turnham (IDA) and G. Gerhardsen, R. Hennion and C. Pineo (consultants) and an updating -2- rission in Mlarch 1974 consisting of Messrs. D. Martinusen, S. MacLeod and D. Turnham. Under parallel financing the United Nations Capital Development Fund (UNCDF) would provide a grant of US$1.5 miUion to go towards the costs of the water supply and health components of the project. II. BACKGROUND A. The Economy of Tanzania 2.01 Tanzania has an average population density of under 15 per km2, but most cf its 14.0 million people are concentrated in a few areas. The popula- tion is increasing at about 2.7% a year. Overall growth of production remains modest, particularly in agriculture whose growth in recent years has not kept pace vith population growth. Per capita GNP is estimated at US$110 (World Bank Atlas 1973). 2.02 Tanzania's socialist development philosophy aims at reducing inequalities in income distribution through emphasis on smallholder agricul- ture and rural development, and through State ownership of important indus- tries, services and large-scale agricultural entterprises. Growth has been behind the Second Five-Year Plan (1969-1974) targets. Due to recent favor- able export prices for coffee, cotton, sisal, diamonds and meat, Tanzania's balance of payments has improved but the full effects of the recent dramatic increase in oil prices have yet to be felt. B. The Agricultural Sector General Background 2.03 Agriculture accounts for roughly 40% of Tanzania's GDP with half of tie sector subsistence production. About 80% of total exports come from agriculture, mainly coffee, cotton, sisal and cashews. Most of the population (94Z) live in rural areas, usually in small isolated communities, and 90Z of the economically active population are engaged in agriculture. 2.04 Most production is from smallholdings, using family labor. Large- scale agriculture in estates produce sisal, coffee, tea and wheat; state farms mainly produce wheat, rice, sisal and livestock. Estate production lhas diminished in importance part±cularly as sisal has declined, and the state farm program remains small. Tanzania's national herd, the second largest in Africa, is grazed extensively over the 40% of the country which is free from tsetse fly infestation. It is largely managed along traditional lines. There is limited use of oxen for cultivation, mainly in the western cotton areas. In years of normal rainfall Tanzania is largely self-sufficient in staples, although some imports of wheat and maize occur in most years; imports of staplds hlave been particularly heavy in recent months. -3- Ujamaa Villages 2.05 The Arusha Declaration of 1967 reaffirmed Tanzania's commitment to a socialist society and identified the ujamaa village as a means of effecting self-reliance and a community approach to development. The ujamaa village was seen as a vehicle for providing economies of scale in delivering improved agricultural techniques and inputs and social services, thereby improving the lives of rural people. Development of ujamaa villages was intended to be voluntary, although because of the excessive zeal among some local officials, this has not always been adhered to in practice. The precise pattern of development within each ujamaa village was to be left to villagers them- selves to decide. Villages would eventually become registered as multi- purpose cooperative societies with powers to own land and to borrow. The Tanganyika African National Union (TANU), Tanzania's only political party, has called on all rural people to support Ujamaa. Government has provided considerable assistance to newly-developed ujamaa villages, chiefly by way of transport of villagers' effects and local building materials for houses and provision of water supplies and other social services. Special suppor: has been con centrated in several poorer regions (including Kigoma). By late 1973, some 2 million persons were members of 5,600 ujamaa villages throughout Tanzania. The extent of collective activity varies among villages and the precise workings of ujamaa are still evolving. Government's initial objective was that all ujamaa productive activity should be communal; early difficulties with communal agriculture have caused the block farm approach, in which each family is responsible for a particular plot, to be the most likely form of production for the foreseeable future. C. Agricultural Marketing and Pricing 2.06 For most major traded crops, Tanzania operates a single marketing channel with a parastatal (state-owned or controlled marketing board or corporation) as the principal focus of the system. The cooperatives (primaries and regional cooperative unions) act as agents of the parastatals in collecting marketable surplus and selling processed products in the regions. The number of parastatals and their crop responsibility has changed frequently in recent years; currently, for maize and beans the relevant parastatal is the National Milling Corporation (NMC), for cotton, the Tanzania Cotton Authority (TCA), and for groundnuts, the General Agricultural Products Export Corporation. 2.07 In collecting produce from the regions, parastatals typically take title to the goods and make payment when the goods are delivered to the wkre- houses of the regional cooperative union or directly to the parastatal's own warehouses. The Unions, borrowing from the National Bank of Commerce for working capital, provide for cash payments to producers through member primary cooperative societies. -4- 2.08 Producer prices, the parastatals' buying and selling prices are set by the Central Government under approval of the Economic Committee of the Cabinet. Producer prices for a particular harvest are set in advance of the planting season and, starting with 1973-74, are uniform, within low and high priority zones, over the entire country. Parastatals' buying prices vary among regions according to differences in costs of collection and handling; their selling prices are uniform. Producers may sell small quantities of food crops outside official channels, often at prices considerably in excess of official producer prices. Unofficial and illegal trading in food crops also continues. At present, official consumer prices for maize, meal and flour are maintained at low levels, in relation to the parastatal's (i.e., NMC's) full costs, and the NMC, which would otherwise incur losses on operations in maize, is compensated by Government on an ad hoc basis. For cotton, on the other hand, operations are sufficiently buoyant to support a 10% ad valorem export tax (partially offset by subsidies to producers for purchased inputs). Input prices are also set centrally but are not on a uniform country-wide basis. Input subsidy schemes are undertaken from time to time to familiarize producers in their use and to increase agriculture production. The Tanzania Cotton Authority currently provides a subsidy of 50% of the cost of inputs used in cotton production. 2.09 Performance among cooperative unions and primary societies varies enormously. Scarcity of skilled manpower to manage, keep the books of, and supervise cooperatives remains a big problem in both Unions and primaries. The shortage of people at all levels who can maintain proper accounts and interpret their results is particularly great. Management of cooperative unions' transport and handling is also a general problem. Cooperatives are discussed further in Annex 1. D. Agriculture Credit (Annex 2) 2.10 The Tanzania Rural Development Bank (TRDB) was founded in 1971, the fourth of a succession of rural credit institutions and the third in the decade following independence. TRDB provides long, medium and short-term credit to farmers through cooperative societies; it also administers special funds on behalf of Government, and is the main credit channel for IDA funds. Ihe standard rates of interest set by Government (currently under review) and chai-ed by TRDB are 8-1/2% on short and medium-term loans and 7-1/2% on long- term loans. Societies on-lend to members at the same rate. At the end of Narch 1973, TRDB loans to societies totalled T Sh 122 million. TRDB is a ;3atisfactory channel for continued lending of IDA funds. -5- III. THE PROJECT AREA A. The Economy of Kigoma Region Location, Climate and Ecology (Annex 3) 3.01 Kigoma region, covering an area of 37,000 km2, lies to the extreme west of Tanzania (Map 10941). The region is bounded by Burundi to the north- west and Zaire to the west, across Lake Tanganyika. Kigoma port, served by the Kigoma-Dar es Salaam branch of East African railways, is the main access to the sea for land-locked Burundi and is also an important base for shipments to and from Rwanda and Zaire. A weekly air service and telephone link operate between Kigoma town and Dar es Salaam. 3.02 The region's elevations range from 1,050 meters to 2,300 meters above sea level. Annual rainfall varies between 600 and 1,600 mm, and the annual mean maximum and minimum temperatures range from 28"C to 12"C. A single rainy season lasts for six to seven months from November to early May followed by a prolonged dry season. Precipitation is reliable and allows a wide range of crops to be grown, with some double planting of short-season crops. 3.03 Climate and topography, together with the effects of past and present human settlement, give rise to three distinct ecological zones (Map 10942): (a) The Highland Zone (altitude 1,500 to 1,750 m, and annual rainfall 1,000 to 1,600 mm) is located above the tsetse fly belt and contains most of the region's cattle. Cattle are more a store of wealth and status than for production; annual offtake is low. Maize, beans and bananas and, in the higher, wetter areas, coffee are the most important crops. The zone is overpopulated. Soils are depleted and erosion is a problem. (b) The Intermediate Zone (Altitude 1,200 to 1,500 m, and annual rainfall 850 to 1,100 mm) is below the highland zone and contains areas of tsetse fly infestation. Population is much less dense than in the highland zone and until recently it had few settle- ments. Cattle are much less common than in the highland zone. (c) The Miombo Zone (altitude 1,000 to 1,200 m, and annual rainfall 600 to 1,000 mm) covers much of the east and southeast. Tsetse infestation throughout the zone has tended to restrict migration from higher population density areas. A distinct sub-zone iS formed between the interior mountains and Lake Tanganyika, extending inland at river mouths. Fishing is the principal activity and population is dense along the northern shore, but, due to steep slopes, is sparse along the southern shore. - ) -- His_tory, Demography mnd Economic Activity 3.04 Kigoma is populated by Bantu peoples, principally the Ha (70% of the total) and the Rundi, who are mostly migrants from Burundi. Kigoma is an area of comparatively late settlement, and with a tradition of relatively 5ree movement from and to other regions of Tanzania and with Burundi. Tusi pastoralists mDved into the highland region during the 19th century and remained the dominant group in terms of wealth and political leadership until the late colonial period. Settlement of white colonists and farmers was discouraged or prevented although a number of missions were opened, especially in the highlands. Communications to the Kigoma port railhead remained poor, thus discouraging the development of commercial crops away from the immediate vicinity of Kigoma town. The number of Asian traders remained small, and almost entirely confined to Kigoma-Ujiji township. The region's only com- mercial bank branch is in Kigoma town. 3.05 The Kigoma region is divided into three districts: Kibondo, Kasulu and Kigoma (Map 10942). The region's population is estimated at about 550,000 in 1973. Over 90% of the population is rural, mainly settled in the highland and intermediate zones. Kigoma-Ujiji township, with a population of about 30,000, is the only sizable urban center. 3.06 Until recently the region experienced a sizable outmigration of young men for low-paying and arduous work on estates in other regions. This movement is evidence of the traditional lack of opportunities in Kigoma to earn cash income. In 1967 the percentage of labor in paid employment (2.6%) was the lowest of any region. Employment has been overwhelmingly in agricul- ture of a low productivity kind. Seed was mainly of unimproved varieties; cultivation was by hand (mainly by women); and almost no fertilizer or other inputs were applied. Families shifted their dwellings every few years as fertility declined. Per capita incomes of most of the rural population are still, very low, in the order of US$20 per year. 3.07 The lack of industry and the backwardness of agriculture are, to some extent, related to the isolation of the Region. Historically, the most active attempts to spur development in Tanzania were direc$td to more accessible regions. Even today, the administrative centers of Kasulu and Kibondo districts have no power supply or telecommunicatlons link with Kigoma town. A simple oequest for a spare part, from anywhere in the region to Kigoma town, for eaxample, must go through the mails oI, as often required, by a special jeep .-ip. According to the 1967 Census, Kigoma region was second to last in jnzania in literacy levels and '; adult population with no schooling (over 'iK,% of tihe total). Since the building of the railway in 1914 until recently little attempt was made to develop Kigoma region. B. Regional Government and Administration (Annex 4) With effect frow 'March 1972) Regional and District administrations -anzania were reorganized and strengthened to promote rural developmuent r-c ravs more effectively. The chief political officer in the Region, the - 7 - Regional Commissioner (appointed by the President), was given cabinet rank. The Regional Development Director (RDD), as head of the regional civil ser- vices, is now equivalant to the Principal Secretary of a central government ministry. The most important effect of the changes was to make all civil servants working in a region in principle responsible to the RDD of that region and not, as before, to central ministries in Dar es Salaam. 3.09 The philosophy behind the decentralization has been that more effective rural development would be promoted if decisions taken at local level were to reflect more closely local views. In turn, local views and support for government programs would be sought through the political process, working through local level development committees. Thus, much more than in the past, the local level political structure, from the ten-family cell basic TANU unit through Ward level, District and ultimately Regional Commissioner, interacts with the civil service in program identification, preparation and subsequent implementation. 3.10 Once approved, responsibility for carrying out the programs rests with the Regional Development Director, supported by the District Development Directors, in respect of the district programs. 3.11 For all of Tanzania the percentage of the budget channeled through the regions is still small (in 1973/74, 13% of the development budget and 20% of recurrent expenditures). However, these percentages are growing as regional planning and management capabilities are further developed. For Kigoma region the recurrent budget allocation for 1973-74, at T Sh 27 million, was up almost 30% on 1972/73 and the development budget, at about T Sh 15 mil- lion, was up almost 100%, due partly to a carry-over of unfinished work from the previous year. The regions have no taxing powers of their own and are bound by national regulatory, pricing and licensing policies. 3.12 In Kigoma region initial problems have centered around staff short- ages and inexperience with the new roles. The extra strain caused by launch- ing Operation Kigoma (para. 3.13) and delivery problems on supplies and spare parts have caused some delays in implementation. Overall, however, the new focus at the regional and district levels of responsibility for planning and implementing projects has now been largely effected and should lead to more active development efforts in the Region than in the past. Operation Kigoma 3.13 Between July 1972 and November 1973 some 76 ujamaa villages con- taining about 24,000 families -- roughly one quarter of the region's rural population -- were established under Operation Kigoma (para. 2.05). The rate of starting new villages and settling families into them has since increased. Usually, villagers are drawn from isolated homesteads scattered in the vicinity of the ujamaa site. The latter is often based on an older village nucleus of perhaps 30-50 families, sometimes with certain facilities already availa;le -- e.g., road connections and a school. Settlement takes place largely during the slack season after harvesting and before land preparation; in the ujamaa - 8 - village, villagers build their own houses largely from niud and bamboo; apart from some transport of bamboo and poles provided by Government in some cases, the houses involve no cash cost. Villagers moving into villages started during 1972 have succeeded in clearing and planting house plots of 0.4 ha (I acre), building houses and rough clearing perhaps half of their initial acreage in village "block farms" for maize, beans, cotton and groundnuts. The initial favoring of wholly-communal block farm has been replaced by a system of indi- vidual plots in contiguous groups under the particular crop. Thus, each "block" of a particular crop contains a number of individual plots, each farmed by a particular family. Initial problems were faced in village siting, and in shortages of transport, spares and skilled manpower of all types. The provision of social infrastructure to the new villages lags considerably behind their establishment. C. Rural Infrastructure Education (Annex 5) 3.14 About 32% of Kigoma region's school age population (35,260 pupils; 35% of whom females) attended primary school in 1972, an increase in absolute numbers of almost 50% in 5 years. The provision of universal free primary education by 1989 is a national goal which, as achieved, will require large increases in the national budget for education. Large numbers of adults in Kigoma region are registered in village adult education classes "under the trees" but attendance has been irregular and measurable results to date unspectacular. The Region's existing primary education facilities are below standard. Tools for school gardens and the few school work shops are non- existent; textbooks and desks are worn and in short supply and teachers' houses and classrooms are often in a poor state of repair. 3.15 In order to provide education that is more relevant to the Region's overwhelmingly rural population the regional education authorities plan in the future to give more attention to teaching practical subjects (e.g., farm- ing, carpentry, masonry and smithery). Priority will be given to providing short courses for teachers in teaching practical subjects and in equipping -raucical workshops in rural primary schools. The courses for teachers could - n-ov-ded at the Region's rural training centers, involving some expansion of these centers for this and other purposes. Priority will also be given to -rcviding textbooks and desks for existing and new classrooms and for teachers' louses. Additional classrooms, also required as the percentage of school Attenders rises, will also be necessary. New ujamaa villages are often located near existing schools which would te expanded as the villages' populations Lncrease. Village Water Supplies (Annex 6) 3.16 Provision of clean water at convenient public points in villages is an important government objective and an important part of its ujamaa .zillage program. In Kigoma the office of the regional water engineer (Maji/K) -9- has been engaged u'-ader Operation Kigoma in providing village water supply systems to newly-established ujamaa villages. Much of the Region is well- endowed with perennial streams, but villagers (usually women) have to walk 3 to 5 km for water from a source that is in varying degrees contaminated; the time so spent is often at the expense of productive work in the fields. The principal focus of Maji/K to date has been on providing pumped supplies which deliver water to 10 to 20 public hydrants located at convenient points througnout the village. The new pumped system if properly installed provides water that is cleaner and safer than the former source (usually a stream). It also saves much time for villagers compared to their former walk to the water source. Despite shortages of transport and materials Maji/K has developed to the point where it can nuow install 10-12 such systems per year. This rate is, however, much slower than the rate of opening new villages and there is now a large backlog of newly-established villages having no improved water supply at all. 3.17 No charge is levied in Tanzania for water delivered to public hydrants. This, coupled with a preference for pumped systems, will result in very large claims on the budget for both capital and recurrent costs. There is need to utilize more cost-effective methods of water supply. For example, simple dugwells, sealed and equipped with handpumps, where adequate shallow groundwater exists, provide clean water at convenient points at a fraction of the cost of pumped systems. Public Health (Annex 7) 3.18 In 1972, national priorities in health services were reordered to give greater attention to preventative medicine and health services in rural areas. The rural treatment system would consist of (a) a rural health center with 5 outlying dispensaries per 50,000 population to cover all non-hospital needs of families and (b) rural hospitals at regional and district levels to provide major curative services. Government's quantitative goals for dis- pensaries and rural hospitals have largely been met, although the spread is uneven and some districts are below national norms. For the rural health centers, however, more than half of the program is still to be completed. The supply of trained staff is a problem but newly-established schools should provide a rapid increase during the next several years in numbers of medical aides and rural medical aides, the personnel who operate, respectively, the rural health centers and dispensaries. 3.19 For Kigoma region, the development of the rural health service is on a par with rural Tanzania as a whole. As well as more rural health centers, however, some additional dispensaries will be needed as the population is re- settled in new ujamaa villages. The staffing constraint will be eased by the newly-created training school in Kigoma for rural medical aides. More needs to be done on preventative medicine. For example, currently nothing is yet being done to introduce improved methods of excreta disposal, a first step in dealing with the internal parasite problem. - 10 - D. Agriculture and Fisheries Agriculture (Annex 3) 3.20 With the exception of the densely-populated areas in the Highlands, traditional agriculture is based on shifting cultivation. Maize, beans, cassava and such minor crops as sweet potatoes, sorghum, pigeon peas and groundnuts are most commonly grown in mixed stand and in varying combinations throughout the Region. Bananas and plantains also form an important part of the diet--raw, cooked or fermented and consumed as beer. They are normally planted close to living quarters where they receive organic manuring in the form of domestic refuse and compound sweepings. No other crops are normally so treated. 3.21 No data are available on traditional cropping patterns. Local opinion and field observations point to maize, bean, cassava mixtures as the most common staple crop grouping, with a length of cropping ranging upwards from about three years depending on fertility and topography. Inorganic fertilizers have rarely been used in the past but during the last two seasons have been increasingly introduced by Government free of charge or subsidized (e.g., by the Tanzania Cotton Authority) to promote their use. Although improved seed of maize and groundnuts are apparently appreciated, their dis- tribution to date has been limited to selected ujamaa villages created since 1972. The regional government intends that inputs in future be provided on credit, thereby avoiding a large budget drain when use of such Inputs becomes widespread. 3.22 The circumstances outlined above prevent an accurate estimate of yields and production under traditional practices. Cash incomes are derived mainly from sales of crops grown surplus to subsistence needs; however arabica coffee in the Highland Zone, oil palm products mainly in Kigoma District, and cotton in the Intermediate Zone, are also of some importance. 3.23 Achievement of improvements to traditional practices ham been seriously hampered by three factors. First, little or no research work has been carried out in Kigoma Region so that extension staff have at best very generalized and only partially proven recommendations to put across to farmrs. Second, the numbers and quality of farmer contact extension workers have been inadequate to make a real impact on the farm population. Third, the pattern of settlement throughout the Region tends towards a scatter of population wijith relatively few nucleated villages. This makes contact with farmers difficult and severely limits the impact of extension workers. 3.24 Marketable surplus in the region has been extremely small. The Ki gomz regional cooperative union, the official channel for handling market- able surplus, is small and weak both financially and managerially. -. 1 1 - Fisheries (Annex 8) 3.25 Fisheries in Kigoma region, mainly on Lake Tanganyika, involves some 10,000 fishermen operating about 7,000 canoes from 35 fishing villages. The most important species caught is dagaa, a small, sardine-like fish that is dried and traded widely throughout Tanzania and neighboring countries. Some improvements in catching technique have been introduced in recent years by way of pressurized lamps for attracting fish during night fishing (the principal form), and beach seines. In addition, three purse seine units (each costing about US$70,000) have begun operating out of Kigoma harbor. A recently commenced FAO/UNDP fisheries project centered in Kigoma will experiment with other simple improvements for traditional fishermen, as well as with improved dagaa drying techniques (currently a problem in the wet season when catches are greatest). Some of these techniques will be avail- able within two to three years; some credit will be necessary so that villagers can adopt them. IV. THE PROJECT A. Brief Description 4.01 The Project, over a five-year disbursement period, would continue to support village development already started under Operation Kigoma and of new villages to be established during the project period. Most of the rural population of the region (some 500,000 people) would benefit in some measure from the project; about half would be directly involved, as residents of about 135 project villages. The main objective of the project would be to double present annual per capita income in the project villages from the equivalent of about US$20 to US$40. Through support for agricultural and cooperative development, regional infrastructure, farmer and staff training and technical assistance, the project would also develop a marketing and credit system, strengthen village-level extension and introduce a program of adaptive agricultural research. These services are increasingly necessary as ujamaa development in Kigoma region moves forward from the initial settle- ment phase to the stage at which village ujamaa communities are to fully sustain themselves on a permanent basis through economic activity. In addition, the level of social services would be improved through provision of education, health facilities and village water supplies. The project would consist of: 1. Investments in eligible villages for public water supplies, education and health facilities, tsetse clearing and feeder roads; 2. Loans to eligible and creditworthy villages for: (a) pro- duction-infrastructure including godowns, maize mills, fish catching and processing equipment and (b) seasonal produc- tion inputs; - 12 - 3. Loans to the regional cooperative union for facilities such as godowns, offices, and transport equipment; 4. Investments in regional supporting services, including: (a) construction and rehabilitation of rural training centers; (b) a small regional radio-telephone system; (c) a livestock pilot project; (d) aerial photography of areas to be developed; and (e) a regional agricultural trials and training center; 5. Technical assistance to include (a) a financial controller, an operations manager and a training supervisor for the regional cooperative union, a land-use planner and trials and training officers for the agricultural trials and training center together with training; (b) backup support such as housing and transport; and (c) a project special fund for use by project management for specific problem-solving; 6. Monitoring and evaluation; and, 7. Project preparation for rural development projects in other regions. The regional government would have overall responsibility for project implementa- tion. The Tanzania Rural Development Bank would be the channel for credit under the project. B. Detailed Features Vi llage Infrastructure 4.02 To be eligible for financing under the project, a Village Site Feasibility Report (VSFR) would have had to be prepared (para. 5.04) demon- strating that the village is appropriately sited in relation to various factors including available agricultural land and water. To assist in this, the project would provide for the aerial photography of about two-thirds of the Region. Water Supply Systems 4.03 Water supply systems under the project would bring potable water year round to public points located so that villagers would generally not have to walk more than 200-300 m for water. The regional water engineer's office (Maji/k) would design for a particular site the type of system that would provide adequate year-round supplies of clean water at minimum cost. Where adequate shallow groundwater exists a number of dug wells, generally - 3- of less than 10 m depth, covered and fitted with handpumps, would be con- structed. In some cases Oravity delivery, direct from a water source, may be possible. WJhere handpumps or gravity feed are not possible a diesel-powered pump system may be required. This is generally a more expensive system to build anJ has higher operational costs, At negotiations assurances were otta4ned tnat village water supply systems would be designed according to minimum cost szecifications, taking into account the alternative systems described above. 4.04 Government policy in Tanzania is to provide domestic water free of charge except if the supply is through a house connection. As part of the project preparation component of the project (para 4.17) a study would be undertaker, to examine the fiscal implications of the water supply component and appropriate methods and procedures for allocating and financing the cost of such services. IDA and Government would exchange views on the study's findings and recommendations. Assurances on the study and exchange of views were obtained at negotiations. Village water supply is discussed furtheL in Annex 6. 4.05 Village-level education facilities financed under the project would iinclude tools and practical workshops, textbooks and desks, teachers' houses, and classrooms according to regional priorities for primary education (para. 3.15). The facilities would also be used for adult education. When a new villaee is started a single classroom would be built in the first year for first grade. An additional classroom would be built in each subsequent year until the full primary school, covering seven standards (or grades), would be completed. Where new villages are established near existing schools they would be rehabilitated and also expanded as necessary. Facilities would be provided in accord with Government's national education norms (para 3.14 and Annex 5). '4.06 Local level health care facilities including rural health centers and simple dispensaries would be provided where health coverage is below national standards for rural areas. Dispensaries are generally of two rooms, stocked with a limited range of medicines and manned by dispensary assistants who have had some training in a regional hospital. The project would also include provision for about 10% of project villages of pit privy slabs and with instruction in their installation and use, as a pilot phase of a program of improved excreta disposal. Many low-lying areas are infested with tsetse flv which transmits sleeping sickness. Where a village is located in such an area, the project would support tsetse clearing of bush in and around the villaga. site, using simple tools and labor-intensive methods (on a partial self-help basis). Feeder roads would join those villages (probably less than hlalf) not already connected to the existing road system. They would be simple tracks, capable of handling loaded lorries, mainly in the dry season. .L; .o vll1ages for Productive Purposes 4.07 Creditworthy villages under the project would be able to borrow from TRDB through the regional cooperative union (para. 5.10) for productive infrastructure, including godowns and maize mills. Villages that are wholly - 14 - or partly dependent on fishing could borrow for improved catching methods (e.g., beach seines) or processing equipment (e.g., simple equipment for drying dagaa during the wet season). Some of the loans would be for catching or processing improvements expected to be proven by the UNDP fisheries project (para. 3.25). 4.08 Creditworthy villages would also borrow from the regional cooperative union which in turn would borrow from TRDB (paras. 5.08 to 5.10) for incremental production inputs including fertilizers, pesticides and, in the case of cotton, ultra low volume (ULV) sprayers. To be eligible for credit a village must be registered as a cooperative society and have in residence a qualified bookkeeper and agricultural extension officer. A new village would not normalJF be eligi- ble for credit before its third season. Loans and Equity Contribution to Regional Cooperative Union for its own Facilities 4.09 The regional cooperative union (Union) would distribute production inputs to project villages and purchase, collect and deliver the villages' marketed surplus to the relevant parastatal for marketing or further process- ing. Once villages are eligible for credit, their marketed surplus and requirements of production inputs are expected to grow rapidly under the project (para. 6.03) and requirements of the Union for godowns, offices and lorries would increase correspondingly. It is expected that the Union would borrow for a core fleet of about 15 lorries, together with necessary service facilities and spare parts. The Union would hire additional transport (from parastatals, the regional government and private sources) during peak periods. Government, as part of its share of project cost, would also make an equity contribution or long-term loan (para. 5.12) as required to cover expected losses during the first four years of the project, estimated at about T Sh 1.2 million (US$170,000). Regional Supporting Services 4.10 A new rural training center would be constructed in Kigoma and the existing centers at Kasulu and Kibondo would be improved and expanded under the project. The three centers would house various training activities accompanying the project, including courses for village officials, teachers and government officers, as well as courses on a variety of rural affairs subjects currently conducted at the existing centers. Courses would also be held for supervisors, tradesmen and operators involved in the construction and operation of water supply systems. 4.11 The project would also include a simple radio-telephone hookup between regional headquarters in Kigoma town and each district headquarters and between each district headquarters and several major villages within each district. 4.12 A regional agricultural trials and training center would be established under the project at a central location that is typical of much of the Region's agriculture, perhaps near Kasulu town. The center annually would provide to - 15 - about 25 primary school leavers, having some farming experience, a combination of classroom and -:ractical training of about one year's duration, that would equip each student to work as an agricultural field assistant in an assigned village. The center would also conduct field trials to test the suitability under regional conditions of seeds and agronomic practices recommended by the national agricultural research center at Ukiriguru. The project would cover the necessary land preparation, physical structures, equipment, housing and staff. Assurances were obtained at negotiations that selection of the site, based on a detailed soil and topographic survey made available to IDA, would be carried out and actual construction started within six months of credit signing, with construction to be completed within 12 nDnths thereafter. Located adjacent to the agricultural trials and training center would be a livestock pilot project whose purpose would be to investigate and demonstrate livestock management systems appropriate to the Region and to introduce breed- ing stock (from types such as the Boran) for the gradual upgrading of the local Ankole cattle. The project would also include funds for a small number of vehicles and lorries to enable the regional administration to carry out its increased activity resulting from the project. Technical Assistance 4.13 In order to strengthen the regional cooperative union (Union) the project would provide an operations manager, a financial controller and a training supervisor. The operations manager would set up and initially operate the Union's distribution, collection, transport, scheduling, storage and handling, quality control and physical planning activities, and train staff. The financial controller would carry out similar functions relating to the Union's accounting, financial reporting, budgeting, credit appraisal and collection, internal auditing, and cash management activities (para. 5.11). Staff training would be a major responsibility for both. The training supervisor would work closely with the DlD's office in establishing a simple village-level bookkeeping control system and in training school leavers from the Region to become village-employed bookkeeper/business managers. He would also assist in initial training and continuing supervision of the Union's Society Credit Officers and in general training and credit matters. 4.14 The land-use planner would advise and assist in the layout of villages, with particular reference to the siting and alignment of block farms on the basis of aerial photography and ground surveys. He would also be responsible for supervision and on-the-job training of regional land-use planning staff. The training officer at the agricultural training and trials center would be principally responsible for planning and conducting training courses at the center, drawing on the assistance of the land-use planner and others. The trials officer would be in charge of the center's trials program which would also involve village level observations and demonstrations located at suitable sites in the Region. The project would finance inputs required for other demonstrations and would also cover during training the subsistence and training costs of village-level bookkeepers and agricultural field assistants. - 16 - 4.15 Reco_mmnded terms of reference and required qualifications for staff to be financed under the project are in Annex 9. The project would also finance required housing and vehicles for these staff. During negotiations assurances were obtained that within six months of effectiveness the operations manager and financial controller would be hired with experience and qualifications acceptable to IDA and on terms and conditions acceptable to ID&; in addition, Government would hire competent and experienced persons as the training super- visor of the Union, the land-use planner and the trials and training officers. 4.16 A number of problems would inevitably arise in a project of this type which cannot be precisely identified in advance. A project special fund would therefore be included for use at the discretion of the Regional Develop- ment Director for consulting services and technical assistance in solving specific problems of project implementation. The fund could be used, for example, to send staff to similar projects elsewhere in Tanzania or nearby countries for insights in solving particular problems or for bringing in specialists for short periods. The management of spare parts and vehicle maintenance in the region is a problem; regional staff could be sent to appro- priate courses or, alternatively, a consultant could be brought in for advice and training. Expenditures in excess of US$30,000 equivalent in any one fiscal year or over US$10,000 equivalent in any one contract would require prior IDA approval. Regional Planning, Project Preparation and Evaluation 4.17 The project would provide support for the further development of Tanzania's rural development program through finance for a team of experts, to be attached to the Prime Minister's Office, to assist the regional staff of other regions in project preparation of rural development projects and regional plans, including examining their policy and fiscal implications. It is expected that three regions would be covered over a two-year period. The program would be directed by a coordinator who would work out of Dar es Salaam. The team would liaise closely with RMEA. With the help of the specialist staff and local expertise, the coordinator would also be responsible for implementing a system of monitoring and evaluation of the Kigoma project. Details of the proposal are further described in Annex 10. C. Project Costs (Annex 11) 4.18 Total project cost, over a five-year disbursement period, is estimated at T Sh 95 million (US$13.3 million), of which US$8.0 million (60%) would be the estimated foreign exchange component. Details are summarized in the following table: - 17 - Foreign Local Ftroei Total Local Fbrin Total Xxch&nKe ----- T8hZ - -----O % Vilage Infrastructure Water Supply 6,206 6,1459 12,665 869 905 1,7714 Health 868 682 1,550 122 96 218 Iduoatima 7,100 8334 15,14314 9914 1,167 2,161. 5 Beder Roads 766 192 958 107 27 1341 20 Taste. Clearing 716 126 842 100 18 118 15 SUB-TOTAL 15,656 1.5,793 31,1449 2,192 2.213 4,405 koans to Villaesr Seasonal Inputs Revolving Fumd 1,541 6,165 7,706 216 863 1,079 8C M1d-Twrm Credit for Production Infraatruotue & 3.120 5.7h 37 57 SUB-TOTAL 3,895 9,285 13,180 5546 130 1,8346 Loa"he h isr.1 OmUlbaO 2,108 3,590 5,698 295 503 798 63 Regional Infrastruatire Regional Trainlg Centero 990 778 1,768 139 109 248 4 Radio-Tol yhone Hookup 154 196 350 22 27 49 56 Agric. Trall/Training Oenter and Livestock Pilot ProJect 498 375 873 70 53 123 43 Vhicle.a 140 420 560 20 58 78 75 Demonstrations 43 171 214 6 24 30 80 Aerial Fhotography 180 1,620 1,800 25 227 252 90 SUB-TOTAL 2,005 3560 5,565 282 498 780 Technical Assistanoe 1,768 4,124 5,892 248 577 825 70 Project. Prearatim and Evaluation j1,481 4 2 208 622 830 75 TOTAL JA.7 71. Centingency Alowanoe Phymical 1,126 1,762 2,888 158 246 404 Price 9,545 14,467 24,012 1,337 2,026 3,363 Ttal6^ Prodot oet Z7,584 57,025 9229 6 7,985 .251 60 - 18 - Costs are based on November 1973 prices, with fertilizer prices updated to February 1974. Price contingencies of 14X for 1974, 11% for 1975 and 7-1/21 per year thereafter have been added to all costs. A physical contingency of 5% has been added to all costs except seasonal credit production Inputs. D. Financing 4.19 Financing of project costs would be shared as follows: T Sh US$ (mllion) (million) x IDA 71 10.0 75 Government of Tanzania 13 1.8 14 UNCDF 11 1.5 11 95 13.3 100 The proposed IDA credit of US$10.0 million would be on standard termn to Government and would cover 751 of total costs. Under parallel financing the United Natie.. Capital Development Fund (UNCDF) would provide a grant of US$1.5 million, to go towards the costs of the water supply and health compo- nents. The IDA credit and UNCDF grant would together cover the foreign exchange cost and two-thirds of local costs, or 861 of total costs. Fund for credit to cooperatives (i.e., the regional cooperative union and village primary societies) would be made available by Goverment to TRDB at 4% annually for 25 years including 5 years' grace under a subsidiary loan agreement acceptable to IDA, the execution of which would be a condition of effectiveness. Because village primary societies are new and have little or no savings, TRDB would on-lend up to 100% of project costs. Villages would pay an interest rate of 8-1/2% for seasonal inputs and 7-1/21 for economic infrastructure, the pre- vailing interest rates throughout Tanzania. Government is currently reviewing these interest rates. TIDB would lend to the Union for its own facilities and for on-lending to villages for seasonal inputs and village infrasture. TRDB in sowe cases would also lend direct to village primary societies for village productive infrastructure. Where the Union on-lends to villages it would not take an interest margin but would cover its costs from its levies on handling villages' produce and inputs. Government would retain funds for all other parts of the project. E. Procurement 4.20 Requirements for fertilizers, pesticides and ULV sprayers (about US$1.1 million) would be bulked with similar item required under other IDA- financed projects in Tanzania. These items together with aerial photography and radio equipmut (about $0.3 million) would be procured through interna- tional competitive bidding in accordance with Bank/IDA Guidelines for - 19 - procurement. Bulking of orders for vehicles and spare parts (about $0.48 million) would not be practicable since they will be purchased in small lots as required over a period of about five years. Purchases would be made follow- ing competitive bidding by local agents of well represented foreign suppliers who are equipped with essential service facilities. In the evaluation of bids local manufacturers would be allowed a preference margin of 15% or the existing rate of customs duties, whichever is lower. Orders of US$30,000 or less would be handled in accordance with existing government procedures which are satisfactory. 4.21 Contracts for construction of classrooms, houses, water supply systems, godowns, dispensaries and other buildings, which are all small, of varied design and geographically scattered, would not be a sufficient size to attract international bidders and would be constructed by a coubination of local competitive bidding and force account. In evaluating bids domestic contractors would be allowed a 7-1/2% preference. F. Disbursement 4.22 Disbursement of funds from the credit would be on the following basis: A. UNCDF would disburse 75% of total expenditures for water supply systems and health facilities. B. IDA would disburse against the following: (a) 90% of civil works for education facilities, feeder roads and tsetse clearing, and housing; (b) 100% of foreign expenditures or 90% of local expenditures for the incremental cost of seasonal inputs; (c) 90% of amounts disbursed to the Union as sub-loans for vehicles and spares and other collecting and marketing facilities and to project villages as sub-loans for productive infrastructure; (d) 100% of foreign expenditures or 90% of local expenditures for aerial photography, radio telephone system, vehicles and spares (other than for the Union), equipment and seasonal inputs for trials and equipment for the rural training centers, classrooms and the pilot livestock program; and, (e) 100% of foreign expenditures and 90% of local expenditures for project preparation, evaluation and technical assistance and related equipment and expenditures from the project special fund. - 20 - Disbursements against A and (a), (b), (d) and (e) of B would be fully documanted. In the case of disbursements against certificate of expenditure (i.e., (c) of B above) documentation would not be submitted to IDA for review but would be available for inspection during IDA project supervision. The incremental oost of seasonal inputs in (b) above would be the amounts spent in a particular year over and above amounts spent in previous years under the project. Any funds remaining in the credit account on completion of the project would be used, at IDA discretion, on further development in Kigoma region. An esti- mated schedule of IDA disbursements is shown as Annex 12. G. Accounts and Audit 4.23 TRDB and Government would maintain separate accounts for disburse- ments and subloans under the project. Both sets of accounts would be audited by independent auditors acceptable to IDA, and the accounts and auditors' reports would be submitted to IDA within six mDnths of the close of each financial year. Auditing of the cooperatives in the Region is the responsi- bility of the audit and supervision fund of the Prime Minister's Office. During negotiations assurances were obtained that audits of cooperatives participating under the project would be completed within six months of the close of each financial year and that to facilitate this, auditors, reporting to the office of the Registrar of Cboperatives in Dar es Salaam$ would be established within the Region prior to any lending to cooperatives under the project. H. EcoloRical and External Effects 4.24 The fertilizers and insecticides to be used in the project would have only a small effect on the local ecology given normal precautions in their use. Endosulphan insecticide is toxic to humans and appropriate safety instructions would be included in all staff and farmer training. The existence of an erosion hazard following from village formation and land consolidation has been fully recognized. Village site selection, based on the Village Site Feasibility Report (para 5.04), together with subsequent housing and farm layout will be in accordance with sound soil and water conservation principles laid down and monitored by project land use planning personnel. V. ORGANIZATION AND MANAUEHENT Regional Goverment 5.01 The project would constitute a major part of the regional govern- ment's development activities and would be implemented through departments of the regional administration. The Regional Development Director (RDD) - 21 - would be responsible for overall project planning and implementation. The Tanzanian Rural Development Bank (TRDB) would be the channel for lending under the project. Village Planning 5.02 The regional government would coordinate planning for the number and location of new villages started in a particular year. The final decisions on new villages and on public investments in existing villages would be embodied in the district and regional plans and budgets as finalized through the annual budget process (Annex 4). 5.03 In order to ensure that new and existing villages to be financed under the project are suitably sited in relation to agricultural land, water and other factors, each site would be studied by relevant district and regional functional managers and specialists to determine the site's economic feasibility. The study team would generally consist of specialists in agri- culture, public works, land planning, water, livestock and finance, supple- mented, as necessary, by specialists in education, forestry and health. 5.04 Each team would prepare a village site feasibility report (VSFR), a detailed assessment of the site's agricultural potential, water and infra- structure availability, village development plan and cost estimates and a signed opinion on the overall feasibility of the site and its maximum sustain- able size. The land planner (para. 4.14) and agriculturalists would use the results of special aerial photographic surveys (para. 4.02). Each team would examine the extent, location and quality of arable land adjacent to the proposed site and estimate the site's maximum carrying capacity, taking into account time taken in walking from homes to block farms. The water engineer would examine alternative means of providing village water and estimate their costs. To facilitate this, irhere appropriate, one or two dug wells would be sunk, as part of the VSFR exercise, to determine the availability of ground- water. VSFRs would not be forwarded to IDA but would be available in Kigoma for review during supervision. 5.05 Government would annually prepare and make available to IDA by March of each year for discussion and agreement an annual development plan for village investments and training to be financed under the project during the following fiscal year. Govermnent would also supply to IDA for comment its annual village development plans covering all villages in the Region. Village planning procedures and a pro forma VSFR are described further in Annex 13. During negotiations assurances were obtained that the procedures outlined in paras. 5.02 through 5.05 and in Annex 13 would be followed in identifying, selecting, planning and implementing investments in project villages. Agriculture 5.06 Each project village would have had an agricultural field ass3stant stationed permanently at the village for at least one season before it could borrow under the project. About 80 agricultural extension workers currently - 22 - work in the Region and would be supplemented by those trained at the regional trials and training center which would be set up under the project (para. 4.12). The field assistants would work with village committees in organizing demonstra- tion plots and in advising villagers on agricultural practices. 5.07 After a village reaches a stage where it would be able to utilize purchased inputs (generally not before its third season of development) it would apply for loans from the regional cooperative union (Union). The village to be eligible for loans, would, among other things, have to become a registered multi-purpose cooperative society and have hired a book- keeper (para. 4.08). The village (assisted by the field assistant and village bookkeeper) would decide on the amount of loan it would apply for and on-lending arrangements with its members. During harvest the village (as a cooperative society) would deduct, from harvest payments to members, outstanding loans for inputs. The Regional Cooperative Union (Union) 5.08 The Union would submit a loan request to TRDB to cover the requests for seasonal inputs by villages which the Union has approved. TRDB would procure (para. 4.20) the inputs for approved loans and would arrange for their shipment to Kigoma. The Union would receive the inputs at Kigoma and arrange for unloading, storage, documentation, making up of village lots, and trans- port (generally around September-November) to the villages. 5.09 During and immediately following harvest (generally May-July) the Union would buy and collect the villages' crops. To effect this the Union would make small cash advances to the villages during harvest. The Union would deduct these advances before making subsequent payments to villages for crop purchases. The absence of any form of banking or other financial network in the Region's rural areas requires the Union to get involved in sizeable credit operations. 5.10 The Union would also borrow medium term from TRDB for its own needs (e.g., trucks) and for on-lending to villages for production infra- stnrcture (e.g., village godowns). Trading and lending procedures under the project are further described in Annex 14. 13.11 The Union would be strengthened by the hiring of an operations manager and a financial controller to set up operating and financial systems, initially operate them and train staff in their use (para. 4.13). The finan- cial controller would screen loan requests from villages in preparing its loan requests to TRDB. He would also monitor the Union's overall financial position, reporting at least monthly to TRDB and, with diagnosis and recommendations, to, the Union's management and the Regional Development Director.' A training super- visor would also be hired whose principal responsibility would be training village-level bookkeepers and providing other assistance on credit matters. 5.12 In order to place the Union on a firmer financial footing the Gov- erment would effect a financial reorganization of the Union. The reorganiza- tion would combine a capital contribution by Goverment (to be used to repay - 23 - part of the Union's outstanding borrowings from the government-owned National Bank of Commerce (NBC)) and a partial write-off and/or rescheduling of re- maining principal and interest by NBC. The objective of the reorganization would be to provide the Union with a level of debt service payments that it can carry in light of anticipated margins, volumes and expenses; specify- ing exactly what this amount is must await completion of the Union's audit- ed accounts. The financial reorganization would be followed by subsequent measures such as a capital contribution (or suitably-deferred long-term loan) by Government sufficient to cover losses, during the first four years of the project, as and if they arise; these are currently projected at about T Sh 1.2 million (US$170,000) (para 6.08). During negotiations assurances were obtained that within six months of effectiveness Government would carry out a financial reorganization of the Union and, in accordance with good financial practice, would take such other measures as might be required from time to time ensure that the Union can service its debts. Tanzania Rural Development Bank 5.13 The Tanzania Rural Development Bank (TRDB), with which IDA has considerable experience in several projects (para 2.10), would be the lending channel for loans to the Union (and ultimate on-lending to the villages). TRDB opened an office in Kigoma in 1972 and plans to increase the number of credit supervisors stationed there as the volume of business expands. TRDB would not lend until the operations manager and financial controller were hired, the union's books had been satisfactorily audited and the financial reorgani- zation of the Union had been effected. TRDB would lend direct to the Union for agricultural inputs, which the Union would on-lend to its member village/ cooperative societies, and for the Union's own needs (e.g., for transport equipment). TRDB would also lend direct to villages or through the Union for village infrastructure. TRDB's appraisal of the Union's periodic loan requests would include examining the individual loan requests of villages to the Union. 5.14 The Union also borrows for working capital needs (chiefly for inventories and cash advances) from the National Bank of Commerce, the official source of such credit in Tanzania. Since these total borrowings affect the Union's ultimate creditworthiness TRDB would also continuously monitor the Union's overall financial situation. Assurances were obtained from TRDB at negotiations that it would strengthen its office in Kigoma as the project requires and follow lending procedures as outlined in paras. 5.13 and 5.14 and in Annex 14. 5.15 Margins, allowed primary societies and Unions in selling and buying prices, affect their respective debt-servicing abilities (para. 6.08). During negotiations assurances were obtained that Government and IDA would discuss from time to time the adequacy of these margins having regard to the financial viability of primary societies and the Union, and reasonable incentives for producers. - 24 - VI. PRO W CTION, MARKETS, FARMER BENEFITS AND FINANCIAL RESULTS Production, Yields and Marketable Surplus 6.01 The following projected crop yields for the principal crops at full development compared with yields under tradiitonal agriculture without the project are based on limited field experiments carried out in Kigoma region (paras 3.21 and 3.22) and on field trials and experience under similar ecologi- cal conditions elsewhere in Tanzania and ne.ghboring countries. Yield increases from existing low levels would be facilitated under the project by the training of farmer contact extension workers, further trials and demonstration of recow- mended practices on farmers' plots, provisioa of a credit and inputs delivery system at the village level, and by the principle of individual responsibility for plots within the block farms in each ujamaa village. The latter factor is of particular significance in promoting the transfer of recommended practices because of the ease of access to farmers for both extension advice and input supply. Traditional At Full Village (vithout project) Development (Yr. 8) … - - - - - - - (kg/ha) - - - - - - - Maize 600 1,460 Beans 280 600 Seed cotton 390 990 Groundnuts 390 600 6.02 A sizeable portion of a typical village's production of food crops under the project would be expected to go for subsistence, at consumption levels above those under traditional settlement and assumed to increase at 3% per capita per year during the project period. The production, subsistence consumption and marketable surplus of an average 350 family village at Year 8 of its development is projected as follows: Maize Beans Seedcotton Groundnuts Village production (tons) 287 122 155 52 Subsistence consumption (tuns) 174 74 - 22 Marketable surplus (tons) 113 48 155 30 o.03 Marketed surplus prior to the project by families under the project is Kiot known but would have been slight. Marketable surplus for the total project is projected as follows: - 25 - Project Year 2 5 8 Tons Value Tons Value Tons Value (OOOs) (T Sh mil.) (0OOs) (T Sh mil.) (000s) (T Sh mil.) Maize 1.5 0.5 8.4 2.9 10.0 3.5 Beans 1.0 0.9 4.9 4.1 4.7 3.9 Seedcotton 1.1 1.4 10.7 13.9 14.3 18.8 Groundnuts - - 0.7 1.0 2.6 3.6 Total value 2.8 21.9 29.9 Assuming 85% of the cotton was exported, the project's annual gross direct foreign exchange earnings at Year 8 would be about US$4.8 million. Crop budgets together with projections of buildup of input needs, cropped areas, and marketed surplus are presented in Annex 15. Markets and Prices 6.04 Tanzania currently produces about 400,000 bales of cotton lint annually of which some 85% is exported, mainly to Asian countries. Project output of cotton would be less than 1/20 of 1% of world production and would have no significant impact on world prices. 6.05 Domestic requirements of maize, for which Tanzania has been a net importer during most years, are projected to continue to grow rapidly. Beans and groundnuts are marketed domestically. For all three, projected marketable surplus under the project at full development is expected to account for a small proportion of the increase in domestic requirements above present levels. A discussion of market prospects for crops expected to be grown by project villages appears in Annex 16. Farmer Benefits 6.06 Incomes of farmers prior to joining the project are not precisely known but are believed to be near subsistence level and are estimated at T Sh 700 (US$100) per family or T Sh 140 (US$20) per capita. By Year 8 of the project, incomes from agriculture are projected to double from T Sh 115 (US$16) to T Sh 230 (US$32) per capita; total income (including estimated off- farm income) would double over a 12-year period from US$20 to US$40 per capita. Financial Results 6.07 Financial models for both a village primary cooperative society and for the Union have been prepared on the basis of projected volumes, operating costs and margins and are shown in Annex 14. At projected volumes and margins - 26 - thc primary society (Annex 14 Table 10) is able to meet its operating costs and debt-service obligations and earn a small surplus which could be used to cover any bad debts. The projected volumes of marketed surplus handled by the primary society could fall by 40% and it could still cover its expenses and debt service obligations. 6.08 The Union (Annex 14 Tables 1-9) is projected to incur losses during its first four years under the project (mainly of a "start up" nature) but thereafter to earn surpluses. The financial viability of both the primary societies and the Union depend on volumes handled and the amount of society and Union margins or levies. In particular, any substantial increase in the level of bad debts (a perennial problem for a number of cooperative unions in Tanzania) would place the Union in a loss position. The level of levies is largely set by Government (para. 2.08 and Annex 14) and would be discussed from time to time with Government (para. 5.15). Government Revenues and Outlays 6.09 The major marketed project output would be cotton, on which an export tax at the rate of 10% ad valorem is levied, and approximately T Sh 30,000 per annum would be recovered from a village at full development through this and other minor tax proceeds. On the other side, large increases in govern- ment recurrent expenditures can be expected; recurrent expenses of Government in project villages at full development would be about T Sh 80,000 per annum. This includes the operation and maintenance of a primary school (at T Sh 50,000 per annum), a village water supply (T Sh 6,500 per annum), and a share of rural medical facilities (T Sh 10,000 per annum), together with the salary of a village agricultural assistant and other minor items. Supervision and overhead staff servicing for these facilities at district and regional levels might add an additional 20 - 25% to the in-village costs, bringing the total to about T Sh 100,000 per annum. 6.10 Not all these costs are strictly incremental to the project, since Government now supports a considerable number of schools and medical facili- ties in the Region, some of which would be available in project villages. In- cremental government outlays under the project might be roughly two-thirds of ernmnnt servauts; allowing for their payments of income tax, project expendi- tures, on balance, would result in a net current account deficit of roughly T Sl 30,000 per village per annum. Tue gradual implementation of Government norms for provision of social services to rural areas would result in similar increases in Government recurytnt costs, with or without the project. To the extelnt that provision of similar levels of social services is attained else- wJhere in Tanzania, Government will need to consider ways in which additional revenue can be raised, either in the villages or from other sources, to cover tihese additional outlays or to provide similar services at lower cost. - 27 - VII. ECONOMIC ANALYSIS 7.01 The 135 villages to be supported under the project would receive varying types of assistance, depending on facilities already available in the village and how far development proceeds during the disbursement period. The economic analysis (Annex 16) examined a representative model village, settled at the beginning of the project period, with no previously existing village facilities. An attempt was made to include in the calculation of costs, estimates for all expenditures taking place in the village relating directly to quantifiable benefits, together with a share of complementary investments supported under the project at regional level (including technical assistance outlays). Directly quantifiable benefits under the project would be derived from improvements in village agriculture as calculated from the farm budgets, with the assumption that without the project farmers would 41- tivate a similar area using traditional practices. 7.02 On the basis of these assumptions, the economic rate of return wculd be 22%. This relatively high rate is due essentially to the very low initial level of agricultural productivity among peasant farmers in the Kigoma region, and the opportunity to reach them in an effective way for the first time. with the concentration of population into the ujamaa villages, In this regard, it is important to allow for the effect of the substantial investments in village social infrastructure provided under the project, since, without the early prospect of such services being available, it is unlikely that many villages would attract settlers in the first place or could retain them through the settlement period. Moreover, the long-run development of ujamaa villages will require on the part of villagers higher general levels of education and health which the social services under the project would help to bring about. There is on this account a plausible case for associating the social elements with the productive elements under the project in a truly integrated way, but it is not possible to evaluate the package in its entirety using rate of return criteria because of the difficulty in quantifying direct benefits from the use of the social service facilities. This question is further considered in Annex 16. However, even if the direct benefits due to social services no more than paid for the costs of providing them, the rate of return under the project would still be 18%. Total capital costs per village are roughly US$75,000 (including an allocation of regional level capital expendi- tures), less than US$50 per capita. About US$50,000 or two-thirds are for social service facilities. 7.03 A relatively high rate of return should be read in the context of the high risks associated with: the experimental and untried nature of ujamaa settlement in Kigoma region, particularly as regards agricultural developinept; and the need to develop a complete input delivery, credit and marketing sys em within the region, to serve the project villages, beginning from a situatioi in which the capacity to provide these services is virtually zero. 7.04 Either directly, because of a failure in village management, or indirectly, because of inefficiencies in marketing or overall administration, the major risk to the project relates to a failure of agricultural productioa - 28 - at the village level. One possible outcome would be where all inputs are provided, but agricultural yields are less than predicted. If gross yields are 20% below those expected, the effect would be to reduce the anticipated rate of return to 5%; with a loss of 25% the project becomes economically worthless. A second type of failure would be where some villages fail entirely and fall apart, while others perform according to expectations. While more disruptive socially perhaps, this type of failure is less serious in terms of impact on the overall rate of return because some inputs are saved. Thus, if 30% of project villages fail after the third year of settlement, the rate lof return under the project could still be 15%. Technical assistance, streng- Ithening of institutions and training would be provided under the project to reduce these risks and the project's potential to improve the incomes of large numbers of poor in a poor region makes the risks worth taking. 7.05 The estimated pre-project income among the bulk of the village families that would benefit directly under the project is the equivalent of US$20 per capita. This compares with an estimated household income for Tanzania as a whole of perhaps US$90-100 capita or 4 to 5 times higher. The project is therefore directed toward meeting the needs of a relatively poor part of the population of Tanzania. 7.06 It is further anticipated as a result of improved agricultural prac- tices under the project that agricultural labor input per farm family would in- crease by 25%. VIII. RECOMMENDATIONS 8.01 During negotiations assurances were obtained from Government that: (a) Village water supply systems would be designed according to a least-cost approach as outlined in para 4.03; further, it would undertake a study under the project into the fiscal implications of the water supply component and appropriate methods of financing the cost of water supply and would consult with IDA on the findings of the study (para 4.04); (b) within six months of signing, it would select a suitable site and commence construction for the regional trials and train- ing center, as outlined in para 4.12; (c) within six months of the credit's effectiveness the operations manager, financial controller and training superviser of the regional cooperative union, and the land use planner and trials and training officers would be hired, as outlined in para 4.15; (d) it would follow auditing procedures as described in para 4.23; - 29 - (e) it would plan project villages and consult with IDA as outlined in paras. 5.02 through 5.05; (f) within six months of effectiveness it would carry out a financial reorganization of the Union, as outlined in para 5.12; and, (g) it would discuss with IDA from time to time the adequacy of cooperative levies (para. 5.15). 8.02 During negotiations, assurances were obtained from TRDB that it would strengthen its Kigoma branch office as necessary and follow lending practices and procedures as outlined in para 5.14. 8.03 A condition of effectiveness of the credit would be that a sub- sidiary loan agreement satisfactory to IDA between Government and TFRB had been executed (para. 4.19). 8.04 The project would be suitable for an IDA credit of US$10 million. ANNEX 1 Page 1 TANZANIA KIGOMA RURAL DEVELOPMENT PROJECT Cooperatives Introduction 1. Cooperatives exist in large numbers throughout Tanzania. By the end of July 1973, there were 2,200 registered cooperatives, including the Cooperative Union of Tanganyika, the Savings and Credit League, 25 cooperative unions, 1,518 cooperative marketing societies, 265 cooperative production and marketing societies, 314 savings and credit societies, and 78 cooperative consumer societies. The existing cooperative structure is in the process of change. The present 25 cooperative unions are to be consolidated and reduced to 18, one for each region. In addition, cooperative societies are to be closely integrated within the program of ujamaa village development. wherein ujamaa villages are expected eventually to become multipurpose cooperative societies, ultimately providing the full range of productive, marketing, financial and consumer services to their members. The existing 265 cooperative production and marketing societies are, in fact, ujamaa villages that have already undergone this transformation. Moreover, it is expected that ujamaa villages, and subsequently their counterparts the multipurpose cooperative societies, will absorb most existing cooperative marketing societies. 2. The cooperative movement of Tanzania has a considerable history. Voluntary "growers associations" existed as early as the 1920's. In 1932, in the Kilimanjaro area, the first agricultural cooperatives were registered to promote coffee as a cash crop for peasant farmers. The movement soon became very popular and spread quickly to other areas where commercial peasant farming was relatively advanced and involved a wide range of economic activities. By 1952, there were 172 registered societies and by 1961 their numbers had increased to 857. Thus, cooperatives grew up spontaneously in the regions of Moshi, Arusha, Bukoba, Mwanza, Ruvuma, Rungwe and Iringa. Shortlv after independence, the Government decided that cooperatives should be established in all parts of the country where they were previously absent. In subsequent years, the Government exercised an increasing degree of super- vision and control. The Government, for example, now sets many of the margins within which the cooperatives have to operate, controls the hiring and firing of all union officials with a salary above T Sh 650 a month, has assumed, direct responsibility for the management of three unions (Dodoma, Mbeya and Bukoba), controls all cooperative education, has assumed virtually all responsibilities for inspection of cooperative societies, has tramsferred to the unions special tasks previously performed by the private sector and has moved to link inextricably the cooperative movement with ujamaa village development. The cooperative movement has thus evolved from a spontaneous effort designed to provide the benefits inherent in economies of scale to its member farmers, into a program initiated, supervised and controlled by Government. ANNEX 1 Page 2 3. The principal responsibility of the cooperative unions, apart from various conasmer, financial, transport and other services they might perform, is to move the marketable surplus of virtually all scheduled crops in virtually all regions of the country from the primary societies to the appropriate mar- keting boards. Affiliated primary societies, in turn, assist the unions in this task by acting as their collecting agents. The cooperatives have been assigned sole buying rights for most crops. However, agricultural production has sometimes suffered from such monopolistic practices, because the coop- eratives have often been marked by poor management, petty corruption and low marketing efficiency. Problems arise, in particular, from an acute shortage of management and accounting skills, an absence of personal accountability within the system and the frequent allocation of unrealistic coastings and margins. Legislation, Organization, and Supervision 4. The first Cooperative Societies Ordinance, stipulating the legal basis for all cooperatives, was enacted in 1932. It was subsequently repealed and replaced by the Cooperative Societies Ordinance of 1968. This legislation is soon to be amended in order to specify the process and criteria by which ujamaa villages are to become multipurpose cooperative societies. As the amended legislation would reflect, it has been suggested that the registration of primary societies would observe three stages, whereby the ujamaa village would be: (i) a probationary multipurpose cooperative society, with full legal status, but with unlimited liability and dependence on the directives of the Registrar of Cooperatives; (ii) a fully registered multipurpose cooperative society, with complete legal independence and status and limited liability; and (iii) in addition, a full-fledged community with powers to mete laws and impose justice. 5. The cooperative structure in Tanzania consists of an apex organiza- tion, The Cooperative Union of Tanganyika (CUT), regional cooperative unions, and constituent primary cooperative societies. CUT's functions as a repre- sentative of the cooperative movement--exercised by agreement with the unions, not by law--appear to have been eroded in recent years. CUT has provided and does still provide advisory, educational and legal services to the countrv's cooperatives. However, many of its functions have been transferred t*._ CGovernment, e.g., CUT's education wing, the Cooperative Education Center, has been absorbed by the Cooperative College at Moshi and many of its legal and advisory activities appear to be shifting to the Prime Minister's Office. 6. Following the Government reorganization of July 1971, the Rural Development Division, headed by the Commissioner for Rural Development, was established to deal with matters relating to cooperatives. The Division contains two principal subdivisions, one on Ujamaa and Cooperatives, and the other on Rural Development Operation and Control, each headed by the respective assistant commissioner (see Chart 1). This central cadre is supported in each ANNEX 1 Page 3 of the 18 Regional Administrative Governments by an ujamaa and cooperative functional manager and his staff. Regional staff consists of ujamaa and cooperative development trainees, assistants and officers, all of whom provide advice, supervision and inspection to ujamaa villages and cooperative societies, both from headquarters and in the field. The Commissioner of Rural Development also serves as The Registrar of Cooperatives, responsible for the registration of all cooperatives. At the regional level, the Regional Development Directors have recently been named Assistant Registrars of Cooperatives and have been charged with registration of all multipurpose cooperative societies within their respective regions. Other institutions within the Prime Minister's Office dealing with the cooperatives include The Unified Cooperative Services Commission and the Audit and Supervision Fund. The former is charged with the recruitment, appointment (including promotions and transfers) and termi- nation of all cooperative personnel earning T Sh 650 per month or more. The latter is responsible for the audit and supervision of the financial records and the preparation of financial statements of all registered cooperatives. 7. The Audit and Supervision Fund currently has a network of 7 zonal offices distributed throughout the country, which it plans to expand into 18 regional offices by December 1974. Total staff consists of 135 personnel, of which only 9 are qualified accountants. Approximately 10 percent of the staff is expatriate, projected to be phased out by June 1976, as qualified Tanzanian staff become available following the necessary training at either the Cooperative College at Moshi or the Institute of Development Management at Morogoro. Only about 50 staff members are currently qualified to audit and supervise the financial accounts of primary societies and but a fraction of these the financial accounts of the cooperative unions. Each staff member could audit approximately 10 societies per month if the accounts were properly maintained, but most societies are insolvent and their books in total dis- array, requiring the accounts first to be reconstructed before they can be audited. To simplify and standardize accounting practices at the society level, a simplified, uniform bookkeeping system for ujamaa villages is in the final stages of preparation. In summary, problems of audit and super- vision of the cooperatives can be expected to persist throughout the country during the next five years. Financial accounts of both primary societies and cooperative unions are rarely prepared adequately for audit and super- vision. Moreover, neither the quality nor quantity of staff of the Audit and Supervision Fund is sufficient to respond satisfactorily to the large task required.- Education and Training 8. The extensive network of ujamaa villages and cooperatives, and the supporting programs and institutions, draw upon a large number and variety of personnel, all of whom require training and education. The only national institutions qualified to provide such services are the Cooperative College at Moshi and the Institute of Development Management at Morogoro, still under construction. The Cooperative College currently offers diploma, final ANNEX 1 Page 4 inspectors, and induction courses to a total of 272 students. The Institute of Development Management presently has 240 students enrolled in a four-year diploma accountancy course, to be doubled upon completion of building con- struction. In addition, rural training centers, correspondence courses, and regional educational programs contribute to education and training at the local level. 9. Under the first cooperative education plan, 287 staff and leaders of primary societies were trained, including 81 society secretaries. An ujamas village secretaries' and treasurers' course was offered to 242 personnel. About 278 staff of the cooperative unions participated in different courses of two to ten weeks' duration. To upgrade staff of the Ujamaa and Cooperative Development Division, a basic inspectors' course was given to 270 personnel and a final inspectors' course of six months to a smaller number of qualified staff. About 50 persons from the cooperative unions, the parastatals and Government completed the cooperative diploma course. Finally, a series of brief courses to cooperative members and leaders were given to over 20,000 persons in 1970 and 12,000 in 1971. In 1970 and 1971, respectively, 712 and 1,052 individuals successfully completed cooperative correspondence courses. 10. Past efforts have, however, only begun to address the pressing needs to train and educate staff of the primary cooperative societies, the cooperative unions, and the Rural Development Division. The ujamaa and cooperative education plan (1973-1979) therefore proposes to establish over the next two years five more zonal cooperative colleges as branches of the Cooperative College at Moshi. The Zonal Colleges will each have an eventual capacity of 150 students, will each require at least six teachers, and will provide courses for primary societies' secretaries, basic and final manage- ment and administration, and basic and final inspectors. Between 1973/74 and 1978/79, it is estimated that 6,900 staff and leaders of ujamaa villages and 6,300 of other primary societies will be trained. In addition, training for approximately 650 union accountants and assistants and 1,100 senior and Junior bookkeepers is projected over the period. 11. Finally, substantial cooperative technical assistance is currently being provided by the five Nordic countries. Of the 30 experts now operating -id"- a2 agreement originally signed in 1968, six are working within the Rural Development Division to establish systems for wholesaling and transport, credit financing, management accounting, promotion and supervision and con- sumer and retail activities, 11 are working on the specific problems of 11 regional cooperative unions, and the remainder are working in the Cooperative and Zonal Colleges. It is intended that the systems developed at the center (transport and primary society bookkeeping systems are near completion) will be both integrated into the cooperative college curriculum and incorporated within the operations of the cooperative unions. ANNEX 1 Page 5 Kigoma Region Coeuerative Organization 12. In Kigoma region, a staff attached to the regional and district Ujamaa and Cooperative Development Offices of approximately 100 personnel provides support, advice, and supervision to ujamaa villages and cooperative societies. The staff is divided as follows: 7 in the regional office, 32 in the Kigoma district office, 33 in the Kasula district office, and 30 in the Kibondo district office. Of these, 8 personnel are employed in the 2 rural training centers in Kasulu and Kibondo, both of which fall under the administ- ration of the Ujamaa and Cooperative Development Office. In addition, there are currently 57 rural development assistants (RDAs) who operate at the village level to mobilize and educate the rural populations in ujarna prin- ciples, to assist villages resettlement, and to coordinate the various government activities. Moreover, approximately 15 cooperative inspectors (CIs) now operate out of the district offices to supervise and inspect the financial performance of the primary societies. However, very few are adequately trained, particularly in accountancy and bookkeeping techtniques, a fact which has contributed to the adverse financial situation of the region's cooperative system. 13. The posts of both RDAs and CIs are now to be transfonmed into that of ujamaa and cooperative development assistant (UCDA). The task of the UCDA will be to mobilize, organize, and train villagers in the skills of forming and managing ujamaa villages and to promote, supervise and counsel ujamaa cooperative societies. RDAs are to be upgraded through correspondence courses and subsequent coursework at the zonal cooperative colleges to fulfill the tasks expected of UCDAs. The Government plans to post eventually one UCDA in each ujamaa village. 14. There are at present in Kigoma region 6 consumer cooperative societies, 5 savings and credit societies, 14 cooperative marketing societies, and 1 cooperative producer society. The 15 marketing and producer societies affiliated with the Kigoma Regional Cooperative Union are estimated to represent a member- ship of approximately 5,000 (out of a rural population of about 100,000 families). The marketing societies handle a variety of crops, particularly paddy, cotton and palm oil, most of which is channeled through the Union (see Table 1). The societies, however, are generally poorly managed and the majority are insolvent. Mbreover, the bulk of agricultural surplus produced by their members, parti- cularly maize and beans, is marketed outside the cooperative system. 15. The cooperative system within the region is now undergoing substantial transformation, as ujamaa villages are established and as multipurpose primary cooperative societies are subsequently formed. The majority of existing market- ing cooperatives are expected to be absorbed by the multipurpose cooperative societies, wyherever the respective memberships overlap. It is planned that ai network of multipurpose primary societies would emerge, whose firm financial and operational management would be assured from the outset, and which would eventually take on the full range of marketing, savings, credit, and consumer functions. ANhNTEX 1 ?age 6 Primarv Societv Registration 16. Several measures, of which registration procedures are the most -niortant, have been planned to assure the viability of the new multipurpose Drimary societies. For a ujamaa village to be fully registered as a multi- Durpose primary cooperative society, the following steps must be taken: (a) an individual constitution and by-laws must be prepared and approved by the villagers; (b) an application form must be orepared and an entrance fee paid; (c) an economic feasibility report must be prepared by the Ujamaa and Coope- rative Development Office; (d) The District Development Conmitcee must approve each application; and (e) all material must be forwardea to the Regional Development Director, who, in his capacity as Assistant Registrar of Coopera- tives, would approve final registration. The economic feasibility report must examine size of the village (minimum size is presently 150 families), planned and Dotential economic activities, quality of leadership, and proven ability to implement programs. At the beginning of each year the District Development Director and his staff determine which ujamaa villages within the respective district might be eligible for registration during the year; and the registration procedures are subsequently initiated by the district office of Ujamaa and Cooperative Development. There are currently three ujamaa villages registered and 12 in the process of registration. 17. Under the project additional measures are proposed to assure the viability of project villages. As a condition for lending by TRDB, each village would be required to have a qualified bookkeeper and agricultural field assistant (AFA). To fulfill the bookkeeping requirements, a UCDA, with the requisite training, would be placed in each creditworthy village for up to two years. During this period he would help to train a bookkeeper, who would be selected by the Village Development Committee (VDC) and paid from the income of the primary society. A Standard VII education would be required. In addition, both the village bookeeper and the UCDA would be trained in elementary bookkeeping, in courses following the curriculum to be laid down by the credit and training supervisor of the Union, to be held in the rural training centers. The curriculum would follow wherever possible that of the zonal cooperative colleges. The training in bookkeeping and accounting pro- cedures should be sufficient for preparation of ledger accounts, trial balances and monthly income statements. Training of the AFAs would be provided within the program of the proposed Agricultural Trials and Training Center. Kigoma Regional Cooperative Unicon 18. The Kigoma Regional Cooperative Union was registered (No. 1463) in 1965 under the Cooperative Societies Ordinance. It acts primarily to market the produce of its 15 affiliated primary societies. It also provides trans- port services, including the distribution of beer and cooking oil within the region. The Union's total fixed assets, including two lorries and a Landrover, currently do not exceed T Sh 150,000. All office and storage space is rented. At present, it has a staff of eight personnel, excluding those required for transport, although an additional five posts have recently been allocated. ANNEX 1 Page 7 19. The Union s procedures for organization and management follow the instructions recently issued by the Rural Development Division. These instructions attempt to improve the quality and delineate the activities of the membership, management committee, and management, and in particular, to prevent the corruption that has undermined parts of the cooperative movement in the past, including the Kigoma Cooperative Union. The General Meeting of the Union, in which generally two to three members of each affiliated society participate, usually convenes once a year to review outstanding Union issues and policy and, in particular, to discuss management's recommendations on prices and wholesaling costs for crops handled by the Union in the coming year. The Union has an Executive Committee, which meets every month, con- sisting of the general manager, a chairman elected by the General Meeting every three years, and a vice-chairman elected each year. In addition, there is a Management Committee, which meets every three months, consisting of ten Members, one-third of whom must resign or stand for reelection each year. All committee members must be approved first by both TANU (Tanzania's political party) and the regional administration before standing for elections. At present, the posts of chairman and vice-chairman and four positions on the management committee are vacant, following a recent reorganization of the Union. 20. The Union's past financial performance has been poor. The last audited accounts for the Union were for the financial year ending April 30, 1972. At that date the Union was suffering a negative balance of T Sh 509,775, as compared to T Sh 932,000 for fiscal 1970/71 and T Sh 736,250 for fiscal 1969/70. In fiscal year 1971/72 current liabilities amounted to T Sh 1.5 million, of which about T Sh 1 million represented bank overdrafts, and current assets amounted to only T Sh 980,000, at least part of which is of questionable value. In fiscal year 1972/73, it is estimated that interest charges alone amounted to T Sh 137,000 and that the NBC overdraft at year's end was almost T Sh 2 million and that outstanding obligations from affiliated societies were approximately T Sh 1.8 million. This persistent insolvency appears to result from two principal causes: (a) a continuing rise in debt balances from primary societies and concomitant increases in the Union's NBC overdraft, arising from unrepaid cash advances for crop purchases; and (b) a perennial operating deficit. 21. To improve the UInion's performance several measures are recommended prior to and during project implementation: (a) the completed audit of the Union's 1972/73 financial statement by time of negotiations; (b) the partial or total write-off and/or rescheduling of the Union's persistent outstanding debt; (c) the substantial strengthening of Union organization and management, including hiring of a new general manager, operations manager, financial controller, credit and training supervisor (see Annex 9), and society credit officers; (d) the establishment of new trading and lending procedures (see Annex 14), in particular, carefully controlling the advancement of cash for crop purchases; and (e) the imposition of similar trading and lending procedures for all non-project affiliated societies, including hiring of society book- keepers. ANNEX 1 Page 8 22. Critical to the efficient disbursement, supervision, and control of NBC cash advances would be the SCOs proposed under the project. In the past, cash has been advanced through the Union to affiliated primary societies for crop purchases without adequate control or supervision. This has resulted in large increasing debts from the societies to the Union and frou the Union to NBC. To prevent such problems in the future, new lending procedures are proposed which rely heavily on the SCOs. The SCOs would control the advance- ment of cash and supervise its use during regular visits to primary societies. In addition, they would regularly report on the financial situation of their respective societies to Union management, and ultimately to NBC and TRDB. To prepare qualified individuals to undertake these functions effectively, the following training is proposed. Form IV leavers would be recruited for training at the Institute of Financial Management, where they would undertake a program similar to that now followed by TRDB's credit supervisors. The course involves 18 months of training in management, economics, accountancy, and related subjects, to which might be added some training in relevant crops by the appropriate marketing boards. Once working for the Union, SCOs should receive salaries commensurate with their high qualifications. The Unified Cooperative Services Comission, or other government entity, might participate in the training and subsequent remuneration of SCOs. TANZANIA KlGOMI RURa DEVELOPSIIT PROJECT ORGAtilZATIONAL CHART OF RURAL DEVELOPMENT DIVIESION Prime Minister's Office Rural Development Division Ujamaa and Cooperative Sub-division Rural Development Operation and Control Sub-division Promotion and Supervision Planning and Development Development Training of Ujamaa and Cooperative| Research Section Operational-Control Section Societies Section Section Audit and Planning and Staff Ujamaa and Supervision Performance Cooperative Training Services Industiial Farm Management Home Economics Coopers ive Analysis and Women .Jea4gMejtL aining Services WhoJesi.te ad Marketing Rural Construction Retail t Coordination Analysis Training Services J | Statistics and i s L R-eporting Lsand Servicing Team Publicity and Radio | Program C | TANZANIA KIGOMA RURAL DEVELOPHENT PROJECT Kigoma Regional Cooperative Union Affiliated Primary Societies and Produce Handled Type of Produce Nguruka Kigoma/Ujiji Kaseke Ukaranga Uvinza Mgambo Mwandiga Kalya Ilagala Kibondo Kasulu Coffee Manyoru Makere Buhoro P-it Paddy x x x I . Cotton x x x x X Coffee x I X x Maize YI Y Palm Oil X x x x_ _ Palm Kernel Xx Y x _ . _ _ Bees Wax x . X Pidgeon Peas x X Beans x Y I X X X X Groundnuts X X _ Simsim X X __ Millet (White) _ X Cassawa Castor Seed x _ . , Sun flower Cashew nuts X SOURCE: Regional Administration ANNEX 1 Table 2 TANZANIA KIGOMA RIUL DEVEIL)P T PROJECT Kizvaa Regional Cooperative Union Union Produce Purchases - 1971-72 Crop Volume (toas) Value (Tah 000's) 1. Paddy 755 572 B. Bs Wax 49 441 p Cotton 325 32B 4. -Palm Kernels 592 302 5. Yellow Bean. 141 92 6. Pas-I Oil 23 68 7. Pidgeon Peas 175 67 8. Groundnuts 42 50 9. Coffee 3 45 i0. Mixed Beans 71 39 11. Millet 98 29 12. Maize 35 14 13. Six Siji 10 214 Aii. Castor Seeds 1 1 15. Cashewnuts I 2321 2063 SOURCE: Audited Accounts of Xigoma Cooperative Union. ANNEX 2 Page 1 TANZANIA KIGOMA RURAL DEVEIPMENT PROJECT The Tanzania Rural Development Bank (TRDB) Resources of TRDB 1. TRDB has an authorized capital of T Sh 100 million, divided in 100 equal shares, to be subscribed exclusively by the Government. The initial subscribed share capital of T Sh 25 million, representing the tangible assets transferred from NDCA (TRDB's predecessor), was raised through successive Government contributions and finally reached T Sh 47 million by the end of 1972. General reserves as of December 31, 1972 were about T Sh 0.7 million, as compared to about 1.7 million at the end of the previous year. This reduction was mainly due to the loss of almost T Sh 1 million recorded for the half-year ending December 1972. 2. Other resources of TRDB comprise short-term deposits of T Sh 26.2 million, 1/ as well as medium and long-term loans totalling approximately T Sh 37 million. Besides a number of IDA credits, the latter include a National Bank of Commerce overdraft of T Sh 3.3 million, loans from SIDA for grain storage facilities and trucks, and a loan from the ICO for the local purchase of seasonal inputs. The bulk of TRDB 's funds available for lending to date have originated from the five IDA credits extended to Tanzania for agricultural projects over the last seven years. Date of Credit Amount Disbursed Credit Credit Amount as of Feb. 28, 1973 No. Purpose Agreement (US$ million) (US$ million) 80-TA General Agriculture Credit Jan. 1966 5.00 5.00 132-TA Beef Ranching Development Oct. 1968 1.30 1.20 217-TA Small Farmer Flue-Cured Tobacco Oct. 1970 9.00 0.55 287-TA Smallholder Tea Mar. 1972 10.80 0.65 382-TA Livestock Development May 1973 18.50 - Total 44.60 7.40 P.s of December 31, 1972, IDA funds represented almost 80% of TRDB's medium- and long-term liabilities with a total of T Sh 28.5 million. 1_ Including T Sh 20 million reserved by the Lint and Seed Marketing Board for cotton purchases. ANNEX 2 Page 2 3. TRDB is also expected to receive special funds from the Government, bearing the name of "soft window," to be administered separately from ordinary capital resources and used for the purpose of extending loans on concessionary terms to finance special development projects. To date, this procedure has not been used by TRDB. Lending Operations 4. TRDB provides short and medium-term loans generally at an interest rate of 8-1/2% and long-term loans at 7-1/2%. Maturities are usually based on the productive capacity and estimated life of the assets to be financed and a rough distinction can be made between the following categories: (a) Short-term loans for agricultural inputs due for repayment within 12 months from the date of disbursement. (b) Medium-term loans repayable over a period of five years. These loans are mainly for farm implements and machinery, trucks, piggeries, poultry farms, fishing boats and equipment and small-scale industries. (c) Long-term loans with a maturity of up to 15 years. These cover major investments such as produce godowns, buildings, coffee pulperies, tea factories, ranches, etc. 5. As a matter of policy, TRDB does not normally finance more than 75% of total project cost and a loan for any single project should not exceed 10% of the Bank's net worth. Eligible borrowers are mainly the Cooperative Union and Societies, the District Development Corporations, the ujamaa villages and registered associations. Crop financing remains the prerogative of the National Bank of Commerce and the various agricultural produce Marketing Boards. 6. TRDB's outstanding loans as of December 31, 1972 amounted to T Sh 134 million, of which 121 million or 90% were with Cooperative Societies. The composition of the outstanding cooperative loan portfolio at the time was as follows: Table 1 Short-term loans : T Sh 69 million or 57% Medium-term loans : T Sh 34 million or 28% Long-term loans : T Sh 18 million or 15% Short-term loans cover seasonal inputs for a number of crops; tobacco, tea, coffee, and cotton currently represent about 95% of the short-term outstanding portfolio. Tobacco is by far the most important crop for TRDB with a share of about 40%. The medium-term loan portfolio is heavy geared toward vehicles (usually trucks), which account for almost 85% of the medium-term funds out- standing, and the remainder is equally shared by tractors and spray equipment. ANNEX 2 Page 3 Long-term loans in the past have mainly been for the construction of godowns, buildings, coffee pulperies, tea factories and coffee rehabilitation. These items represented about 85% of the long-term outstanding portfolio at the end of 1972. 7. Four of the eighteen regions--Iringa, Tabora, Mbeya and Mwanza-- currently hold almost 60% of the total outstanding loan portfolio, while other regions such as Central, Kigoma and Morogoro are hardly represented at all. Investments from IDA funds represent about 20% of the total amount outstanding and are largely in the form of short-term loans for tea, coffee and cotton inputs. Arrears Situation and Credit Recovery 8. The accumulation of overdues has been a constant feature of TRDB's operations over the last three years, and at the end of December 1972, these amounted to T Sh 35.6 million, or 26.5% of the loan portfolio. This situation arose mainly from bad debts on medium and long-term loans inherited by TRDB from its predecessors; overdues on the loans extended by TRDB itself were only 7% of the outstanding amount in February 1973. 9. Pursuant to IDA's request that a statement of measures to reduce the amount of arrears be prepared, TRDB recently submitted a detailed proposal for rescheduling and writing off a number of bad debts. The proposal received IDA's approval and loans amounting to T Sh 9.9 million were rescheduled, while a balance of T Sh 4.4 million was written off. This combined with a large collection of overdues totalling T Sh 7.2 million during the first quarter of 1973, reduced total arrears to T Sh 11.3 million (that is, 9% of the new outstending portfolio), fully covered by the provision for bad debts. 10. Along with the above measures, TRDB expressed its intention to improve credit recovery in the future. This would be achieved through a series of measures including claims against stocks held by the delinquent unions, repossession of vehicles, seizure of cattle, etc. In some cases, TRDB would seek to obtain a lien on crop sales to the National Agricultural Product Marketing Board (NAPB) or on fees paid by NAPB to the Unions for the rental of godowns. TRDB also plans to take court action against societies which refuse to pay in spite of their good liquidity position. The determi- nation of TRDB to recover its overdues is beyond doubt,. but its success hinges upon the backing which the Government will be willing to provide. Organizational Structure and Management 11. The major element affecting TRDB's structure has been the regional- ization of the country. TRDB's former 58 field offices were replaced by 18 !-. Ioral offices, each having a regional representative and a credit super- visor. The responsibilities of the new regional offices include pre- investment studies, project appraisal and supervision, technical assistance and the maintenance of permanent contacts with borrowers, regional repre- sentatives of Ministries and branches of the National Bank of Commerce. ANNEX 2 Page 4 As a result of this reorganization, the operations of TRDB will undoubtedly gain in effectiveness, and some improvements in credit recovery may be expected. On the other hand, the staffing and operation of such offices vll be an additional burden on TRDB's already strained financial resources, since the regional allocation of staff has been made irrespective of the nature of the Bank's involvement in any given region. The overall increase in staff was from 90 before the regionalization to about 180 today. A recent development, which should have a positive bearing on TRDB's future operations, was the appointment of six UNDP specialists who will reinforce Management's skill in a number of key areas: namely, farm management, rural credit, marketing, livestock and agricultural machinery. All six specialists have now joined TRDB and are expected to play a key role in project appraisal. ANNEX 3 Page 1 TANZANIA KIGOMA RURAL DEVELOPMENT PROJECT Agriculture in Kigoma Region Climate, Soils and Ecology 1. A variable land form characterizes the region with elevations ranging from 1,050 m to 2,300 m above sea level. Most of the region forms part of the Central Plateau, an immense, undulating peneplain with a profile of gentle slopes and shallow valleys with large areas of swamp bordering the main water courses. Elevation increases gradually to the northwest, then abruptly into the Kibondo and Kasulu highlands which border Burundi. The Malagarasi River, which rises in the highlands north of Kigoma and forms the Tanzania-Burundi boundary for about 160 km, drains most of Kigoma Region. 2. Annual rainfall in the region is between 500 and 1,100 mm (Table 1) and the annual mean maximum and minimum temperatures range from 28° to 12°C respectively. A single rainy season lasts for six to seven months from November to early May followed by a prolonged dry season. The Malagarasi swamp, Lake Tanganyika and the highlands act as climate modifiers and, in the highlands, this results in higher annual rainfall and slightly lower annual maximum and minimum temperatures than those that prevail in lower lying areas. Precipitation is reliable and allows a wide range of crops to be cultivated with some double planting of short-season crops. 3. Climate and topography, together with the effects of past and present human settlement, give rise to three distinct agro-ecological zones (Map 10942): (a) The Highland Zone. A continuation of the Burundi mountains, which follows the northwestern boundary of the region, is the wettest of the three. The Zone has a rolling to hilly physiography with incised seasonal streams and rivers. Elevation ranges from 1,500 m to 2,300 m, becoming higher, hillier and generally more dissected near the Burundi border. Annual rainfall is normally above 1,000 mm being highest on the northern slopes. For the most part, soils are well drained, dark red to reddish brown sandy clay loams. Hill crests often have shallow soils and there is severe rill and gully erosion in areas of high population pressure. The climax vegetation has long since been eradicated, but there are some remnants of Brachystegia woodland, and higher levels were probably under broad-leaved forest. Medium to tall Hyparrhenia-Sporobolus grassland covers much of the ANNEX 3 Page 2 Zone and in the north there are areas of Hyparrhenia- Combretum bushed grassland. Glades in the highland forests were probably the first settled areas of Kigoma region, but most of the Zone has now been degraded by over- cultivation and soils are heavily depleted of phosphorous and nitrogen. The Zone covers roughly 510,000 ha of which perhaps 20% is cultivable. Little unoccupied land suitable for cultivation remains and holdings are small (about 1 ha). A wide range of crops is grown, of which maize, beans, and bananas are the most important. Coffee is grown in the higher, wetter areas. Most of the region's cattle are concentrated in the Highland Zone. (b) The Intermediate Zone lies between the densely populated highlands and the almost uninhabited Miombo woodlands. The Zone is narrow and has an undulating physiography with occasional hills and a predominance of perennial streams and rivers. Elevation ranges from 1,200 m to 1,500 m with an annual rainfall of 850 mm to 1,100 mm. Soils that are dusky red to darkish red sandy clay loams predominate and there is slight gully erosion on the steeper slopes. Overall, the Zone is sparsely populated though out-migration from the Highland Zone has concentrated in some localized areas. A climax vegetation of Brachystegia Julbernardia woodland covers much of the Zone with a tall understory of Panicum maximum and Hyparrhenia spp. Soils are moderately fertile but significant responses to phosphorous and nitrogen applications are expected. The Zone covers roughly 300,000 ha of which an estimated 50% is cultivable. Settlement, however, has been restricted by the incidence of tsetse fly, Clossina morsitans, which is common in the uncleared Brachystegia woodland. There are also a number of endemic sleeping sickness areas (Map 10942). Holding sizes are larger than in the Righland Zone (1.5 to 2.0 ha) with a larger proportion of the cultivated areas left to fallow after four to five years of cropping. Most of the population live in scattered homesteads and until recently, there were few nucleated villages. Maize and cassava are the major subsistence crops; beans, bananas, groundnuts, sweet potatoes and a range of other vegetable crops are also grown and in recent years cotton has been promoted by Government staff. In the Intermediate zone cattle and small stock ownership is less common than in the Highland Zone. ANLNEX 3 Page 3 (c) The Miombo Zone covers most of the east and southeast areas of the region and is characterized by an almost featureless, gently undulating peneplain with occasional small hills and large areas of swampy depressions. Streams and rivers from the highland areas converge to form wide, shallow river valleys; in this Zone perennial water supplies are less accessible. The ranges in elevation are from 1,000 m to 1,200 m. Few climatic recordings have been made in this Zone, but annual precipitation is expected to vary between 600 mm and 1,000 mm and is known to be reliable. Soil on the upper slopes and crests go from being deep, well drained dusky red sandy loams to sandy clay loams lying over dark red clay loams. The lower slopes and depressions have imperfectly drained dark grey to black clays. Brachystegia Julbernardia woodland with an understory of tall Panicum maximum and Fyparrhenia spp. covers most of the Zone, but there are some unforested alluvial pockets that have attracted human settlement. Soils of the Miombo belt are typically deficient in nitrogen and phosphorous, but the region has large areas of a typical, tall, thick trunked woodland with soils of moderate to good fertility offering a sound potential for agriculture. Tsetse infestation throtighout the Zone has restricted migration from areas of high population pressure, although some villages have been established during previous regional resettlement programs, and there are scattered settlements along the line of rail. Traditionally, the Zone has been important for honey gathering, hunting and fishing. CROP PRODUCTION 4. Maize, cassava, beans and bananas are the staple foods in the project area, supplemented by such minor crops as sweet potatoes, sorghum, pigeon peas and groundnuts. Cash incomes for farmers are largely derived from sales of these crops surplus to consumption needs; however, coffee, oil palm products and cotton are of localized importance and al:so make a small contribution to the economy. Flue-cured tobacco is also under inves- tigation as a cash crop at two sites in the Kigoma district. 5. Data on crop production and marketing are extremely limited. Results of a crop yield survey carried out during the 1972 National Agricul- tural Census are not yet available. A small proportion of production passes through cooperative marketing channels (no more than about 2,500 tons per year) and a larger, uncontrolled and unquantified export trade is carried out by private traders to neighboring Burundi and Zaire. ANNEX 3 Page 4 INSTITUTIONS Ministry of Agriculture (Kilimo) 6. Responsibility for agricultural development rests mainly with the Ministrv of Agriculture (Kilimo) whose regional staff is divided into crop production and animal health and tsetse control departments controlled by regional functional managers at the Senior Field Officer level. In late 1973 technical staffing of the two departments totalled some 241 located as indicated below: District Regional Kigoma Kasulu Kibondo Total Crop Production Senior Field Officer (SFO) 1 - - - 1 Field Officer (FO) 3 2 2 2 9 Assistant Field Officer (AFO) 6 28 23 17 74 Field Assistant (FA) _ 6 14 17 37 Sub-total 10 36 39 36 121 Animal Ilealth and Tsetse Control Senior Field Officer (SFO) 1 - - - 1 Field Officer (FO) - - - 1 2 Assistant Field Officer (AFO) 3 8 7 9 27 Field Assistant (FA) - 22 35 33 90 Sub-total 4 30 42 43 120 7. Over 50% of total technical staff in Kilimo, i.e., the FA grade, have received little or no formal technical training. They are mainly primary school leavers who have been given some in-service instruction in sp-cific activities, such as maize production, or dip operation. Except ahere specific functions, such as dip operation, are called for their present usefulness is generally very limited. Of the remainder almost 42% are in the AFO grade, a level of staff based on two to three years of technical training after secondary school. Mlost are strangers to the region and thus are initially unfamiliar with its people and agricultural conditions. They are usually young men whose lack of local background compounds their overall lack of experience, further limiting their usefulness as agents for agricul- tural development. Fortunately, at the FO and SFO level, the region has a number of experienced and dedicated men capable of providing support and guidance to lower level staff. However, numbers are small in relation to the task and their impact is lessened by difficulties of communications and ANNEX 3 Page 5 the lack of firm information on which to base recommendations to farmers and hence instructions to subordinate staff. The project would seek to utilize the skill and experience of such staff much more fully by the intro- duction of training and trials programs (see below) and the development of technical backup services for field workers. Tanzania Cotton Authority (TCA) 8. In addition to the regional cooperative union and the Tanzania Rural Development Bank (TRDB) which are described in Annexes 1 and 2 respectively, one other institution, the Tanzania Cotton Authority (TCA), plays a potentially highly significant role in the agricultural development of the region. Established in March 1973 as the successor to the Lint and Seed Marketing Board, TCA is charged, inter alia, with responsibility for promoting the development and improvement of the country's cotton industry. This responsibility includes organization of seed supplies, marketing of seed cotton, ginning, seed crushing and the regulation and control of the marketing and export of cotton lint. 9. Because it is a new organization with considerably wider scope than its predecessor, TCA has had to recruit additional senior personnel from a number of other agencies including Kilimo. It is still substantially under- staffed, particularly at the field level, and in areas such as Kigoma where the crop has been of comparatively minor importance. 10. TCA operates a scheme whereby insecticides, sprayers and fertilizers are distributed to farmers at 50% of delivered cost. Limited supplies of knapsack sprayer and DDT/Sevin insecticide together with some 270 tons of Sulphate of Ammonia were provided for the 1973/74 planting season. Plans for distribution and demonstration were being drafted late in 1973 but arrange- ments for distribution and sale were far from finalized at that time. 11. To summarize, although Kilimo has a comparatively large number of staff in the region, most are either inadequately trained, inexperienced or both. Bad communications and lack of firm recommendations for improved practices compound the problem of providing ujamaa villages with extension staff capable of servicing the needs of villagers. A further complication is the present inability of TCA to support its expansion plans for cotton to any meaningful extent. These facts combine to dictate that initial plans for agricultural development must be based on simple measures that can be seen to pay off and that can be understood and transmitted to farmers by village-level extension workers who have received only limited technical training, as explained below in more detail. TRAINING NEEDS 12. It is anticipated that national plans for training general agricul- turists at AFO level will result in a yearly outturn of about 300 by 1978. Competition from other regions, projects and agricultural parastatal organi- zations such as TCA, is likely to be such that Kigoma region would receive ANNEX 3 Page 6 no more (and probably less) than 5% of outturn, or at most 10-15 AFO's per year. Because of the low level of general education in Kigoma, comparatively few pupils progress to secondary education and hence eligibility for AFO training. Therefore, such AFO's as may be posted to Kigoma are likely to come from other regions and be unfamiliar with its social and agricultural conditions. Thus unfamiliarity is apparent in AFO's recently posted to the region, to the extent that they require a period of at least one year of acclimatization to Kigoma conditions before becoming effective extension workers. The problem of staff shortage and orientation could be overcome by a limited regional training program of village level workers as outlined below. The Kigoma Training and Trials Center 13. The primary objective would be to train village level extension workers of the FA grade in the rudiments of crop husbandry as they apply under Kigoma conditions. Primary school leavers who are natives of Kigoma would be selected for an initial nine to ten month course at the Kigoma Training and Trials Center. The Center, which would be provided under the project, would be centrally located, at the site typical of the Intermediate agro-ecological zone. The training wing would have a capacity of 25 board- ing students and be staffed by one FO and two AFOs with experience of Kigoma conditions. Teaching staff would be seconded from the regional establishment, would continue to be responsible to the regional functional mnanager, and would be assisted from time to time by the staff of the trials wing of the Center on such topics as fertilizer responses. 14. Training would consist of simple classroom work on the elements of crop production, pest and disease control, agricultural mensuration and extension methods, together with extensive practical work during which classroom lessons would be put into effect in the field. The Center would have a training farm made up of a number of units of crop sequences recom- mended for ujamaa village block farms that would be operated by groups of sttudents. The curriculum for basic training of FA's would be developed by regional staff in consultation with education specialists from the Department of Mlanpower Development in Kilimo's Dar es Salaam office. 5. Successful completion of the basic course of training would result $ appointment to the FA grade and posting to an ujamaa village. To the extent possible, newly qualified FA's would spend their first crop season helping more experienced staff. Thereafter each would assume responsibility for promotion of agricultural development in an ujamaa village. 16. Subsequent to their basic training, FA's would receive specific in-service courses on subjects such as cotton spraying and maize cultural practices. These courses would be given by Training Center staff to all village extension personnel on a district-by-district basis. Courses would e timed such that narticipants would apply their lessons immediately on ANNEX 3 Page 7 returning to their villages. Additionally, the Center would be used for short residential courses such as seed cotton grading, seed multiplication and land use planning for Kilimo and TCA staff during the two to three month period between FA courses. Field Assistants who had proved themselves worthy of further training should be considered for acceptance at Central Agricultural Training Institutes (CATI) operated by Kilimo on a national basis. Successful-completion of CATI courses would result in promotion to AFO level with prospects of further advancement. 17. Because of the urgent need to train village level extension staff, establishment of the Training and Trials Center would receive high priority. Regional authorities would undertake to select and provisionally set aside a suitable 250-ha site in the Intermediate agro-ecological zone preferably in Kasulu district. Site selection would be on the basis of a topographic and detailed soil survey which would be made available to IDA within three months of credit signing. 18. As the national training program gathers momentum and the need for regional basic training lessens, the purpose of the training wing would change toward provision of residential short courses for established staff and farmers. Such courses would include orientation for new staff and tovics of particular regional significance based on results obtained at the trials wing of the Center. RESEARCH REQUIREMENTS 19. Kigoma region comes under the jurisdiction of Kilimo's Western Research Station at Ukiriguru some 30 km south of Mwanza, and 350 km north- east of Kigoma. The region has received little attention in the field of crop research and since 1966/67 has not been included in any territorial or district trial programs. Since no comprehensive investigations to test variety and yield responses have been conducted throughout the region's main agro-economic zones, the 1Western Research Center's published trials results that refer to Kigoma are not representative of the region as a whole. Never- theless, the marked responses which have been obtained from improved cultural practices, new varieties and fertilizer applications in similar ecological areas in Tanzania and elsewhere in eastern Africa indicate the considerable scope for crop production development in the region. The Kigoma Training and Trials Center 20. Because there is a dearth of information on all crops currently cultivated, it is important that initial investigations should focus sharply on rro': of major economic and nutritional significance. The trials wing of the Training and Trials Center proposed for inclusion under the project would concentrate primarily on the maize, cotton, bean, and groundnut crops. ANNEX 3 Page 8 Work would be directed towards testing the applicability of recommendations from the Ukiriguru Research Station to Kigoma region conditions by field trials at the proposed Center and at selected sites in the region. In addition, the Center would be used as the Kigoma location for any relevant territorial or related trial programs operated from Ukiriguru. Research requirements for the different crops are outlined below. Maize 21. Of the several composite varieties bred in Tanzania for cultivation above about 900 m altitude, Ukiriguru Composite A (UCA) significantly out- yielded the local variety in Kigoma by over 55Z in the 1966/67 territorial trials. Although no further trials have been carried out since this time, considerable quantities of UCA seed have been distributed to farmers in the region. Performance has apparently been satisfactory as the variety appears to be widely acceptable. Future trials should test new varieties developed at Ukiriguru against Local and UCA. 22. The apparent superiority of UCA over Local on farmers' farms should be further refined by observations on varieties of response to such basic treatments as planting date, spacing, number of weedings and stem borer control. Additionally, fertilizer response studies should be initiated with particular reference to the economics of application and the form of nitro- genous and phosphate fertilizers best suited to the predominantly acid and Vhosphate-deficient soils encountered over much of the Intermediate and Miombo zones. Cotton 23. Current recommendations for planting the UK64 variety in early mid-December are based on trials carried out in the mid-1960s. District variety trials should be reintroduced together with time of planting, spacing and weeding observations. The present reconmendation for pest control using DDT, Sevin and Dimethoate and conventional knapsack sprayers in a six-to- eight spray regime is likely to be superseded shortly. Trials at Ukiriguru have indicated the superiority of endosulfan applied using ultra low volume (ULV) techniques through battery-powered hand sprayers. -This recommendation is incorporated in project proposals, but before widespread application by farmers, the trial wing of the proposed Center would carry out investigations with particular reference to type and timing of insect attack with a view to formulating optimal spraying regimes. Although a chlorinated hydrocarbon, endosulfan does not accumulate in the body and is thus less hazardous to human health than is DDT. 24. No broad cotton fertilizer trials have been conducted in the region. Although research on this matter is required, general indications from trials in comparable situations elsewhere in Tanzania indicate signi- ficant response to nitrogenous and phosphate fertilizers. These should be ANNEX 3 Page 9 investigated under Kigcra conditions particularly to test the feasibility of prolonging maturation by nitrogenous manuring and thus utilizing the totentially long growing season by developing large framed plants with heavw yielding capabilities. Beans 25. There have been no field trials in Kigoma on beans despite their nutritional and economic importance. It is not intended that an elaborate program should be launched, rather, the Trials Officer should consult with appropriate specialists of the Ukiriguru Research Center to design a program of trials relevant to the needs of the region. This might include time of planting, spacing, fertilizer response, and insect control investigations together with some limited variety testing. However, experience elsewhere in this field on beans for local consumption points to strong preferences for indigenous types and a need to concentrate on increasing their potential productivity. Groundnuts 26. The groundnut crop appears to have a promising future in the Intermediate and Miombo zones of the region where it is already a popular minor crop. The local, Mwitunde-type variety has the benefit of a degree of Rosette resistance, but is inferior to the larger seeded Bukene type, recently introduced, on the grounds of kernel quality and yield. Again, the trials program for the crop would be designed in consultation with specialists at Ukiriguru with particular reference to the needs of the Droject. In addition to varietal testing it would need to focus on time of planting, spacing, seed dressing, incidence and control of Cercospora leaf spot, and response to fertilizer, particularly phosphates and possibly sulphur. As in other fertilizer investigations, the economic aspects of application under farmers' conditions would be closely examined before formulation of recommendations and their transmittal to extension staff. Crop Sequences 27. Although trials of this nature would be unlikely to provide meaning- ful results within the life of the project, it is nonetheless imperative that thev should be initiated without delay. Their importance stems from the fact that ujamaaization is founded on permanent settlement and prolonged cropping. Soil-carrying capacity under these circumstances is unknown, but field observa- tions clearly indicate the need to establish viable crop sequences. Design of investigations should draw on work already underway at Ukiriguru. As for other activities at the Center, emphasis should be on simplicity and strict adherence to the needs of the region. ANNEX 3 Page 10 The Trials Wing 28. As indicated above, field trials would be initiated at the Training and Trials Center proposed for inclusion under this project (paragraph 20). They would be complemented to the extent possible by outstation trials at locations in the different districts of the region in order to provide fully meaningful results. The trials wing would be headed by a professional officer with experience of agronomic research in Tanzania and preferably elsewhere. Discussions in Dar es Salaam during appraisal indicated the probable availability of such personnel with in-service training at the Maize and Wheat Improvement Center (CIMMYT) in Mexico. The Trials Officer would be assisted at the Center by three AFO's. The proposed program of out- station trials would be formulated in consultation with the District and Regional Agricultural Managers. The Trials Officer would be professionally responsible to the officer in charge of the Ukiriguru Research Station, al- though his administrative and day-to-day supervision would be provided by the Regional Development Director through the Regional Agricultural Manager. Seed Multiplication 29. In addition to the investigational work proposed, the trials wing would also undertake to provide nucleus stocks of improved seed of maize and groundnuts for distribution and further multiplication at established ujamaa villages, as indicated below (paragraphs 52, 53, 54). It would also provide newly established villages under the project with initial seed requirements for proposed demonstrations. CROP PRODUCTION IN UJAMAA VILLAGES 30. Villages established under Operation Kigoma by late 1973 were just beginning to develop crop production plans at that time. One acre (0.4 ha) house plots had been cleared and planted and block farms were at different stages of clearing in readiness for planting with crops such as maize, beans and cotton in the 1973/74 crop season. Experience in 1972/73 w-i*h communal farms had, with very few exceptions, proved unsatisfactory. Crors had been planted late and poorly tended, and yields were disappointing -r. consequence. As a result most village development councils, with the concurrence of district and regional authorities, opted to change their 01ock farm operations from a communal to an individual basis. In practice this meant the allocation of a one acre plot to each family on each of the block farms proposed for operation by the village. Villagers would then clear, plant and harvest their individual plots and he free to dispose of any surplus over subsistence through the village primary cooperative society. In most cases plot allocation had been made on the basis of contiguous blocks conforming to the 10 family cells. ANNEX 3 Page 1 1 31. Most commonly in 1973/74 two such block farms were to be planted, one with maize and/or maize and beans, and the other with cotton. Attain- ment of the initial targetted block farm plantings of two acres per family appeared unlikely in many cases because of the heavy clearing required and understandable preoccupation with house plots and house building. This problem, however, should diminish over time as villages become established and calls on labor for activities such as water carrying and village estab- lishment are lessened. Project Proposals 32. The organization outlined above, i.e., individual plots on block farms, would be followed in villages financed under the project, but the initial clearing target would be no more than one and one half acres, instead of two acres of block farm per family. Cropping Sequence 33. A suitable cropping sequence has still to be established for the area. For planning purposes in the two lowland zones, where the bulk of resettlement is to be expected, a sequence of cotton then maize followed by late planted beans, groundnuts, and a four-year bush or cassava fallow is proposed. Farmers would, at full development, thus operate their one acre house plots plus one acre each of cotton, maize/beans and one-half to one acre of groundnuts with a possible cassava fallow to follow or a total of three-and-one-half to four acres (1.4 to 1.6 ha). Block Farms 34. All villages that are to be financed under the project would be the subjects of satisfactory Village Site Feasibility Reports (VSFR), which are described in detail in Annex 13. From the viewpoint of cropping acti- vities this would mean an adequate supply of land of a quality and topography suitable for sustained cropping within ready walking distance (say 2 km) of the village living area. Thus a 350-family village would have at least 1,000 ha (350 x 7 ac) of such land in reasonably contiguous blocks. Block farms would be demarcated by the land-use planning staff of Kilimo. It would be allocated by the village development council or an appropriate committee, probably in 10-acre blocks to cell leaders who would then sub- allocate to the individuals within their cells. Labor Availability 35. Local opinion has it that farmers in traditional settlements operate about three to four acres, usually in plots scattered round the homestead. These estimates are supported by a limited survey carried out ANNEX 3 Page 12 in 1968. 1/ The average family size is about five persons, probably equi- valent to a little over two working adult equivalents per family. The size of traditional holdings is probably largely governed by labor availability during peak seasons, such as land preparation, weeding and harvest of the staple maize and bean crops. Introduction of additional crops as proposed under the project would tend to spread labor needs more uniformly, providing more total employment and significantly increased returns to labor. An approximate estimate of labor requirements for a typical family following pronosed project practices is provided in Table 2. This indicates the probability that labor would be available when needed in the crop year vith a surplus for clearing and non-agricultural activities. Cropping Practices Proposed under the Project 36. Earlier sections have indicated the paucity of relevant research data and the generally low level of technical assistance at the village. These considerations together with the fact that most ujamaa villages and their inhabitants have had limited exposure to modern agricultural technology, suggest the need for careful introduction of improved cropping practices. These would be designed to increase productivity, marketable surplus and hence cash income at minimum risk to farmers. In no cases are there specific recommendations for improved practices based on series of field trials carried out in the Kigoma Region. The practices proposed are extrapolated from experi- ence elsewhere. It may be argued with validity that farmers organized with a fair degree of political fervour in a 10-family cell basis, and cultivating the same crop on contiguous plots are much more readily accessible than most small-scale producers and are therefore likely to be highly responsive to innovations. This could lead to higher production than has been estimated. Exnerience over the first years of the project together with the results of field trials and demonstrations will provide the basis for revision of present production estimates. Demonstrations 37. Although it is claimed that most farmers in the region are now generally aware of the benefits of modern inputs, such as improved seed fertilizer and insecticides, field observations indicate a need for demon- strations to show farmers how to maximize input effects. For this reason, s)rovision is made under the Project for the supply of inputs for demonstrations _r ujamaa villages, preferably by selected farmers, under 4v supervision of oxtension staff. Demonstration would be based on packages of practices and inputs as defined by crops below. 1/ Agricultural Economy of Kigoma Region by A. A. Moody, Agricultural Economics Dept., Western Research Centre, Ukiriguru, M?anza 1968, Mimeographed. ANNEX 3 Page 13 Maize 38. Considerable quantities of UCA seed are already in the region. Sta&e I is defined as the use of such seed with no other refinements, thus differentiating between traditional maize cultivation using the local variety. A yield increase owing to variety approximating 25% is assumed, from 535 to 675 lb shelled grain per acre (600-750 kg/ha). Stage II would involve timely planting, use of UCA seed, spacing at three plants per stand at three feet on three-foot ridges, two or three weedings, and dusting with DDT 5% dust against stem borer (Busseola spp). Yields under this regime are estimated at about 1,000 lb shelled grain per acre (1,100 kg/ha) an increment of 325 lb/ac (48%) over Stage I. 39. Stage III would differ from Stage II in the introduction of nitrogen and phosphate fertilizer. Calcium Ammonium Nitrate (CAN) plus Di Ammonium Phosphate (DAP) would be applied, each at 56 lb/ac (63 kg/ha). This is the level proposed for planning purposes but it may well need to be amended following field trials and changes in price relationships during the course of the project. Proposed application levels are estimated to result in yields of 1,700 lb shelled grain per ac (1,900 kg/ha), an increment of 700 lb/ac (70%) over Stage II. 40. Details of estimated effects of these treatments are given in Annex 15 Table 1. Salient points of interest are summarized below on an acreage (ha) basis. Level of Husbandry Maize Traditional Stage I Stage II . e III Yield (lb/ac) 535 675 1,020 1,700 (kg/ha) (600) (720) (1,140) (1,900) Gross Return (T Sh) 85 108 163 272 (per ha) (210) (267) (403) (672) Material Cost (T Sh) 7 8 17 95 (per ha) (17) (20) (42) (235) Labor Cost (mandays) 45 50 60 66 (per ha) (111) (123) (148) (163) Net Return Excluding Labor (T Sh) 78 100 146 177 (per ha) (193) (247) (361) (437) Return to Labor (T Sh/manday) 1.76 2.00 2.43 2.6P 41. For planning purposes the rate of uptake of the different stages of maize cultivation has been estimated as follows: ANNEX 3 Page 14 Project No. of Villages in Different Stages Total No. of Year I II III Vllaites 2 5 10 5 20 3 5 25 15 45 4 5 35 30 70 5 5 45 45 95 It is recognized that there will be variations in uptake by different farmers in any one village. The foregoing assumption implies this fact. Beans 42. Nothing specific has been established by field trials for improving bean yields in the Kigoma region. It is known that a late (April/May) planted sole crop of local yellow beans yields at least double that of an early (November/December) planting. This is due largely to reduced competition and improved seed setting. Project proposals include one basic recommendation for beans on block farms. This is for planting through maize, spacing at two feet on three-foot ridges, and two or three weedings, depending on whether or not the previous maize crop has received fertilizer with the latter being the factor differentiating between Stages II and III. 43. Details of the effects of these treatments are given in Annex 15 Table 2. They may be summarized on an acreage (ha) basis as follows: Level of Husbandry Beans Traditional Stage I Stage II Yield (lb/ac) 250 500 600 (kg/ha) (280) (560) (670) Gross Return (T Sh) 95 190 228 (per ha) (235) (470) (565) Mfaterial Cost (T Sh) 8 14 14 (per ha) (20) (35) (35) Labor Costs (mandays) 24 45 50 (per ha) (60) (111) (123) ULet Return Excluding Labor (T Sh) 87 176 214 (per ha) (215) (435) (551) Return to Labor (T Sh/manday) 3.62 3.92 4.28 Rate of uptake is assumed to be the same as for maize, since the crops would are grown in association. ANNEX 3 Page 15 Cotton 44. Recommendations proposed for improving cotton yields are based on experience in the Western Cotton Growing Area of Tanzania and in locations elsewhere having similar soil, climatic and pest problems. Farmers in Kigoma Region growing UK64 cotton without insectidal control, planted in January at suboptimal spacing, without fertilizer, improperly thinned and inadequately weeded obtain yields approximating 350 lb of seed cotton per acre (390 kg/ha), under the Traditional system. Under these conditions the proportion of first grade (AR) to second grade (BR) seed cotton approxi- mates 55:45. Mid-December planting at the recommended spacing of two feet on three-foot ridges, application of CAN and DAP each at 90 lb/ac, thinning to two plants per stand, two or three weedings, and insect control by six sprays at seven to ten day intervals is estimated to give yields approxi- mating 1,000 lb of seed cotton per acre (1,120 kg/ha), the recommended Advanced level of production. Under these conditions, because of control of insect pests, mainly boll worms, the proportlon of AR to BR grade of seed cotton is estimated at 80:20. Current control measures are based on use of DDT, Sevin and Dimethoate applied from a conventional knapsack sprayer. Project proposals are based on experience in Tanzania and else- where using Endosulfan applied through battery powered ultra low volume (ULV) sprayers. One such sprayer would be adequate for five acres of cotton, or two for each 10-family unit cell. 45. Details of the effects of these treatments are given in Annex 15, Table 3. They may be summarized on an acreage (ha) basis as follows: Level of Husbandry Seed Cotton Traditional Advanced Yield (lb/ac) 350 1,000 (kg/ha) (390) (1,120) Gross Return (T Sh) 177 602 (per ha) (437) (1,487) Meterial Cost (T Sh) 1 243 (per ha) (2) (600) Labor Costs (mandays) 62 83 (per ha) (153) (205) Net Return Excluding Labor (T Sh) 176 359 (per ha) (435) (887) Return to Labor (T Sh/manday) 2.84 4.33 46. For planning purposes the rate of uptake of the different stages of cotton cultivation has been estimated as follows: ANNEX 3 Page 16 '.oject No. of V_la _s~ in Differe, t S_:ages Total No. of Year Trad ,ciona 4dve -iced Villages 2 10 0 20 3 15 30 45 4 15 55 70 5 15 80 95 As for maize, it is recognized rh3t there will be var.acion in uptake bj different farmers in any one vl'!Age. Groundnuts 47. Groundnuts are presently a popular but comparatively minor crop in the Intermediate and Miombo agro-ecological zones. NLo trial results are available but conditions in the region are .u'fficiently atEmiar to those in other groundnut growing areas to permit realistlt assApt-ions for plan- ning purposes as to the effects of a simple package' reuommended practices. The traditional Muitunde variety, suitable only for e-, late ani frequently interplanted at suboptimal spacing without fertll=ter, is estLm.ated to yield 350 lb of shelled nuts per acre (390 kg/ha). Ite Bukene variety,.of confectionery quality has been introduced within the past five years. It is popular with farmers because of its higher yield porenti&l and larger kernels which earn a premium price. Planted as a so1e crop at a spacing approximating nine inches on three-foot ridges with a dressing of single superphosphate of 112 lb/acre (125 kg/ha), 't is aazuaied that this variety would yield 700 lb of kernels per acre (780 kg/hs;, i.e., double the yield from the assumed traditional practices. 48. Full details of the two types of groundnut production are giveni in Annex 15 Table 4. They are summarized on an acreage (ha) basis below: Level of HuabWdry Groundnuts Traditional Advance. Yield (lb/ac) 350 .0C (kg/ha) (390) (090) Gross Return (T Sh) 198 .1433 (per ha) (489) (1,0 69) Material Costs (T Sh) 26 93 (per ha) (64) (230) .r'or Inputs (mandays) 54 87 (per ha) (133) (215) Net Return Excluding Labor (T Sh) 172 340 (per ha) (425) (839) ;k-turn to Labor (T Sh/manday) 3.19 3.91 ANNEX 3 Page 17 49. It is not intended that improved practices for groundnuts would be introduced until Year Four of the project when a program of demonstrations would be mounted. Py this time several years of trials would have been conducted at the proposed trials wing of the regional Training and Trials Center. The results of these investigations may dictate a different form of recommendation; however, the relative orders of magnitude of benefits are likely to be larger rather than smaller than those assumed above. For planning purposes it is estimated that 11 of the longest established villages would adopt improved practices for groundnuts by project Year Five. Supply of Inputs 50. With the exception of supplies of improved seed, inputs would be provided on credit through each village cooperative via the regional cooperative union by the Tanzania Rural Development Bank (TRDB). Such credit would not be made available until the village had demonstrated its eligibilitv for credit by registration as a primary cooperative society. Further details on village and individual eligibility criteria, and the mechanics of input procurement and distribution, and credit administration are in Annex 14. 51. Cotton seed is currently distributed at no charge by TCA. It is assumed that this arrangement will continue, with the costs of the operation reflected in producer price. 52. Improved maize seed, currently Ukiriguru Composite A, has been provided free by the Regional Administration using the Regional Development Fund to purchase foundation stocks from State farm in other regions that were subsequently bulked on block farms in newly established ujamaa vl"¶ages. This procedure would be modified under the project such that the prop(: Training and Trials Center would multiply breeders' seed to provide nucleus stocks for further multiplication on isolation blocks in ujamaa villages. Seed would be changed every three years so that a 350 family village would need to set aside a six-acre (2.45 ha) multiplication block for this purpose. Assuming a seed yield of 1,500 lb/ac (1,675 kg/ha) this area would supply sufficient seed to plant the village block farm (350 acres). 53. Improved groundnut seed would similarly be produced in the village frorm nucleus material supplied by the Training and Trials Center. It has been assumed that farmers would grow an average of 0.5 acres of groundnuts. Thus, assuming a seed rate of 70 lb/ac and replacement every three years, a 350-family village would require some 4,100 lb of seed per year. With improved groundnuts yielding 700 lb of seed per acre, some six acres of groundnuts would need to be planted for seed each year. 54. Seed multiplication at the village level would be organized by the Field Assistant in consultation with the village development council. The past practice of seed production on a communal farm should be resisted because of the difficulty of obtaining labor when needed. Rather, the village should pay labor to work the seed farms--which would need to be isolated from other crops because of the risk of outcrossing--exchanging the seed with farmers for an equivalent quantity of commercial grain or kernels, and utilizing this material to finance the operation. ANNEX 3 Page 18 LIVESTOCK DEVELOPMENT AND TSETSE CONTROL 55. Animal production in Kigoma region is severely restricted by tick-borne diseases, trypanosomiasis and internal parasites such as liver fluke. Most of the livestock population is concentrated above 1,500 m in the Highland Agro-ecological Zone above the wooded habitat of the tsetse flies (Glossina morsitans) which trasmit trypanosomiasis. About two-thirds of the cattle population is located in the highland area of Rasulu District. Livestock census results for 1955, 1961 and 1971 show an apparent increase in cattle numbers at the expense of sheep and goats. 1973 census figures are in course of preparation. Year Cattle Sheep Goats 1955 55,800 30,500 106,900 1961 65,100 39,200 151,000 1971 78,000 25,200 108,500 56. Cattle are almost entirely Ankole, well suited to extensive management but of poor carcass quality, slow maturing and with low repro- duction rates. Herd size is small at two to five head, and cattle are kept mainly as prestige symbols. There is no tradition of using cattle for draft purposes and the numbers and beef quality of local animals are such that substantial numbers of Zebu slaughter stock are imported annually, as indicated in the following table showing recorded number of animals slaughtered in Kigoma Region. Year Sheep Goats Cattle Total Imported Local Local % 1969 792 3,732 5,262 2,132 3,130 60 1970 733 3,342 5,708 2,468 3,240 57 1971 948 4,461 5,938 2,559 3,379 57 1972 1,157 5,769 7,664 2,527 5,137 67 These data are collected during meat inspection and are almost certainly low estimates of actual slaughter, particularly of sheep and goats. Never- theless, they still imply an extremely low off-take, below 5% in 1971, which is indicative of low calving and high calf mortality rates. The local Ankoles are probably slaughtered around six years of age at a weight of 200-250 kg, fetching T Sh 250-300 on the average in 1973. 57. The regional administration operates a regional network of dips to control ticks, and hence tick-borne diseases. Field staff with veterinary training provide a limited diagnostic and treatment service for other diseases in the three districts of the region. Dip establishment is on the basis of a cattle population of 500 within an 8-km radius. Below this level tick control is effected by stirrup pump sprays. In 1973, Kilimo operated 32 such dips and proposed construction and staffing of a further 20 in the 1973/74 financial year. ANNEX 3 Page 19 Tsetse Control 58. Despite the importance of tick-borne diseases, livestock develop- ment is most severely constrained by trypanosomiasis. As indicated in Map 10942 the presence of tsetse flies, and hence the disease, effectively contains the cattle population within less than 20% of the region. Chemotherapy is generally considered to be uneconomic for livestock. The effect of ujamaa- ization and the move from more densely populated, and hence cleared, areas into the Intermediate and Miombo zones will be to extend the fly-free area, since G. morsitans cannot breed and survive without bush cover. This process will take some years in each village, i.e., until the farming area has been totally cleared. 59. Because of the overall benefits of tsetse clearing on livestock and human health, the ongoing, satisfactory system of payments to villagers fer. this purpose would be continued under the project. Clearing costs vary according to the size and density of the tree and bush population. Kil imo experience indicates typical clearing costs approximating 40-50 mandavs/ha in the Intermediate zone, rising to 50-60 mandays/ha in the Miombo zone. Payment at the full minimum daily rate would be prohibitively expensive and unwarranted, since farmers should not be paid to clear their own plots. On the other hand during the first period of establishment of an uiamaa village, farmers are unlikely to make even a small cash income since most are moving to a new locality, opening new farms and building a house. Payment of a token amount for clearing one or two acres per familv in the first year provides cash for basic necessities, such as salt, and ensures an area of farmland for subsequent cropping. In 1973 such payrl., ranged between T Sh 5.00 and T Sh 15.00 per acre depending on the density of bush cover. These levels would be continued under the project. 60. Assessment of the rate, and of the need for tsetse clearing would be part of the Village Site Feasibility Report (VSFR) referred to earlier. Tsetse control staff of Kilimo would carry out a survey in the village area to assess the tsetse fly population and the amount of clearing involved. The field survey and report by an AFO trained in tsetse control assisted by two FA's, would take approximately 14 days for each village area. Such men are already in post in each district and are supervised by the Regional Tsetse Control Officer. Pilot Livestock Project 61. Present tsetse-free areas are inadequate to support current cattle and human population, hence the erosion and overgrazing typical of the Highland zone. New ujamaa villages created in the Intermediate and Miombo Zones will have a greater chance of siphoning off surplus population from the highlands if they can hold out cattle-keeping possibilities to livestock owuners. As indicated earlier, this can only occur following clearing, and the creation of areas free of tsetse fly. Additional attractions would be the provision of veterinary services and opportunities to upgrade local ANNEX 3 Page 20 stock. Apart from the obvious benefit of reducing the highlands cattle nopulation, such a move would also permit the introduction of cattle into lowland farming systems. The benefits of mixed crop and livestock farming would be considerable from such standpoints as land utilization, soil fertility maintenance, and human nutrition. 62. It will be some years before cattle keeping will be practicable in many ujamaa villages, although action needs to be initiated in the near future to investigate stock and land management practice appropriate to the region with reference to both highland and lowland zones. Such investigations should also be related to attempts to upgrade the local Ankole cattle by crossbreeding with improved types such as the Boran. 63. The project would finance the initiation of a pilot project along these lines that would be located together with the proposed Training and Trials Center. A small herd of 30 improved Boran heifers with two bulls would be established at the Center. Its purpose would be twofold: first to investigate and demonstrate livestock management systems incorporating utilization of crop residues and fodder crops, and second, in the longer term, to produce breeding stock to be used for upgrading local Ankole cattle in ujamaa villages that had adequate grazing land and had undertaken to participate in livestock health and improvement programs. Implicit in these objectives is the need to demonstrate the benefits of a commercial orienta- tion to cattle keeping in the region. Although the primary objective would be integration of cattle keeping and cropping to promote beef production, the possibility of training draft animals would also be investigated. It is unlikely that these activities would make much of an impact on ujamaa villages during the life of the project; however, the results obtained would provide the information necessary for preparation of subsequent investment proposals. AN x7?:j Tabu, TANZANIA KIGOMA RURAL DEVELOPMENT PROJECT Mean Monthly Rainfall Records to December 1970 Station Kigoma Kasulu Elevation above Mean Sea Level (m) 885 1,380 Recording Period (years) 32 39 Rainfall (mm) January 127.7 165.1 February 120.4 158.3 March 151.4 178.8 April 151.5 197.8 May 52.3 53.3 June 5.6 3.7 July 1.9 0.5 August 2.4 2.1 September 14.7 21.1 October 53.8 66.5 November 131.5 135.3 December 151.9 '76.2 Mean Annual Total 965.1 1,158.7 Source: East African Meteorological Department. ANNEX 3 Table 2 TANZANIA KIIQ4A RURAL DEV`LOMIENT PlDJ3CT Estimated Labor Needs and Availability Per 3.5 Acre Farm (madsYs)A' Crop Acreage Jan. Feb. Mar. Apr. May june July Aup. §Ot. Nov. Moc. Total Houseplot 1 8 6 7 7 5 - - - 5 8 8 6 60 Maisez/ 1 8 6 - 5 5 2 - - 10 10 14 6 66 Beans/ 1 _ - 10 5 5 6 6 - - 10 8 - 50 Cotton 1 8 7 6 2 6 10 4 10 8 8 - 12 81 rounuts 0.5 9 8 7 - 10 15 16 _ - - 10 10 87 Monthly Total 33 27 30 19 31 33 28 10 23 36 40 34 34s Labor AvailablaV/ 40 40 40 40 40 40 40 40 40 40 40 40 480 Net Surplus 7 13 10 21 9 7 12 30 17 4 NIL 6 136 1/ Aesuea adoption of noat labor-intensive techniques, i.e. at full development, and a mean six hours workday. In fact, the average workday would probably be divided between the houseplot and the block farm. t/ Maize and beans are grown on the same land, i.e. beans are planted in maize fields about one month before maize harvest. / sumes two adult equivalents per family of five, working an average of 20 days per month. Sources Annex15 Tables 1 through 4 and mission estimates. Note: Despite national adoption of the metric system, actual usage in the region is still with acres. Accordingly, the calculationa in this table are based on acres. ANNEX ' Table 3 TANZANIA lIGGfA RURAL DEVELtOPMENT PROJECT Herd Projection for Boran Nucleus Herd PROJECT YEAR Herd Age .5niFpOritt~On in years 1 2 3 4 5 6 7 8 9 10 11 12 on-ard 3+ 25 23 29 33 33 33 33 33 33 13 Hei fer. 2-3 27 10 9 7 7 7 7 7 7 7 Subtot.al Breeding Females 27 25 33 38 40 40 40 40 40 40 40 llel ters 1-2 27 11 9 12 12 14 14 14 14 14 14 Hei,ers 0-1 11 9 12 12 14 14 14 14 14 14 14 Ills in S,rvtce 2+ 2 2 2 2 2 2 2 2 2 2 2 2 R;"I s 1-2 11 8 12 11 14 14 14 14 14 14 1I'lls 0-1 11 8 12 11 14 14 14 14 14 14 14 Tot.l No-sbers 29 51 66 76 87 93 98 98 98 98 98 98 e.a- of Stock C Iled Bl lIs 1 1 1 Ci.led Cows 3+ 1 1 3 4 6 6 6 6 6 6 Pl I ed Hei fers 2-3 1 1 1 1 1 1 1 1 (H ,lled Heifers 1-2 In Calf Heifers 2-3 4 4 6 6 6 6 b Rreeding RlOls 2+ 11 8 12 11 14 14 14 14 14 T-t.al Sales 1 14 11 21 22 28 27 27 28 27 In Palt Heifers 2-3 30 Breeding Bulls 2+ 2 1 1 1 Total Porch,ses 37 1 1 1 Btrths {lelfer Calves 12 10 13 13 15 15 15 15 15 15 15 15 Boll Calves 12 9 13 12 15 15 15 15 15 15 15 15 Total Bi rths 24 19 26 25 30 30 30 30 30 30 30 30 Deaths Co-s 3+ 1 1 1 1 1 1 1 1 1 1 HrL IrS 2-3 3 Hei fers 1-2 Heifers 0-1 1 1 1 1 1 1 1 1 1 1 I Bulls in Service 2+ B-Ils 1-2 Bulls 0-1 1 1 1 1 I I 1 1 1 1 1 Total Deaths 3 2 3 3 3 3 3 3 3 3 3 3 Technical Coefficients Ca>ns Rare (71 ~~~ ~~~ ~~~ ~~~~~90 75 75 75 75 75 75 75 75 75 75 Ntortah ty Rate 17.) Cows 3+ 3 3 3 3 3 3 3 3 3 3 Heifers 2-3 io./ 3 3 3 3 03 3 3 3 3 Heifers 1-2 3 3 3 3 3 3 3 3 3 3 Heifers 0-1 10 10 10 10 10 10 10 10 10 10 Bulls In service 2+ Bulls1 1-2 3 3 3 3 3 3 3 3 3 3 Boils6 0-1 10 10 10 10 10 10 10 10 10 10 Culling Rate (I) CH es 3+ 5 5 10 15 20 20 20 20 20 20 Heifers 2-3 5 5 5 5 5 5 5 5 5 H/eifers 1-2 5 5 5 5 5 5 5 5 5 -1/ Incl.des loss in transit together with normal mortality. ANNEX 4 Page 1 TANZANIA KIGOMA RURAL DEVELOPMENT PROJECT GOVERNMENT AND ADMINISTRATION IN THE KIGOMA REGION Regional Government After Decentralization 1. With effect from March 1972, Regional and District administations in Tanzania were reorganized and strengthened to promote rural development programrs more effectively. Chart 8466 shows the main features of the new organogram. The chief political officer in the region, the Regional Com- missioner (appointed by the President), has been given cabinet rank. The Regional Development Director (RDD), as head of the regional civil service, is now equivalent to the Principal Secretary of a central government minis- try. The most important effect of the changes was to make all civil servants working in a region in principle responsible to the RDD of that region and not, as before, to central ministries in Dar es Salaam. 2. The new District authorities, helped by the Regional Secretariat, identify and prepare projects geared to the needs and potential of their areas and supervise their implementation. Programs and projects are re- viewed and collated in the region and subsequently submitted in the form of. budget proposals accompanying a regional plan to the Prime Minister's Office (which Office has responsibility for the overall planning of rural development). Following review and clearance at this level, regional pro- posals become a rural development program for the nation and eventually, with proposals from the central ministries and financing implications pre- pared by Treasury, part of the national plan. Following clearance of the plan and budget by the Economic Committee of the Cabinet, final authority resides in the National Assembly, which body approves separate regional as well as national budget estimates. 3. As indicated in the organogram, and detailed in the Organizational Guide 1/, the District Development Director is responsible for obtaining the participation and commitment of local communities and for ensuring that gov- ernnment staff work closely with the people in preparing, and executing the programs. He is assisted by three staff officers (for planning, finance and personnel) and by eight to ten district functional managers (functions or sectors as indicated in Chart 8466) in charge of technical and specialist services at district level. '7 'The New Organization: A Management Guide." Government of Tailania, January 1972. ANN tX 4 egionol i s rc,is' to s the Ita n,)rŽT.en dti.trict prepara- . ivn and implc eniat _ _-n and generat g-ov- >niaent objecti ves and policies. The .-egional Developmerdt OiroeAtox tlhus Ja-i.ntains a flow of plans a-;id record of achievement frour 0rh- districcts to zh cernter, while aztempting to ensure an appropriate flow of technical ass1stance and resotirces to the districts. Regional organization closely miriors organization at district level but in general, regional staff are nfic .:3erior and experienced3 than their district counterparts, Staf.t Officers assist lbv RDI) in the tasks of coordination and control buit are also expected to assist district officers with profes- ..ional advice ;i wA .bn e to the Pr,i) ard to district counterparts, though, in principle, operav !a` respofl:> n'iity Is veste(d in ;,F district level faanagement. (An exception is the Regional Water Engineer who takes personal responsibility for the implementatiors of all water programs in the region.) Regional functional managers also liaise with the central departments to 2nsure that Districts get the resourues needed to carry out their programs. in sunmary, with * ontrol exercised over the whole range of technical gov- *ernment services as these operate a, local level, the tuse of integrated or comprehensive approacnes, =ore in Keeping, with riral development objectives, is facilitated. The coTnbination} ot authority, responsit4ility and account- ability focused at the District and Regi.onal level should lead to much more active promotional efforLs than it; the past, particiolarly in the more back- ~-ard and isolated rfgioals that were so easily neglected under the old system. 'c-Xp2rience ill thi tll"- < Administrar-ve structtire. ii 'K.igoma region follow those described in the proceeding section; curretltily, ,ere are approximately 120 higher level 1/ staff distributed over the Regional office (sited in Kigoma-Ujiji tcwnship) and the three cistrict offices of K(igoima, Kasulu, and Kibondo. IlRese are consideiably blelow establishr-at staffing leve's (which allow for ab.ut 200 posts), And at levels just belew District Finetional Manager -- ., among senior and middlc level profess.`onal grades, only about one- ihalf of the 60 established posts a e filled. As well a3 s& , equireme:ts o, the project sn respect of ada(itlonal 'tnctional speciali._is, discussed elsewhere in the Report, some strengthening rn the accountting and pla-aning Set tio-,s at District level is particularly nrgent. 1n Kasulu and Kibondo, ttie Planning Officer has no supporting staff In contrast to an establishment of s.:- posts, while the District Finance ficers are in ani even worse position, with zero staff in contrast to a ;Xo t establishment of about 20. In addition to staff ae with most of the new regional staff act familiar with con,ditions it thc region, somux rime must elapse before re- -Jonai planning and pioject prepa;atrAon t-a[i become fully effective. Moreover, of necessity, relatively inexpertencer! or newly promoted staff now occupy U Higher Execut.ve :.zkote _j!7cer ;iA. 3hbove, inclutiing junior level -rofessitvnal 3 taff. ANNEX 4 Page 3 many of the posts that carry major responsibility for plan and project pre- paration. Often the background of previous experience does not chime par- ticularlv well with the predominantly rural setting in Kigoma. In general, regional headquarters staff, having greater experience and seniority, are called upon to take a more directly executive and operational role than the job descriptions, of para 4 might suggest. The differences are sharpest at district ltevel, between old style civil service administration and the pro- motional-and management orientation called for under the new system, and many disttict officer wvill at first need consi4tible help in coping with their more demanding new jobs;. S., Finally, the launching of Operation Kigoma, following hard after tlie,initial decentralization and reorganization of government in the region, resulted in a very considerable additional strain on regional and district staff. In these special circumstances, it is no criticism of the administra- t on that proper financial control and budgeting are hardly yet firmly es- tablished nor that planning, either at district or regional level, is still in its infancy. Difficulties are also occasioned by comparative remoteness from Dar es Salaam including tardy and uncertain delivery of vital spare parts and materials (especially for water supply systems, and water trans- port). These problems have resulted in significant delays in completing programs in respect of water supply, and school building and ujamaa devel- opment in general. Regional Finances 9. At present, the proportion of government resource for develo --nt that is channeled tlLrough the regions is still relatively small. For }9Xi/ 74, the regions are expected to account for expenditures totalling T Sh 170 million or 13% of tne capital resources available to government (excluding expenditures relating to the TanZam railway). Already, however, this is double the amount allocated in 1972/73, the first full year after decen- tralization, and the share is likely to grow as project preparation and management capabilities are further developed in the region. 1/ 10. For Kigoma region, development expenditures totalling some T Sn 7 million were authorized for 1972/73, including a substantial increase from the original (pre-decentralization) budget estimate of T Sh 4.3 million. Of the total, T Sh 1.5 million remained unspent at the end of the period, due partly to the difficulties with supplies and transport mentioned under para 8 above. In addition, the sharp acceleration in regional construction activity levels, called under Operation Kigoma and associated programs, was difficult to accomplish in the few months available after start up. 1/ 1:I respect of recurrent expenditures, the regional share is about one- fifth of government expenditures, and totalled about T Sh 460 million in 1972/73. ANNEX 4 Page 4 11. For the current fiscal year 1973/74, authorizations may exceed T Slh 15 million, an increase of more than 100% over expenditures in the previous vy..r. Part of the increase, about one quarter of the total, is to take account of unfinished work from 1972/73 and the effect of rising costs. Major budget headings are as shown in Table 1. Table 1: PROPOSED DEVELOPMENT EXPENDITURES 1973/74: KIGOMA REGION T Sh Million Notes Operation Kigoma expenses 3.0 Allocated by PMO quarterly Regional Development Fund 0.9 Allocated 6 monthly, for small projects Rural Water Supplies 4.3 Includes a request for 2.0 million to complete the 1972/73 program Primarv School Facilities 2.1 Includes a request for 0.9 million for cost increases Agriculture and Forestry 0.8 Rural Health Facilities 0.5 Includes 0.1 million supplementary request Others, including Urban Programs 3.8 TOTAL 15.4 Sources: Regional authorities and published estimates. 12. For the recurrent budget, the approved estimates for 1973/74 total T Sh 26.6 million up from 20.7 million in 1972/73. Of the total, pri- mary education, with a budget of T Sh 5.6 million, is the largest item, fol- lowed by regional administration 4.6 million, extension and other agricul- tural services 2.6 million and road maintenance 1.9 million. A noteworthy feature is the substantial increase in allocations for maintenance, repair and renewal of equipment and structures; including the road program, in- creasing from T Sh 2.3 million to T Sh 4.2 million. 13. Up to the present time, no powers have been granted to regional authorities in respect of taxation, nor local discretion or deviation in the application of national policies regarding regulatory practices, licensing, marketing of produce or pricing. All funds continue to be channeled to the Treasury in Dar es Salaam (some very minor sources for local revenue have in ANNEX 4 Page 5 fact been abolished). There is thus a somewhat lopsided ability to generate programs of expenditure on the part of regional government without the stimulus to economy in resource use likely to result from more reliance on local financing. For the longer term, as regional programs of expenditure, recurrent and capital, become more important, a degree of fiscal self-reli- ance at regional level should be considered. As discussed elsehwere (Annex 16), the introduction of user charges at village level fo: services could be one of the more promising avenues leading in this direction. The Budgetary Process 14. The philosophy behind the decentralization scheme has been that more effective rural development would be promoted if decisions taken at local level were to reflect more closely the views of local people. In turn, local views and support for government programs would be sought through the political process, working through local level development committees. Thus, much more than in the past, the local level political structure, form the ten family cell basic TANU Unit through Ward level, District and ul- timately Regional Commissioner, interacts with the civil service in program identification, preparation and subsequent implementation. 15. The process relating to the preparation of the regional development budget can be said to begin in the districts with meetings at Ward level of Development Committees, made up from village leaders of the basic TANU ten- family cell units, plus civil servants at Ward level -- agricultural tech- nicians and rural development assistants, and chaired by the Ward Exec"-ive Officer. This group operates mainly at the very early stages of progr i. identification, for example, in regard to possible sites for new ujamaa villages (see Annex 13). Their proposals, together with others from district and regional staff are considered by a District Development Planning Committee with the District Area Commissioner as Chairman and district functional managers together with a selection of Ward-level elected councilors as m2mbers. Subsequent draft plans are considered by a District Development Council and District Executive Committee of TANU before transmission to regional level bodies. 16. Based on the three district plans, a similar review process culminates in the transmiss_on of a draft regional plan and accompanying budget through the Regional Executive Committee of TANU to the Prime Minister's office, in principle in March, i.e., four months prior to the commencement of the financial year in question. Subsequently, discussions with the Prime Minister's office take place, prior to finalizing the program. In practice, this procedure has not been fully established until this current year, due to the newness of the institutions and procedures. 17. Once approved, responsibility for carrying out the programs rests with the Regional Development Director, supported by the District Development Directors, in respect of the district programs. The district programs are not, however, separately identified in budget estimates, except for certain ANNEX 4 Page 6 items financed through foreign aid appropriations (e.g., rural health centers and rural water supply systems). 18. Accounts are the responsibility of the Regional Finance Officer, supported by district finance officers and the accounting sections of the functional managers. The system draws heavily on long-established procedures, centered around "votes" or "sub-votes" under which expenditures on specified objects to the defined limits are authorized. "Vote books" are kept in which expenditures falling under that particular authority to spend are recorded and spending agencies are issued with a "Warrant of Fund" as authorization instrument, detailing the votes or sub-votes and amounts made available there- under. Funds may be issued for expenditures over periods of varying length witnin t'1e accounting period; warrants are usually issued on a monthly basis from tne office of the Regional Finance Officer. Transfers of funds within sub-votes may be authorized at regional level; transfers between sub-votes require treasury approval. In principle, provided expenditures are conscientiously recorded under the relevant item and checked in sufficient detail, it would be pos- sible to identify expenditures down to the level of particular activities in identified villages using these procedures. But to use the system for this purpose, as would be required under the project, (and as in principle is already required by government), a considerable strengthening of accountancy staff at district level would be necessary. 20. Not part of the accounting process, but related to it, are the Im- plementation Reports, prepared by the regional planning secretariat on a quart- erly basis detailing progress through the year in meeting the objectives of the regional plan and expenditures authorised under the budget. The report for the final quarter of the year includes the Annual Implementation Report, which is close to a post mortem on the year's activities, including details of special difficulties and reasons for under or over-spending in relation to the original and supplementary estimates. ANNEX 5 Page 1 TANZANIA KIGOMA RURAL DEVELOPMENT PROJECT Education National Education Policies The national education goals of the government are to: (a) Expand a revised scheme of a seven-year primary education (including academic and practical subjects) to 100% of the school-age population by 1989 (presently 37%; 40% in Standard I); (b) Eradicate illiteracy through a massive campaign to be ended by 1975 and then provide facilities for a continuing adult education scheme, partly aided by part-time volunteer teachers; (c) Establish in each ujamaa village a Community Education Center consisting of school premises, teacher's houses, workshops (domestic science, carpentry, masonry, smithery and a school farm), rural dispensary, day care center for children, and assembly hall. 2. Achieving these goals will be difficult and expensive to the na- tional treasury. Education centers have been provided under several Bank- financed projects, the least expensive of which would have an investment cost of about T Sh230,000 (US$34,000) or about US$100 per pupil with recur- rent costs of about T Sh 60,000 per year (US$25 per pupil). The goal of uniiversal primary education would require annual capital outlays of T Sh 115 million (US$17 million) compared to the present annual development budget for pricmarv education of about T Sh 15 million (about US$2 million). This fiscal burden may require reducing the rate of expansion of primary educa- tion or developing innovative and less expensive ways of disseminating basic education among the different age groups of the population. Additional thioughlt may well be warranted as to whether the traditional formalized pri- mary education is the best and most cost-effective way to help rural groups with thie problems they are likely to face. Iducation in Kigoma Region 3. Primary Education. About 32% of school age population (35,260 -,s about 35% of whom were females) attended primary schools in 1972, an increase in absolute numbers of about 50% in 5 years. The 861 primary teachiers provided a teacher-pupil ratio of 1:45 which is approximately in line ANNEX 5 Page 2 with the national average. Mlost teachers are from regions other than Kigoma and have some knowledge of farming but little teacher training in craft skills. Primary teacher's salaries start at T Sh 320 per month, somewhat above the minimum wage and some 5 to 10 times the income of families whose children they are teaching. -4. Existing primary education facilities in the region are generally below standard with teacher's houses and classrooms often in a poor state of repair with variances depending to some extent on the interest of the head teacher. In order to obtain better housing teachers try to be transferred to other regions as the school construction program has fallen behind because of slhortage of resources. The self-help contribution has been less than hoped for: furniture is scarce; there are desks for less than half the students and in some classrooms there are no desks; textbooks are insufficient and out of date with only a few having been received in the region during the last three years. There are no school libraries. In a few schools a room for practical training has been provided but without equipment or tools. Each school has a garden in which students work a few hours each day, using tools brought from home (when possible). School farms could be made more effective by adding equipment and teaching basic agricultural skills. 5. Adult Education. An extensive network of adult classes "under the trees" has been established in all villages. The main subjects taught are reading, writing, and political education, along with health and agriculture in some instances. The materials used are those prepared and printed in quantitv for the RIESCO/UNDP experimental programs. Although registration is hipn, attendance in these classes that assemble three times a week is irregular and in cases where a literacy test has been given (after 15-month courses) the percentage of success seems to be about 10X) 6z. The teaching staff is composed of certified school teachers and a number of volunteer teachers (who are paid T Sh 30 per month when funds are available from reserves) who have completed the Standard VII of primary schools. The volunteer teachers receive special training in seminars organized at the district rural training centers and/or through special courses organized in local sclhools. 7. Administration, Planning and Finance. Following decentralization, thie administration of Primary Education and Adult Education is carried out at the regional and district levels by an education officer, an inspector, acI adult education officer, and a supplies and statistics officer. S. Estimates of capital and recurrent expenditures are prepared annualLy by the district and regionai education officers and are included iLl thit total regional budget request. The total regional allocation for FY72/73 for education was T Sh 8 million (recurrent and capital) of which about T Sh 3,700,000 was for teachers' salaries. ANNEX 5 Page 3 9. Requests and allocations are established on an annual basis with- out benefit of long-term planning and with no provision for future allocations. 10. Priorities. In coping with rapidly increasing primary student enrollments the regional budget requests in recent years have emphasized the building of additional classrooms and teachers' houses. By maintaining this priority, an academically poor network of conventional schools deliver- ing a type of formal education not well suited to the needs of much of the rural population could result. Accordingly, regional authorities consider that future priorities will be oriented more to the practical needs of the rural population. The following priorities have been established: (a) Initiating short special courses at the rural training centers to train some teachers in practical work such as carpentry, farming, masonry, and smithery. Local skilled artisans could be utilized for some of the training. (b) Tools and workshops for practical training to be provided for schools having five or more classes (i.e., from Standard V). Workshops and tools could also be used in adult education. (c) Desks, textbooks and copy books. Most existing schools require a complete issue. (d) Teachers' houses. (e) Additional classrooms and furniture. 11. The project would provide funds for tools, textbooks, desks, shops, additions to rural training centers, teachers' houses and classrooms in accordance with the above priorities. Cost estimates together with calcula- tions of numbers of each of the above items required under the project are provided in Annex 11, Table 3. ANNEX 6 Page 1 TANZANIA KIGOHA RURAL DEVELOPMENT PROJECT RURAL WATER SUPPLY A. General Historv 1. In 1970 the Ministry of Water Development and Power (known as Maji 1/) was established in Tanzania to consolidate with in one ministry the functions of rural water supply development and irrigation, in addition to providing urban water supply and power. The Ministry's main activity is water develop- ment: the power sector is limited to planning and control with implementation being the responsibility of the Tanzanian Electrical Supply Company TANESCO). 2. In 1972 the administrative and implementation responsiblities of the ministries in Tanzania were regionally decentralized and direct control of the rtural and urban water supply programs passed to the appropriate re- gions. In the water sector today the principal tasks of the Ministry are: (a) Providing overall direction through planning and control, setting guidelines and standards; (h) Initiating, planning and executing national water projects, including those in major urban areas such as Dar es Salaam- (c) Advising and assisting the regions in all technical matters; (d) Procurement; and, (e) Manpower development. 3. The regional administration. under the Regional Development Director (RDD)) has full responsibilitv for the design, construction, operation and maintenance of all rural water supply projects. In the Kigoma region, the regional administration is also responsible for the operation and maintenance of the urban sYstems. The Regional Water Engineer, in his capacity as func- tional manager of water development, has direct control over the design, con- struction and operation of all water supply schemes. He deals with Maji 1/ In this annex Maji/K will be used for the Kigoma Regional Water Office. ANNEX 6 Page 2 headquarters through the office of the RDD except when he consults directly on technical matters. 4. Tanzania has embarked on an ambitious program of no more than twenty years to provide adequate and clean water for domestic use to all its people. Increasing emphasis is being placed on rural water supply systems. Of the 1973/74 budget allocated for urban and rural water supply construction, 43% was budgeted for rural water supplies. The promise of clean water is one of the key inducements for encouraging people to move into ujamaa villages to exploit the resources of the country collectively. Organization 5. Since 1965 the organization of Maji has been reviewed by four dif- ferent groups of consultants. Each study has resulted in reforms, keeping the Ministry in an almost constant state of change. The basic organization of Maji currently is as follows: IPrnipal Secre,tar; I ,--s earcn n Planniag &! - Opera1tons II Finance & TrUaini YProjec; Construction Maintenance I AdminIstration Div. XPrep. Div.| IDivision Division Division Including Procurement. 6. In 1973, Maji including regional and district offices had about 7,000 permanent employees, the majority working on construction and main- tenance. Of these, more than 220 were expatriates from 22 countries, who held senior positions at headquerters or in the regions as executive en- gineers, designing engineers, mechanical inspectors, and technical assist- ants (in planning, project preparation, construction, operation, and main- tenance). Some of the expatriates were provided through bilateral tech- nical assistance programs: the remainder are paid on a contract basis. Many of the expatriates have no counterparts working with them, so little advantage is taken of the possibility of providing Tanzanian technicians with in-service training. ANNEX 6 Page 3 7. Regionally, except for a research and training division, the Office of the Regional Water Engineer has the same basic organization as Maji head- quarters. Under the new system of decentralization, the regional Maji office is dependent on the Regional Commissioner's Office (i.e., the central coordinating body in the region), with planning, finance and personnel being controlled by the Regional Development Director. Funding 8. Capital for the water supply program comes from both national and external sources insofar as there has been external bilateral assistance by way of loans, grants and technical assistance (through provision of both advisers and regular staff). The exact distribution of development funds between the Ministry and the regions differs from year to year. Large scale projects are under the control of the Ministry, whereas, virtually all con- struction related to rural water supplies is carried out by the regions. This is illustrated-by the distribution of the water development budget for 1973/1974 below. 1/ Water Development Budget - 1973/1974 Distribution Ministry Regions Total of Funds TSh million % T Sh million % T Sh million X Rural supplies 3.1 5 5.9 95 63.0 100 Urban supplies 78.9 95 4.2 5 83.1 100 Proj. Prep. 20.4 86 3.3 14 23.7 100 Training 2.0 100 0 0 2.0 1030 Total 104.4 61 67.4 39 171.8 100 9. Revenue from all water supply systems collected from households witlh water connections goes to the national treasury. Funds for operation and maintenance of all systems, both urban and rural must come from the na- tional treasury through an annual budget for recurrent expenditures. For 1973/74 recurrent expenditures were budgeted at T Sh 77 million, 54% by the regions and 46% by the Ministry. As the projects in operation in rural areas increases the recurrent expenditures (including 60% for personnel costs) will rise. 1/ Estimates of Public Development Expenditures 1973/74. ANNEX 6 Page 4 B. Kigoma Region Organization and Staff 10. The Maji organization structure for the Kigoma region is similar to that of the Ministry in Dar es Salaam. Shown below is the structure in- cluding numbers of technical and non-technical staff for each division. MAJI/K 1 Kigom-! Rn,-Tional 'W1 tor -iner| 1 --i-i ,, r1. i01 i -intr ,O,2ancI |, !'K .p- 'rD-i., 'vio. tr; -- viOI -3 iL i ontnneaiiisrv ' ' D.iV iS -o ti ol t Ih -i J ! LxI?cr 3 uporvisor Office Supervisor 0 .i.-t. ci.nicail u 40 46 29 The regional staff totals 263, including personnel working on both urban (Kigoma, Kasulu and Kibondo) and rural systems and stationed at the regional depot in Kigoma and in the district offices in Kasulu and Kibondo. A con- siderably smaller depot is under construction in Kibondo while office and storage space for the Kasulu District is provided in the space used for the Kasulu town water supply system. ANNEX 6 Page 5 11, Except fo- 7umo' operators. construction crews with foremen and works inspectors, tne above staff all wore in the Kigoma Regional Depot. When the Kibond:l District Depot is completed, the operatiorn and maintenance staff for that district, including supportinig staff, equipment, and vehicles will be moved to the new depot, thus permitting cioser supervision and main- tenance for the Kibondo region. 12. Although the Regional Water Engineer is a Tanzaniaii, most of the techiLicians and supervisors are expatriates. Because only one of the ex- patriates is working with a Tanzanian counterpart, little advantage is being taken of in-service training possibilities. Res orisibilities of the Regional Office 13. The Maji/K office has overall responsibilitv for all phases of rural and urban water stipplies. For urban stupplies, this responsibility includes billing and collection for house-connected water services. opera- tjif and maintenance, and extensions of improvements to the systems serv- ing 1igoma-UjiJi, Kasulu and Kibondo, where the approximate populations are 29,000 5,000 and 5,000 respectively. 14. For the rural systems, the responsibility includes field surveys and studies for proposed villages, preparing designs (which are later re- viewed and approved by the Maji office in Dar es Salaam); ordering materials and supplies through Maji in Dar es Salaam; constructing new systems and extensions, and operations and maintenance of all systems in the Region. Physical Plant and Equipment 15. The regional offices, stroehouse and workshops, are housed in a modern depot in Kigoma. The offices are adequate for the current staff, but practically no provision has been made for recuired additional personnel if there is a major expansion of the present program. Space is available within the compound for additional offices and working areas, although pro- tected storage space for plastic pipe is needed. Activities 16, Surveys and Investigations. To date village sites have been selected without paving adequate attention to the probable source of water supplv. This has resulted in the construction of systems requiring two to three km of pipe from the intake to the village and in one system, an additional two km of pipe to reach the storage tank location. It has also resulted in systems pumping against a head of 100 m or more, requiring installation of expensive pumping equipment, longer and stronger pipes for the rising main nndl morn costly operation and maintenance. Some of these problems could ,ilve been avoided had the Maji office participated in village site selec- tion. (This involveme;C would be provided for under the project; Annex 13). ANNEX 6 Page 6 17. Once the site is selected and the decision made that a system will be installed, a survey is made by the design office of Maji/K, including a site plan with the necessary sections and details of the pipe, fittings and references to standard plans that have been prepared for intakes, pump houses and installations, storage tanks and public watering points. Lists of materials are compiled and a cost estimate is prepared. Requisitions are pre- pared for the pipe, fittings and other materials and forwarded to the Maji office in Dar es Salaam where the requisitions are filled from stock, from the local market, or ordered from outside when necessary. From three to nine months and sometimes a year is required to obtain all of the items required for one system. 18. Pumps and engines are not specified by the region but are selected at Dar es Salaam based on the characteristics indicated by the Maji/K office. To avoid the need for multiplicity of spare parts, the 2aji office in Dar es Salaam is trying to standardize the pumps and engines to be utilized. 19. The engineers are handicapped by lack of data on stream flows and on the area's groundwater possibilities, although water is fairly plentiful in the Kigoma region from streams, a few rivers and some lakes. NORAD of the Norwegian government recently signed an agreement with Tanzania to con- duct a meteorological survey of the Kigoma region including the gauging of flows in some of the larger rivers. A groundwater survey will not be done owing to the cost and the fact that the region's agriculture cannot utilize widespread groundwater irrigation at this time. 20. Design. The systems for Kigoma region are being designed based on experience and the guidance of a set of lecture notes, prepared at Ubungu training school. The lecture notes stress the design of pumped systems, pay- ing almost no attention to seeking the lowest cost (capital and recurrent) system. Neither the merits of a hand-pumped shallow well nor the possibili- ties of developing protected wells are mentioned and little emphasis is placed on quality of water furnished at public watering points. Although designers specify gravel and sand packing around pipes to provide filtering actioni at the system's intake, lack of supervision during construction sometimes results in a direct connection from the source to the system. 21. A few gravity systems are being designed, however, the majority of systems designed to date are pumped systems using river water collected with perforated pipe laid in a bed of gravel and sand near the river. The water is pumped from a well lined with concrete rings to an elevated storage tank made of concrete blocks tied together with horizontal reinforcing bars placed in grooves cast in the concrete blocks. In most cases, the pumps utilized are centrifugal pumps that are driven by diesel engines. 22. Estimated Costs. Both capital and recurrent costs at a particular site vary widely according to the type of system installed. Where adequate groundwater supplies exist, digging a shallow well provides by far the ANNEX 7 Page 1 TANZANIA KIGOMA RURAL DEVELOPMENT PROJECT PUBLIC HEALTH COMPONENT A. General Tanzanian Priorities and Problems 1. Although their precise effect on economic growth is difficult to measure, short life expectancy and a high incidence of illness in Tanzania adversely affect the productivity of the labor force. In Tanzania, 41Z of the population is affected by tuberculosis of the lung. Child mortality is high: one child in ten dies before reaching 5 years of age in rural areas, due to malnutrition, pneumonia or diarrhea. Infectious and parasitic diseases, includ- ing malaria, affect two thirds of the country's inhabitants. ?. In 1972, following a reorganization of the Ministry of Health, national priorities were switched toward preventive medicine and health services for rural areas. At the same time, the establishment of a Manpower Development and Training Division in the Ministry gave recognition to the cirtical import- ance of health education and training for medical staff in the short and medium term future. 3. The Government's strategy is to develop an economical rural based system organized around: (a) a Rural Health Center with 5 outlying dispensaries per 50,000 population, designed to cater for the entire health needs of the family except those which can be provided only in hospitals, and to organize preventive campaigns such as nutrition education, environmental sanitation, mzternal and child health services and immunization. The centers consist of simple buildings for handling out-patients, maternal and child welfare clinics, maternity units, and holding beds. Rural dispensaries, designed to serve a population of approximately 10,000, cater for the more simple medical needs of the community. Each dispensary consists of an office, a store and a treatment room and is intended to be manned by a Rural Medical Aid, who would be a grad- uate of a three-year medical course; and (b) rural hospitals at regional and district levels, equipped to provide major curative services. These facilities are complemented by health services at the national level, provided in consul- tation hospitals in Dar es Salaam, Moshi and Mwanza and in three government ~iospitals specializing in tuberculosis, mental disease and obstetrics respectively. 4. Tanzania has surpassed its targets on a nationwide basis for dispen- saries and for rural hospital beds (1 per 1,000 population), although these facilities are unevenly spread and below minimum targets in some districts. Hlowever, the number of Rural Health Centers increased during 1969-1972 to only 'O of the required number. Further development will be aimed at (a) achieving ANNEX 7 Page 2 the target for Rural Health Centers and (b) expanding rural facilities and dispensaries, as required for population growth and for equality of distribution. In 1972/73, 11 Rural Health Centers were built, bringing the total to 100. For 1973/74, the target is an additional 29 centers plus 100 more rural dispen- saries (of which there are currently over 1500 in operation). 5. Tanzania can meet all its targets for rural health facilities within the next ten years, without having to increase the recent average rate of annual investment in this area. External aid will probably provide a substantial por- tion of investment requirements (up to 50%), as in the past. To finance the additional recurrent costs of operating the system, the Goverment has recently decided to increase its recurrent health budget by at least 10% p. a. (compared with 5.7% p. a., 1970/71-1972/73). This would raise the share of health in the total Government recurrent budget from 8Z in 1972/73 to approximately 9.5% by 1980. 6. The strategy calls for medical auxiliaries to operate the country's health care facilities at all levels, under the overall supervision of physi- cians. Key auxiliary personnel are Medical Assistants (MAs) and Rural Medical Aides (RMAs) who are to operate, respectively, the rural health centers and the dispensaries. Together with Assistant Medical Officers, 1/ MAs also staff the hospitals substituting to a large extent for physicians. 7. The supply of auxiliary personnel has not been adequate for the proper staffing of Tanzania's health care facilities, despite a considerable expansion in recent years. In 1972 there were only 578 trained Rural Medical Aids and 335 Medical Assistants for the whole country. But the annual output of RMAs and }As increased from 60 in 1965 to 150 in 1971 with the establishment of three new schools and it is intended that the combined yearly output would increase to a level of 480 per annum by 1978. Current shortages of Medical Officers are even more critical and in consequence the health care given by the auxiliaries in the rural areas has suffered from a lack of professional guidance. Of about 495 doctors now working in the country, less than 200 are citizens. Fifteen medical graduates from the Faculty of Medicine in Dar es Salaam qualified in 1972. With the goal of 700 citizen doctors by 1980, it has been decided to have an annual intake of 48 Tanzanians at the Faculty of Medicine in Dar es Salaam and 16 at other colleges within and outside of East Africa. Funding 8. The development budget for 1973/74 was approved in an amount of T Sh 58 mIllion (US$8.5 million), of which T Sh 23 million, or 40% of the total is allocated for regional expenditures (mainly for health centers and dispensaries). Total recurrent expenses increased from T Sh 152 million in 1972/73 to an estimated T Sh 206 million for 1973/74. (This is equivalent to a little over US$2.00 per person). Regional health services account for T Sh 119 million, or 58% of this. 1/ An assistant Medical Officer is an upgraded Medical Assistant. ANNEX 7 Page 3 B. Kigoma Region Existing Facilities 9. On a population/hospital bed basis, Kigoma is well above the national standard of 1,000 per bed with an average of 730 per bed in the region as a whole. There are now four rural health centers operating in the region, with one addition scheduled for construction in the current fiscal year. In relation to rural population, there will still be over 100,000 people per health center, at the beginning of the disbursement period under the proposed project, more than twice as many as the targetting 1 to 50,000 national standard. 1/ 10. There are 61 dispensaries currently operating at village level, (mostly in ujamaa villages), with an additonal 12 under construction from the 1972/73 program. Including the 12 under construction, rural population per dispensary will be 7,200 by mid 1974, ranging from 8,100 in the Kasulu district to 5,800 in Kigoma (rural) district. These ratios are well within the original national target of one dispensary per 10,000 of rural population, and within the 1974 national average one one per 8,800 of rural population. However, it has become clear that one dispensary per five villages or 10,000 people is not, in practice, enough to provide coverage for more scattered and remote elements of the rural population. Currently, additional facilities are allocated on a weighting system taking into account the ratios of population to dispensary at District level, the porportion or rural population residing in ujamaa village and share of population more than 10 km from any medical facility. For 1973/74, however, this weighting scheme did not give any additional dispensaries to Kigoma Region (100 new dispensaries were allocated). 11. This is currently a critical problem in the region, particulai,, with the considerable expansion over the past two years in physical facilities. Because of the shortage of doctors, the rural medical centers are operated largely by Medical Assistants and Rural Medical Aides, supported by a scattering of trained nurses, midwives and health auxiliaries. 12. The daily load on the health center staff is demonstrated by the number of outpatients attended by each staff member (excluding midwives) at three typical centers (assuming that each one handles the same number of out- patients during a six-day week): Bitale - 80 per staff member per day Mabama - 61 per staff member per day Kakonko - 60 per staff member per day In addition, the in-patients must be attended and, theoretically, dispensaries supervised; however, lack of transport, even if staff were available, makes this practically impossible. 1/ However, there are two important Voluntary Agency health facilities in Kasulu district which perform some functions of a rural health center. ANNEX 7 Page 4 13. Staffing of the dispensaries is even more precarious. In the 14 dispensaries operating in Kigoma District during 1972, there were a total of six rural medical aids, ten auxiliaries and nine midwives. The 25 medical personnel attended a total of 376,655 people in 1972, an average of almost 50 per staff member per day. In Kasulu district it was reported that there were only 2 Rural Medical Aides in the 18 dispensaries operating in 1972, and in Kibondo, three for eighteen dispensaries, the other being operated by mid- wives and health auxiliaries. The intent is for each dispensary to have at least an RMA, a midwife and a health auxiliary. 14. There are one-year hospital training courses for nursing assistants, village mid-wives and six month courses for health auxiliaries. (Essentially a course for imparting first aid service information.) 15. To increase the supply of Rural Medical Aids, a training school was started in Kigoma in April 1973, with an expected annual output of 30 graduates per annum from 1976 onwards. 16. Ad hoc seminars and refresher courses are conducted in the rural training centers into subjects such as the treatment and prevention of malaria, gastro-enteritis, dysentery, lep Dsy and TB. Administration/Supplies 17. The chief medical officers in the hospitals and health centers in Kigoma region are also the head administrators, resulting in a loss of valuable medical services and less than adequate administrative services. The regional hospital is supposed to supervise the district hospitals which in turn supervise the rural health centers and dispensaries and there is a need for trained administrative personnel at each level. Courses for training medical personnel in elements of administrative practice are needed. 18. Lack of transport also makes supervision and maintenance of the necessary links between the various levels of health service extremely difficult. The referral system is almost impossible to operate under current conditions. Furnishings and drug supplies in the medical centers and dispensaries are also scanty. Preventive Services, Environmental Sanitation and Health Education 19. While the importance of preventive services is now widely recognized, health workers are still heavily engaged in curative medicine. The budget for preventive services has increased about two and a half times for 1973/74 from the 1972/73 level, but the total is still well under 10% of expenditures on curative services (and only 6% if rural health centers and clinics are added to the curative side). 20. Some health education principles are taught during adult education courses in ujamaa villages. However, these efforts need to be increased in order to teach people the advantages and proper use of good health practices. ANNEX 7 Page 5 21. Internal parasites (e.g., hookworm), are an imnDortant health problem in the region. The problem can be attacked effectively and cheaply by install- ing a concrete slab in each household pit privy. A hand-dug pit of a meter or more in depth, capped with a suitable slab with squat-type hole, is effective in combating the parasite cycle. Public education through the schools and adult education classes is necessary to teach villagers of the need for and use of the new method. Teaching must begin with teaching the village teachers and health attendants. Making slabs available, instructing teachers and village leaders in their use, and persuading villagers to install and use them will take time. However, without such an involvement there is a danger that health benefits as might otherwise flow from the provision of controlled-access water supplies will not be realized. 22. In regard to water itself however, surveillance of water quality being distributed in towns and villages is inadequate. Samples are taken from the source when a new system is to be installed and are sent to Dar es Salaam for chemical analysis, but little bacteriological analysis is being carried out. No routine tests are carried out. The Ministry of Water Development and Power is considering establishing a water quality laboratory in each regional water office. It is suggested that in Kigoma this might be installed at the hospital in combination with the laboratory there, thus strengthening that laboratory and not installing a separate facility with separate equipment and personnel. Financing of Health Care 23. Following substantial capital outlays in 1972/73, the current program is modest, the major item being an additional rural health center, to cost T Sh 400,000. The effects of the 1972/73 capital program do, however, show up as a 51% increase in the recurrent budget, which increased from T Sh 3.3 mil- lion to T Sh 4.9 million. Details are shown in Table 1: Table 1: Recurrent Expenditures on health care Kigoma Region 1972/73 1973/74 ----T Sh million --- Medical Staff Salaries 1.38 2.09 Other Staff Salaries 0.45 0.71 Drugs and Hospital Supplies 0.77 0.86 Other Expenditures 0.66 1.28 '4. Recalling that the region is still well below authorized staffing levels in respect of medical positions, the costs of medical care are already onerous. The recurrent costs, under present scales for dispensaries and rural health centers, should be about T Sh 14,000 and T Sh 130,000 per annum. ANNEX 7 Page 6 Project elements for health care: 25. In general, Kigoma region is not below average for the nation in respect of the rural health facilities and it is anticipated that the major effort for the next few years should be towards improving the quality of service, through more adequate staffing and furnishings for the existing facilities. However, the population movements with the settlement of new villages would justify additional dispensary facilities particularly as some settlements on the miombo woodland frontier would be relatively remote from existing services. Accordingly, provision has been included for construction of 10 new village dispensaries, to be sited in ujamaa villages accepted under the project. 26. Kigoma region is still well below national targets in respect of rural health centers. It is anticipated that the number of additional staff to man these centers will gradually improve over the project period, so justifying some additonal construction. On current expectations, some 30 new centers will be built each year over the next several years in the nation and by the close of disbursements under the project it is anticipated that the ratio of rural population to health centers will have falled to 60,00n to 1. On the basis of maintaining a regional pace of development in line with development in the nation as a whole, it is proposed to finance an additional three rural health centers under the project, to serve the needs of the project villages. To allow time for training and adequate staffing of existing facilities, expendi- tures under this component are phased to take place mainly in the last two years of the project. ANNEX 8 Page 1 TANZANIA KIGOMA RURAL DEVELOPMENT PROJECT Fisheries - Kigoma Region Introduction 1. Tanzania is one of the largest producers of fish in Africa. Total fish production, although varying considerably from year to year, has greatly increased during the past decade. In 1970 the total fish catch was 188,600 tons live weight, of which 25,400 tons equivalent were exported, although in 1972 production declined to 157,000 tons. Production over the last decade has increased over three times, although some of this may be due to better reporting. Some 80-90% of the catch is taken in inland waters. Fish accounts nationally for perhaps a third of total animal protein intake, and consumption in 1970 was 13.1 kg per capita. 2. The fisheries industry is mostly carried out by artisanal fishermen operating with relatively primitive vessels and gear on a small scale and individual basis. The marine fisheries, carried out primarily around the islands of Zanzibar, Pemba and Mafia and in the inshore areas of mainland Tan- zania, employs some 5,350 fishermen and 2,000 vessels using a wide range of gear to fish shrimp, lobster, snapper, shark and many marine invertebrates. On Lake Victoria, over 12,000 fishermen employing almost 3,800 canoes are active in pursuing inshore fisheries for tilapia and catfish with beach seines and gill nets. On Lake Tanganyika, some 10,500 fishermen engage in fisheries of dagaa, small sardine-like species, and of its predators, primarily Nile perch. Some 7,000 canoes with light attraction techniques and "lusenga" nets, as well as beach seine and gill nets, are used. Fisheries Production in Kigoma Region 3. The fisheries of Kigoma region are carried out primarily on Lake Tanganyika. The lake has surface of 32,900 km2, of which 13,500 km2 or 41% belong to Tanzania, and an average depth of 700 m and maximum depth of 1,435 m. The total lake catch within Kigoma region has been growing in recent years and has been in the order of 45,000 to 50,000 tons in the last several years. Increases were due primarily to the growth of the fishing community together _.t- introduction of synthetic fibers in fishing gear and other improvements. aximum sustainable yield from the Kigoma waters of Lake Tanganyika is only FouT being studied but is estimated at double current production levels. The principal pelagic resources of the lake include four species of prey (all dagaa) and four of predator. The relative importance of dagaa varied from 43% of totn' catch in 1967 to 88% in 1970, 94% in 1971, and 68% in 1972. The fish resources of the lake and their environment are described in Appendix A. ANNEX 8 Page 2 4. The total number of fishermen along the Kigoma lakeshore, mostly located in some 35 villages, has increased from approximately 5,100 in 1970 to 10,500 in 1972, according to annual reports of the Fisheries Division. These fishermen operate about 7,000 dugout or planked canoes. About two-thirds of the fishermen catch dagaa mostly with a scoop net known as "lusenga", canoes and kerosene pressure lamps, but also, in recent years, with beach seine nets. The remaining one-third fish the predator species, principally with gill nets, but also with such other methods as line and hook and beach seine. As both dagaa and its predators are attracted by light, most fishing is carried out during the 22-23 relatively moonless nights, although bad weather reduces to about 18 the average nights of active fishing per month. 5. Natural conditions vary considerably along the lakeshore. While some villages are relatively well protected against the prevailing winds, most are exposed to the frequent rough weather and large waves. While parts of the lake's edge are smooth and gently sloping, others are steep and full of obstacles. For all villages, the natural setting determines the fishing methods and equip- ment to be used (e.g., scoop net, gill net, beach seine net, line and hook). Villages also vary according to their accessibility to roads, services, market outlets and extent of farming activity. In introducing new techniques the unique features of each viUage must be taken into account; this has not al- ways been done in the past. 6. The traditional fishing methods, as well as natural conditions, vary considerably: (a) The most common method --"lusenga" net, kerosene lamp and canoe-- involves the relatively small investment of about T Sh 1,000 and the participation of two fishermen. The average catch is approximately 5-7 tons per man per year. Notwithstanding the modest investment required, the equipment is often owned by middlemen who, according to some reports, exact as much as four-fifths of the catch. Although evidence is often con- flicting and unclear, it would appear that a reform of the ownership and sharing patterns might induce a favorable impact on fishermen's welfare. Ujamaa fishing villages, for example, divide their profits by thirds and allocate evenly for mis- cellaneous village expenses, a special bank account, and the individual fishermen; (b) Beach seine units consist of 3 boats, 3 kerosene lamps and 1 net costing approximately T Sh 8,600, and require about 20 men to haul in the net. Mechanized boats can be introduced, as they have been in the two ujamaa villages, to set the net, but problems of repair and maintenance must first be overcome. Beach seine units can be used for both night and daytime fishing, and smooth and shallow beaches, without obstructions, are most suitable; and ANNEX 8 Page 3 (c) Gill nets are designed for big fish and are used mostly in the river, swamps and delta area of the Malagarasi. Gill nets, which have been given by Government to six villages, are used almost exclusively by some villages while used in combination with dagaa fishing techniques in other villages. 7. Since 1969, 17 beach seine nets, 158 lamps, 3 mechanized boats, 32 canoes and 158 gill nets have been given to ujamaa villages by Government. Little if any information is available on the results obtained from this equip- ment. Moreover, it is unclear whether the recipient villages have been able to generate sufficient surplus to replace the equipment when it wears out. 8. In addition to traditional fishing practices, a small number of Med- iterranean-type purse seine units have been introduced to Kigoma waters in recent years. They are similar to those now used extensively elsewhere on Lake Tanganyika (Burundi has 16, Zambia 4, and Zaire 24 units). Each unit consists of one purse seine net 300 m long and 110 m deep, one mechanized boat 15-18 m long and powered by a 100-140 hp engine, one netboat and 5 lightboats. Two units are owned and one rented by Kigoma Uvuvi Ltd. (KUL), a company established in August 1972 and owned jointly by the Kigoma Development Corpora- tion, the Dar es Salaam Development Corporation and a private individual, and one unit is owned and operated by a Greek fisherman. The units belonging to KUL were purchased secondhand to T Sh 365,000 and T Sh 370,000 and the nets, also secondhand, were purchased for an additional T Sh 70,000 each. Each unit is manned by a crew of 30 and a skipper. Total annual operating costs are roughly estimated at T Sh 330,400 and annual yields at 500 tons per unit. Profits of these units rely heavily on catches of larger fish for the local market rather than on dagaa; because of the small size of the local market landed prices (and profits) drop sharply when large catches are delivered. Appendix B presents a comparison of the costs, yields and manpower requirements of the nurse seine and more traditional techniques. 9. Some fisheries within Kigoma region are being or could be undertaken outside Lake Tanganyika. Fishing takes place in the marshes and lagoons of the Nguruka area, where six fishing villages are located, and in some of the rivers, primarily the Malagarasi River. It is estimated that some 2,000-3,000 fishermen may be engaged, mostly on a part-time and subsistence basis, in the annual production of approximately 3,000 tons. In addition, some subsistence fish farming has been practiced and although techniques are still in an embry- onic stage and many problems (e.g., water supply and retention, selection .f soil and species, fertilizing and feeding methods) must still be resolved, long-term prospects may be promising; more study and development work is required. Fisheries Processing and MarketinR 10. Virtually all dagaa is sun-dried. Traditional drying methods, how- ever, are generally unsatisfactory. The fish is often stored in baskets for several hours before drying, thus adversely affecting the final quality; the ANNEX 8 Page 4 fish is placed directly on the sand for drying to which it adheres and which is difficult to remove; the climate of the humid rainy season induces spoilage before drying and rain at any time during the 24 hours necessary to complete drying destroys the fish's quality. Kigoma Uvuvi Ltd. have introduced some improvements in the processing of dagaa by using metal screens for drying and a closed building for weighing and storage. Other improved techniques merit testing under local conditions. Big fish, on the other hand, are now generally smoked in the villages before distribution. Possibilities of freezing and refrigerated railroad transport to consuming centers are currently under review. 11. No statistics are available on the quantity of dried dagaa produced each year. Estimates range between 6,000 and 12,000 tons. 1/ Some 650 tons of production was recorded for export to Zaire, Burundi and Zambia in 1972, although these countries recorded much higher corresponding import statistics, indicating a sizable illegal traffic. About 1,200 tons of dried dagaa were transported by rail from Kigoma to Dar es Salaan and other destinations. Of the 6,560 tons of big fish harvested in 1972, it is assumed that most was smoked and distributed throughout Tanzania. 12. Very little information is available on the marketing and distribution of fish within the region. It is not know, for example, what the current and potential markets for big fish and dagaa are or which geographic areas and income groups make up these markets. Likewise, many of the middleman's oper- ations are unclear, relating, for example, to his operating procedures, his average captial and profit margins, and his ownership pattern. It is known, however, that marketing and distribution are complex, involving, inter-alia, a diversity of regional customs and difficulties in transport. Fish are usually taken when caught to a landing station for sale to fish traders. Fish are sold by piece, baskets or piles. In villages infrequently visited by traders or far from consuming centers, fishermen themselves undertake the processing of fish, but generally the fish are processed by the traders. Sun-dried or smoked fish, because they are easier to handle, have a wide distribution, being carried by traders ovef long distances to consuming centers by head load, bicycle, bus, lorry, boat and train. Fresh fish marketed in Kigoma are kept under running water, without ice or cold storage. At the close of market, the unsold fish are smoked to prevent spoilage and to improve appearance for marketing the following day. Individual traders sell between 5 and 25 pieces of big fish (1-3 kg each) per day in the market place. 13. About 1,175 tons of dried dagaa have been transported elsewhere in Tanzania, according to reports of the Fisheries Division. Most presumably traveled by rail, for which the charges to Dar es Salaam are T Sh 29.10 per 100 kg (2 bags) and to Tabora T Sh 11.25. 1/ The ratio of wet to dry dagaa has been reported to be as little as 3:1 and as much as 6:1, while 4.5:1 is perhaps the best estimate for a properly dried product. ANNEX 8 Page 5 14. The price of fish is obtained by bargaining and differs widely with a variety of factors, including quality and quantity, supply and demand. For example, during the rainy season, when productivity of the lake is highest and conditions for drying unfavorable, the price of dagaa can be three times that of the dry season. Moreover, the price of fresh dagaa caught by the purse seine units and sold in Kigoma varies from T Sh 0.20 to 0.75 per kg depending primarily on the quantities landed. According to some sources, the producer price of dried dagaa ranges form T Sh 60 to T Sh 200 per 50 kg bag, or T Sh 1.20 to T Sh 4.00 per kg. Because fish prices are not at present controlled and because a portion of the fishermen's catch is often paid to the middleman for use of boat or equipment, the middleman may reap far greater profits than the fisherman. Moreover, the fisherman has no way to check whether he receives correct payment for his share of the quantity originally delivered. In Kigoma town in October 1973, the retail price of dried dagaa was the equivalent of about T Sh 2.75 per kg. Prices of dried dagaa have reportedly been generally rising over the last five years. In comparison, the producer price of big fish ranges from T Sh 2 to T Sh 3 per kg and up, and observes considerable variation among species. 15. The larger predator species such as Nile perch have a lucrative export market if properly caught, handled, frozen, packed, stored and exported. The local "opportunity cost" of such species is therefore high. Apart from a small market in Dar es Salaam it is doubtful that many persons in Tanzania could afford to pay the prices that these larger species would be worth (assum- ing required processing and export marketing procedures could eventually be developed). Therefore, such fish have little roles in improving the nutrition levels of Tanzania's predominantly rural population. That role must continue to be performed by the much cheaper and well known dried dagaa. Government Programs and Capabilities 16. Rapid and extensive development of the fisheries sector is a goal of the Government of Tanzania. Likewise, it is committed to the improved economic and social welfare of the fishing community and to uJamaa village development as the appropriate vehicle of such improvement. The Government has recently accelerated efforts to achieve these two objectives. First, it plans the development, probably through Kigoma Uvuvi Ltd., of an industrial fisheries project in Kigoma town, including additional purse seine units, a fish receiv- ing station, a boat building yard and refrigeration facilities. Provision in the 1973/74 budget has been made for a demonstration unit (T Sh 153,000), a fish receiving station (T Sh 500,000), and a boatyard (T Sh 600,000). Manage- ment of Kigoma Uvuvi Ltd. requires strengthening even for its present level ~:f operations. Management experienced in operating fishing enterprises, in conditions similar to Kigoma's is necessary. An incentive scheme for paying crews on fishing vessels should also be considered. Initial development should be modest pending results of the UNDP/FAO project's market, processing and catching studies and activities. Secondly, it plans to increase incomes of the lakeside fishermen through the introduction of improved techniques and equipment to ujamaa fishing villages. A 1973/74 budgetary allocation for a mobile service unit (T Sh 300,000) has been made as a beginning to such a program. ANNEX 8 Page 6 17. The conception and coordination of plans and policies for development of the fisheries in the Tanzanian waters of Lake Tanganyika warrant strength- ening. A development strategy directed at both the modern and industrial sectors embodies potential conflicts, particularly at points of production and marketing. Maximum production and fishermen's welfare are not necessarily consistent. In addition, planning seldom demonstrates the consumer orientation required. Mbreover, the different components and parties involved in a common development strategy require coordinating. 18. A UNDP/FAO Lake Tanganyika Fisheries Research and Development Project was recently approved to provide the basis for the optimum development of the lake's resources. The project's work program, already started, should gain considerable impetus when a project manager, to be stationed in Kigoma, is appointed. The information and technical assistance to be provided by the project should contribute significantly to properly planned and coordinated development. The project sets out to assess the lake's fish stocks in order to identify maximum sustainable yields and corresponding investment opportun- ities; to assist in improving the lake's fishery statistics services; to im- prove fishing methods, the exploitation of untapped resources, and the fish marketing structure; and to introduce improved fish processing methods. Under the project will be hired a project manager, fishery biologist, stock assess- ment specialist, fishery statistician, master fisherman, fisheries technologist, fish processing technologist, fish production and marketing economist, socio- logist, marine engineers and consultants on boat construction, civil engineer- ing, investment, and other services. The close coordination and consolidation of the work of the UNDP/FAO project with the operations of the regional fisheries staff would be highly desirable for project success and staff development. 19. Because so little information on the subject is available, the marketing part of the Project is of special significance. It will hopefully provide new information, inter-alia, on: i) The impact on the existing fish- ermen of industrial fisheries development; ii) The present and future market for dried, fresh and frozen fish and the subsequent ice production and refrig- eration requirements, possibly using existing transport and refrigeration facilities to experiment with the processing and marketing of fresh and frozen fish; iii) Protein deficiency in the hinterland villages of the region and the use of fish as a means to improve nutrition levels, including reference to such specific operational considerations as identifying target groups, income levels of target groups, profit margins, price variations, consumer preferences, and preservation characteristics; and iv) The existing and recoimnended owner- ship and marketing patterns of artisanal fishermen and middlemen. 20. The project hopefully will also examine the income generating cap- abilities of the lakeside ujamaa village, including whether the equipment already granted by Goverrment has generated sufficient income for replacement when worn out and existing and alternative systems to allocate the benefits of communal production. ANNEX 8 Page 7 21. Important experimentation will be undertaken of possible new equipment and techniques, including: i) Small purse seine units, including appropriate siting, chiromila nets, pelagic longlining, and gill netting; ii) Innovation and extension in the traditional sector through experimentation with lift nets, trimarans, catamarans, benthic longlines, improved benthic gill netting and handling, and the application of oars, sail and outboard engine to traditional fishing methods. The experiments in the traditional sector could best be carried out at the village level under village conditions and in conjunction with local fishery officers. The techniques of one region might be tested in another, e.g., light fishing in the southern part of the lake where it is currently practiced less. 22. Because poor processing methods render the fisherman victim of both the weather and the middleman, improvements in processing, pa,rticularly the drving of dagaa, will be given particular attention. 23. The Regional Fisheries Division has a sizable staff which, following upgrading and reorientation and in conjunction with the UNDP/FAO project, should provide the impetus for fisheries development on Lake Tanganyika. The Regional Fisheries Division is staffed by 67 personnel, of which 6 are assist- ants, 45 field assistants, 10 assistant field officers, and 6 field officers. There are 6 staff in Kasulu and Kibondo Districts, and 30 in Kigoma town. In addition in 8 fisheries stations along the coast of Lake Tanganyika are perhaps 20 staff, whose principal responsibilities are the collection of statistical information. 24. The training of regional fisheries staff has been primarily insti- tutional, and although representing a range of professional levels and disci- plines, has involved little field work. Emphasis has been more on the theoret- ical than the practical, more on biology than engineering and economics, more on statistical compilation than on new equipment and techniques. A wider range of disciplines in the training of fisheries personnel, and more interchange of personnel among regions, may be useful. Consequently, one of the principal constraints to regional fisheries development is the lack of technical know- ledge and experience relating to a variety of new fishing methods and equipment. If an independent and dynamic fisheries services is to be established, capable of the required innovation and extension work, then the training and activities of regional fisheries staff must be given a more practical on-the-job focus. Two 8-m boats under construction in Mwanza, for training in purse seine fishing, represent a first step. In addition, a sizable training component is included within the UNDP/FAO project which will hopefullv become more important as the project develops further. 25. Following government decentralization in 1972, Regional Administra- ti>a was made responsible for planning and implementation of projects in the regions and the ttinistries in Dar es Salaam for technical support. In the case of fisheries, some confusion continues to exist over the respective roles of the region and center on project preparation and investment. The roles of a ANNEX 8 Page 8 number of national parastatal organizations now being formed to service the fisheries sector (the proposed Tanzania Fisheries Development Corporation and its subsidiary, the Tanzanian Boat-building Corporation, the Tanzanian Fish Processing Corporation, and the National Cold Chain Operation) also require clarification. Project for World Bank Support 26. A pilot fisheries component was proposed by the preparation report on the Kigoma Integrated Rural Development Program, submitted by the Govern- ment in May 1973 and amended in October 1973 during the visit of the Appraisal Mission. It was proposed that the component include: (a) The introduction of small purse seine units to 14 lakeshore villages to be promoted and assisted through the services of a qualified fisherman; (b) The construction of a boatyard in Kigoma town, with expatriate manasement and training; (c) The improvement of lake fish carrier services; and (d) The provision of simple, improved fishing gear, in the form of beach seine nets, small powered vessels, light attraction units and drying racks to lakeshore ujamaa villages. Tqtal estimated cost of the project as revised and expanded would be approx- imately US$0.8 million. 27. A fisheries credit fund, to finance ujamaa fishing villages with simple new techniques and equipment, has been included under the project. Such a component would serve to accelerate ujamaa village development and im- prove the welfare of its inhabitants. Powever, the proposals for additional purse seine units, a boatbuilding yard, and improved carrier facilities were felt to be either inappropriate or premature at the time and warranted further study and/or elaboration. In general, insufficient information existed upon which to make investment decisions that would have a profound and as yet un- predictable impact on the fisheries and fishing community of Lake Tanganyika. Specifically, it was determined that: (a) The small purse seine units, requiring safe berthage and various services, must operate out of a harbor, probably Kigoma, and not from the scattered villages. They would thus make no contribution to ujamaa village development. Instead,they would employ only a limited number (about 30 each boat) of fishermen, who would be required to live in and work from an urban setting (Kigoma town); (b) Little is currently known of the effect on fish prices of any marked increase in industrial fishing. A drop in prices (especially of dagaa), which is quite possible, would ANNEX 8 APPENDIX A Page 1 THE RESOURCES AND ENVIRONMENT OF LAKE TANGANYIKA 1. The major pelagic resources of interest in the fishery development of Lake Tanganyika include four species of predators and three of prey. The predators are Lates augustiphous (called Nile perch, gombe, capitain and some- times sangala), L. marine (called sangala), L. microlepis (called nonzi), and Luciolates stappersi (called makeke). The prey are the sardines Limnothrissa and Stolothrissa (both called dagaa) and juvenile Luciolates (called dagaa also). The sardines are quite short-lived, maturing in 6-12 months, while the predators are much longer-lived. The Benthic resources in greatest demand include Sangala and Boulengerichromis (kune), as well as several of lesser importance. 2. Lake Tanganyika is a meromictic lake (never mixes top to bottom). Below about 170 meters the lake is toxic and devoid of life, due to hydrogen sulfide, other products of decomposition, and an absence of oxygen. The water is very clear and gives the appearance of low productivity, but high tempera- tures (23-28C) and an interchange of nutrient salts between the dense toxic layers and the surface waters maintains high productivity levels. The lake temperature stratifies and develops a thermocline at 60-80 meters in the wet season, but is mixed above the toxic layers in the windy dry season. 3. WTind significantly affects the lake's productivity. The most con- sistently strong winds occur from July to October. These North-South winds produce a tilting of the temperature layers and turbulence at lee and windward shores. This turbulence probably brings nutrients into the well-lighted surface waters and substantially increases production of phytoplankton and zooplankton. The pelagic sardines enjoy a rapid growth in biomass during this period Deak about October-November and then declining to a low in April-May. Areas of sufficient turbulence to increase local production may include at least the Zambia and Burundi ends of the lake, and embayments near Kalemie in Zaire and Karema in Tanzania. Little is known about turbulence-induced nroduc- tivitv in the lake, but the Kalemie and Karema areas appear to be conductive to an upwelling because of a more gradually sloping of the bottom. 4. The pelagic stocks off Kigoma seem to be holding up well, and the area should be able to sustain greater exploitation. Increased exploitation, how- ever, should be carried out extremely cautiously. Purse seines, lift nets, chiromila nets, longlines and drifting gill nets all offer potential for in- creased harvest. In addition, the benthic fishery may be developed with long- line, gill net, and with trawl in a few locations. Monofilament gill nets also offer possible gains in the clear waters of the lake. 5. Little is known of the homogeneity or heterogeneity of pelagic stocks. They may consist of one, a few, or many populations. They may be highly mobile or remain in one area for long periods. Stock heterogeneity, as well as esti- mates of growth, mortality, and biomass is of critical importance in assessing ANNEX 8 APPENDIX A Page 2 effects of the expanded fisheries on traditional harvests. In addition, virtually nothing is known about the density of pelagic stock that could be ^ished outside Burundi, Zambia, Kalemie, and the area very near Kigoma. In- dications suggest that fishable stocks exist in several areas of Tanzanian waters, although traditional fisheries are all limited to five km of the lake shore. Large areas of the coast are not fished because of inhabitable terrain on the steep, rocky shores. Moreover, little dagaa fishing occurs at night south of Kipili for reasons not well known, except that daylight beach seining i very successful there. KICMMA RURAL DEVELOPMENT PROJECT Rough Comparative Data on Lake Tanganyika Fishing Techniques Purse Seine i,each Seine (1b m, 3 light 6oats) 1 Man hours/dat ' 8 6 2. Manpower Rtequirements 31 24 2 3. Capital Costs a. Plank Canoes 2,400 TSPi 800 rsn (dugout canoes) (3,600 TSh) (1,200 TSh) bo Purse Seine boats 419,300 TSn c. Net 80,000 TSh 6,000 TSli 1 50 TSP d. Lamps 700 TSh 330 TSh 220 TSh 4. Depreciation ao Plank 6anoes 4 years 4 years (dugout canoes) (8 years) (8 years) bo Purse Seine uoats 10 years co Net 5 years 3 years 2 years d. Lamps 2 years 2 years 2 jears 5 Annual Lield 600 tons 25 tons 6 tons 2/ 6O Lnnual Itevenue 1,0>0,000 TSiI 43,750 Wh 10,500 DSP 7. Minus: a. Operatina Costs 330,000 TShP bo Depreciation 58,350 TSh 2,7 50 00 B. 8et Income 661 ,650 TSh 40,230 TSh 9,815 TS h, 9. Net Income/Man 21,000 TSh 1 ,700 rsn ,100 TSP 1/ Figures ass-me occasional second shift, witLh 25 additional manpower. 2/ Figures assume 1 .75 TSh per kg. ANNEX 9 Page 1 TANZANIA KIGOMA RURAL DEVELOPMENT PROJECT Recommended Qualifications and Responsibilities of Staff Financed Under the Project 1. Operations Manager of Regional Cooperative Union (Union) - Would be directly responsible to the General Manager of the regional cooperative union. Qualifications - Bachelor's degree or equivalent work experience and at least ten years' successful work experience in handling logistical matters, at least three of which in a senior position. - Must be capable of designing and instituting documents and procedures for the control of collection, transport, storage and transfer of large numbers of lots of a variety of goods. - Must have experience in maintaining and operating road transport equipment. - Experience in handling agricultural commodities would be useful. - Work experience in a developing country (preferably Africa) essential. Personal Characteristics - Person must be adaptable. - If candidate does not speak Swahili, some facility with languages would be necessary. - Must be able to cooperate with a variety of other agencies to get the job accomplished. - Must be capable of coping with ambiguities of roles and relationships. - Must have an ability to train subordinates and delegate authority while retaining efficiency. ANNEX 9 Page 2 Duties and Responsibilities: - Setting up as necessary and initially managing the Union's systems and procedures for: transport, warehousing, vehicle operation and maintenance, shipping, vehicle hire and leasing, quality grades and standards and control, forecasting and scheduling. The Operations Manager would give great attention to staff training. During the last six months of his stay the Operations Manager would be expected to be an adviser only, leaving actual operating to the person selected ,as his replacement. - Planning the Union's facilities and equipment requirements. - Liaising with villages, transport firms, parastatals, TRDB and regional government in handling crop inputs and outputs, vehicle use and repair, and other matters. - Hiring transport from other parties to handle the Union's peak season requirements and leasing out the Union's own transport during slack periods. - Working closely with the financial controller in the credit program to village primary societies. - Staff training and occasional general training at the rural training centers. - Reporting at least monthly to the Union's management and to the Regional Development Director (RDD) on progress, problems and issues, together with recommendations. - Initiating requests for procurement for the Union's supplies, spares and equipment. 2. Financial Controller and Deputy General Manager - Would be directly responsible to the General Manager of the regional cooperative union. Qualifications: - Qualified accountant (e.g., chartered accountant). - At least ten years' successful work experience partly in a developing country of which three years would have been as financial controller or equivalent Involving accounting, budgeting, credit management and collection, cash and inventory management and financial reporting. ANNEX 9 Page 3 Personal Characteristics: - Same as for Operations Manager except that knowledge of Swahili less imDortant. Duties and Responsibilities: - Setting up and initially operating the Union's systems and procedures for: bookkeeping and accounting; budgeting; financial and cash forecasting; credit management and collections; cash and inventory management; financial reporting; and internal auditing. The Financial Controller would give great attention to staff training. During the last six months of his assignment the Financial Controller would be expected to be an advisor only, leaving actual operating to his deputy. - Liaising with villages, parastatals, TRDB, other banks and regional government. - Working with the "Training Supervisor" in setting up Village-level standardized accounts and control systems, super- vising their implementation; and from time to time, participating in training village bookkeepers and others. - Screening and recommending action on all requests for credit from primary societies. - Procurement. - Reporting at least monthly to the Union management, the RDD, and, as required, TRDB, on the Union's overall financial situation; together with relevant projections and identification of problems or issues and their recommended solution. - Reporting at least quarterly to IDA as required. 3. Training Supervisor General: The project would involve supporting, through providing infrastructure and loans, a large number of villages (estimated at 135 during the five year disbursement period). To qualify for loans the villages would first have had to become registered as "multipurpose primary cooperative societies" (ANNEX 14). In addition, each village (primary cooperative society) is projected to become a sizeable enterprise involved in handling members' crop sales and purchases of inputs, storage of inventories, record-keeping, handling large amounts of cash and other assets. Adequate business management and bookkeeping arrange- ,ments are therefore required. To qualify for loans a village requires a fulltime hired bookkeeper, probably from a village in the region and trained ANNEX 9 Page 4 in the region. Therefore, a training supervisor, a principal task of *whom would be training and supervising village-level bookkeepers, is essential. He would report directly to the Financial Controller. Oualifications: - Bachelor's degree in commerce or eauivalent, or a recognized accounting qualification. - Experience in developing and instituting simple accounting and control systems, selecting successful candidates and supervising initial working performance. - Experience in small enterprise management and organization: goal setting, simple personnel practices, commercial practice (e. g., contracts), planning and record-keeping and credit management. - Some experience witlL cooperatives and with agriculture useful. - Adequate experience in developing countries essential. - He should be an active man and be willing to spend considerable amount of his time visiting villages. Personal Characteristics: - Same as Operations Manager of Union. Duties and Responsibilities: - Preparing a standard, simple bookkeeping and control system to be followed by the Region's primary cooperative societies. In this, he would liaise closely with the Regional Development Director's (RDD) office and would examine similar systems prepared elsewhere in Tanzania. - In close cooperation with the RDD's office, establishing and imple- menting programs of classroom and on-the-job training to develop village-level bookkeepers. Course candidates would have completed Standard VII. Training might consist of two to four weeks' classroom instruction on the workings of the village standard set of accounts, followed perhaps by four months' on-the-job training under close supervision. Initially, the training supervisor would have to provide considerable on-the-job supervision; as the project developed he would be assisted by the Union's Society Credit Officers (S00's) and by previously trained village-level bookkeepers and secretaries. ANNEX 9 Page 5 - Assisting in initial training and continuing supervision of the work of SCOs. - Providing general assistance on the Union's credit matters. - Acting as deputy to the Financial Controller, as required. 4. Land Use Planning Officer Qualifications: Formal training in agriculture, soil science or land resource use, land surverying and photogrammetry. Previous experience in land resource work and associated land and agricultural planning would be essential. Duties and Responsibilities: - Undertake a photo analysis of the project area and prepare land use resource maps and inventory of the area. - Prepare land use resource maps of proposed village areas for incorp- oration in Village Site Feasibility Reports (VSFR). - Advise on the location of roads and other infrastructure proposed for ujamaa villages in VSFR. - Give general guidance on land resource use with particular reference to soil and water conservation systems for ujamaa villages to the Regional Development Director and his staff. - Concurrently with the above tasks, train Tanzanian staff in land use and associated techniques. The Land Use Planning Officer would be directly responsible to the Regional Agricultural Functional Manager. 5. Senior Training Officer Qualifications: Formal training in general agriculture with particular reference to small scale production of annual crops. Training experience under Tanzanian conditions, fluency in Swahili and exposure to conditions in Kigoma Region would be essential. Duties and Responsibilities: - In consultation with training specialists in the Manpower Development Derartment of Kilimo and the Regional Agricultural Manager to devise and develop suitable curricula for "induction" courses to be given to farmer contact extension personnel at the proposed Kigoma Region Training and Trials Center. ANNEX 9 Page 6 In consultation with the Regional Agricultural Manager and the Trials Officer to devise and develop appropriate short courses for in-service training of extension personnel, and for village leaders in crop production and related subjects. Operation of the training wing of the oroposed Training and Trials Center including student selection, discipline, supervision of subordinate staff and teaching of selected courses. - Liaison with the Trials Officer in utilization of joint facilities at the proposed Training and Trials Center. The Senior Training Officer would be directly responsible to the Regional Agricultural Functional Manager. 6. Trials Officer Qualifications: Bachelor's degree in agriculture or related subject, with at least three years subsequent experience in field trial techniques under Tanzanian conditions utilizing simple statistical layouts. Duties and Responsibilities: - In consultation with the Director of the Ukiriguru Research Center and the Regional Agricultural Manager to formulate and carry out programs of field investigations on the major crops of Kigoma Region with particular reference to maize, cotton and groundnuts. To the extent possible, field trials should be replicated at suitable locations of the proposed Training and Trials Center and should be designed to provide information which is readily applicable by farmers in ujamaa villages. - In consultation as above to advise on the form and content of demon- strations of improved practices to be carried out in ujamaa villages. Such demonstrations would be designed to popularize cultural practices of proven viability under farmer's conditions in Kigoma. - Multiplication of nucleus stocks of seed of maize and groundnut varieties in consultation with the Regional Agricultural Manager. The Trials Officer would be directly responsible to the Regional Agricultural Manager in administration matters and to the Director of the Ukiriguru Research Station on professional affairs. 7. General Manager, Kigoma Regional Cooperative Union Qualifications: A senior civil servant or equivalent position in the private sector with at least 15 years experience in responsible posts and several years in a senior executive position in a major public corporation or private company. ANNEX 9 Page 7 In previous employment he would have undertaken senior administrative responsibility and would have a thorough understanding of business and/or cooperative procedures. Duties and Responsibilities: - Overall supervision of all operations of the regional cooperative union. - Preparation and submission of draft policy statements and execution of policy directives from the board of the regional cooperative union. - Preparation and submission of progress reports to the board of the regional cooperative union and such other agencies as may require them, e.g., IDA, TRDB, NBC. - Supervision of Union relationships with primary cooperative societies. The General Manager would be responsible to the Board of Management of the regional cooperative union. ANNEX 10 Page 1 TANZANIA KIGOMA RURAL DEVELOPMENT PROJECT Regional Planning, Project Preparation and Evaluation Introduction 1. Following the decentralization of government (Annex 4), development of government's investment program will tend to depend more and more on the regional capability to prepare a stream of investment proposals within a co- herent regional framework or plan. While staff at the center retain powers and functions relating to overall economic policies--including pricing, ex- change, taxation and interest rate policies, the 18 regions have effectively taken over responsibility for all but 'national' projects (as for example, the relocation of the capital city). As detailed in Annex 4, relatively in- experienced or newly promoted staff now occupy many of the posts that carry major responsibility for plan and project preparation. Thus the need for additional skills relating to regional planning, project identification and preparation, while always urgent, has been accentuated as a consequence of the new policy emphasis. Accordingly, support for an intensified technical assistance effort for regional planning and project preparation, to include responsibility for the evaluation and monitoring of experience in Kigoma, would be included under the project. 2. Most of the new regions are heavily rural and agricultural in orientation, and the integrated rural development approach, based on ujamaa villages, is to be the general model. Hence, in these special circumstances, activities relating to rural development project preparation would emcompass the major task of regional planning. 1/ A team of three or four appropriately qualified and experienced personnel, to work in a region over a period of 6 to 8 months in association with regional staff, would be able to outline a regional plan, involving the identification and preliminary preparation of projects suitable for national or international financing. It is anticipated that the association of high level regional staff in close working relation- ship with technical assistance personnel would yield important training benefits. 3. A direct link between the technical assistance effort and the Kigoma project would be envisaged through the involvement of the technical assistance group with the early stage implementation in Kigoma region, including the establishment of a suitable monitoring/evaluation system. The Kigoma project is important to Tanzania as a first or pioneering effort paving the way for subsequent programs that explore the potential for achieving substantially increased incomes in cash and kind, among groups whose need for income can 1/ Particularly since shaping or influencing development through fiscal means is ruled out by the limitations on regional authority. ANNEX 10 Page 2 reasonably be described as very great by any standard. The projects will cover the second stage in the development cycle of an ujamaa village--that stage where concern shifts from the process of resettlement and organizing of the village community to the process of developing the newly established community as a viable and self-reliant economic unit. There are great uncertainties in moving into this second phase and the full range of problems is likely to show up only as the program gets underway. Hence a flexible and innovative management response will be needed, and this would be facilitated by an organized system for documenting and monitoring the progress of the project. Such a facility would also ensure that the experience and problems in Kigoma are fully recorded for the benefit of Tanzania's programs more generally. Staffing and Organization 4. Skills required in the preparation of such programs are those of an economist, an agriculturalist/land use planner, and a water engineer (who would be sufficiently broad-based to take responsibility for general engineer- ing aspects of other physical structures). A specialist in cooperatives and marketing, including training aspects, would also be very desirable; it is anticipated that such a person would be supplied by the Tanzania Rural Develop- ment Bank (TRDB) or through short term consultancies. Other specialist skills required for short periods would cover education and training, sociological matters, health, fisheries and forestry. In addition, provision has been included for consultants to work primarily at national level, to contribute to the development of national policies in respect of a number of problems common to the regions such as: system of regional planning; training; the provision of health care, water and education; and possible fiscal solutions to the problems of financing the regional programs. 5. The Project Preparation Team Coordinator would ideally be a Bank Staff member of some years standing. He should have experience of procedures relating to project preparation and implementation, and also be familiar with the analysis of sectoral and national policy issues. Ilis duties would include: (a) Assisting government with the recruitment of staff for the technical assistance team, including consultants. (b) Providing overall direction and guidance to the team, including specific work plans and schedules. (c) Logistic support and, as necessary, liaison between central government agencies and technical assistance staff in the field. (d) Initial responsibility for the monitoring and evaluation of the Kigoma project. (e) General liaison between the Bank and the Tanzania government for matters relating to the technical assistance program, based on thorough knowledge of relevant practices and procedures of both the Bank and government. ANNEX 10 Page 3 It is anticipated that the coordinator would be subject to direction from the Prime Minister's office. In regard to operational matters relating to the Bank, he would liaise with the Bank's office in Nairobi. 6. All members of the permanent team must have experience in project preparation and/or implementation in rural areas in developing countries, preferably in Africa. Staff would be sufficiently senior as to be able to work closely with high level regional staff in the preparation of projects and regional plans, and provide guidance and advice to them. 7. The Economist should be experienced in practical economic planning, including regional planning, and familiar with sources and uses of economic data. He should also have experience of peasant farming systems, including marketing and credit problems. He should be able to present arguments and analyses cogently and lucidly, both in a verbal and written form. He would take a leading role in developing a regional strategy or overall plan and relating project preparation to such a plan. He would ensure that the regional proposals were in conformity with national guidelines and objectives for regional and rural development, and with local needs and sentiments. He would work closely with regional staff in collecting and collating local sources of data and in developing appropriate systems of economic intelligence for the region. 8. The Agriculturalist should have extensive experience of tropical farming systems, preferably in Africa. He should be thoroughly familiar with all aspects of advisory and extension services, including the application of research findings under field conditions, and replication of improve practices. A knowledge of Swahili would be a decided advantage. He should also be aware of the economic implications of his recommendations for agricultural programs and projects. He would take major responsibility for the identification and preparation of investment programs to assist the development of small farmers, including the establishment of appropriate regional organization, and local systems of training. He would liase very closely with personnel dealing with marketing of outputs and distribution of inputs to farmers. 9. The Engineer should be thoroughly experienced with work 'in the bush', involving flexibility and innovative ability in coping with adaption of local materials and structures, understanding of labor-intensive construction methods, and managing with uncertain supplies and periodic shortages. He would be responsible for ensuring that the engineering features of any proposed projects were basically sound and reasonable in relation to local needs and skills. He would also be able to provide rough costings as a basis for financial and economic assessment of the proposed investments. 10. It is anticipated that the first step under this program would be the appointment of the Coordinator, who would initially work closely with the Tanzanian government and IDA in Washington and Nairobi, to (a) recruit for the regional teams, (b) agree on a list of regions for the team to work in, and (c) locate suitable specialist assistance for design of the monitoring/ evaluation of Kigoma project. ANNEX 10 Page 4 11. Finance would be included under the project for a two year program, sufficient for the regional teams to service three regions. While much of their work, and the work undertaken centrally, would be generally relevant to the rural areas of Tanzania, a large number of regions could not obtain the direct assistance of the team. Depending on experience with the technical assistance effort proposed, and on experience with Kigoma project in general, consideration could be given at a later stage to a more ambitious program of technical assistance relating to rural development. The Monitoring/Evaluation of Kigoma Project 12. After consultation with central government agencies and the regional administration, the detailed design and initial implementation of the system would be undertaken in the region by the technical assistance team supplemented with specialist assistance, at the beginning of the project. A consultant analyst would take major responsibility for overall supervision, including field work, arranging for local supporting services and liaison with local researchers (if he were not himself a local researcher). It would be expected that he would be available from time to time over at least a five year period to supervise ongoing work and to take responsibility for the analysis of the results. He should have some years of experience in undertaking empirical research in rural areas in developing countries preferably in Africa. He would probably have published work to his credit. He might hold a senior appointment in an established university or research institute or recognized business enterprise specialized in research work. 13. Major inputs for the evaluation of the project are likely to take the form of an initial base-line survey, to establish conditions at the point of departure of the project, and a terminal survey, five years later. The evaluation should attempt to explain developments under the project, examin- ing technical, economic, managerial, sociological and anthropological factors. The initial intensive phase of work would also provide the framework for an ongoing recording system at a more modest level, to be undertaken under the responsibility of the regional and district planning officers. At this level, the effort should focus on the needs of management for a form of early warning system, so that incipient problems can be rectified quickly or otherwise dealt with at minimum disturbance and cost. It is expected that this work would need to be supplemented from time to time with ad hoc special studies, dealing with specific problems. The planning officer would be assisted from time to time by the consultant analyst, some whose visits might well be timed to coincide with supervision missions. These data, together with other material, should enable a proper analysis and interpretation of experience under the project, including success In terms of initial objectives and in the light of problems subsequently encountered. 14. It would be inappropriate at this stage to attempt a detailed description of the proposed monitoring/evaluation activities. However, some of the high priority issues for investigation would be the following: ANNEX 10 Page 5 (a) Village development Relating to a sample of villages (including some non-project villages), covering the major ecological zones i.e., the highlands, the intermediate zone, the miombo woodland (with tsetse infestation), and lakeshore fishing: i. Progress in project relative to non-project villages, for various years following settlement of the project villages, having regard to - clearance and cultivation of the block farms (total acreage and acreage per family), adoption of improved techniques, use of fertilizers and other inputs, cropping patterns; yields and yield variation per cultivator and per village; use of credit, acquisition of infrastructure. ii. demographic characteristics of selected villages - sex and age structure, composition of families, social or tribal structure, types of family; stability of village population, including mDvement into and out of the village area. iii. Work activity, to include level of activity, seasonal variation, labor utilization per crop and cultivation technique; by age group and sex; off farm activities; availability of additional labor. iv. Income and consumption, per village and per farm family, covering subsistence consumption and marketed surplus; off farm income; disposal of cash income (cattle purchase, bride price, consumer goods - effective of promixity of local market centers, role of village shop and the STC); savings. v. Adequacy of nutrition - calorie, protein and vitamin deficiency; possible upgrading through the introduction of a dagaa dietary supplement. An ad hoc study of the spending patterns of the village elite today - e.g. school teachers, agricultural extension workers, village cooperative officers - might provide useful insights as to what might happen after incomes begin to rise under the project. (b) Use of Village facilities i. Water supplies - to include costs of maintenance and operation frequency of repair and periods in relative to periods out of service, for the different types of system (with particular reference to the dug well/hand pump vs the piped systems); estimation of time savings relative to use.of traditional water sources; social attitudes to dug wells and other facilities; continuing use of traditional systems if any. ANNEX 10 Page 6 ii. Education - profile of village school teachers - social and geographic orgin, length of experience, marital status; relationship of these factors to stability in villages (also to include reference to the availability of teachers' houses in the villages); school attendance and non-attendance - relationship to the availability of school places and village size; effectiveness and intensity of use of the school workship, longevity of tools and facilities. iii. Health Care: Intensity of use of dispensary - relationship to type of service provided (personnel available), and location; ditto for the rural medical center; costs of providing these services; development of and response to the public health program - including excreta disposal schemes. (c) Intensive studies of villages which turn out to be extremely success- ful (say the best 1 or 2 of those at a particular stage of development) with a view to identifying causal factors for replication elsewhere. Ditto for extremely unsuccessful villages. (d) Regional level studies i. Motor transport - costs and operating efficiency of trucking operations, for use in determing the eventual needs and costings of the regional union, including mileages operated, length of daily runs, maintenance and repair costs, frequency of breakdown records. ii. Performance of the local construction industry - availability of local contractors, length of time taken to complete contracts, use of skilled labor; construction costs of local materials vs cement blocks. iii. Impact of village development on the region - activity of major market centers; private transport to and from villages; use of bus services; expansion of the banking system. TANZANIA KIGO4A RURAL DEVELOPMENT PROJECT Cost Estimates Ut Year) (TSh 'OOOs) I II III IV_ V Total US$ 'OOOs A. Villa-te Infrastructure Water Supply (Table 2) 2,507 2,520 2,533 2,546 2,559 12,665 1,774 Health (Table 2) - 110 410 540 490 1,550 217 Education (Table 3) 3,595 3,225 2,898 2,894 2,822 15,434 2,162 Feeder Roads (Table 4) - 202 252 252 252 958 134 Tsetse Clearing (Table 4) 214 160 160 160 148 842 118 6,316 6.217 6,253 6,392 6,271 31,449 4,405 B. Loans to Villages Seasonal Inputs Revolving Fund (Annex 15, Table 9) - 630 2R'l 2,i*67 2,658 7,70A 1.079 Med. -Term Credit for Prod. Infrast. (Table 5) - 1.059 1,396 1,426 1.593 5,474 767 Sub-Total - 1,689 3,247 3,993 4,251 13,180 T1 _ C. Loans and luNlt Contribution tf ReSional 967 1,431 1,052 811 1,437 5,698 798 Cooverative Vnion (Table 6) D. Regional Infrastructure Rural Training Centers (Table 7) 670 1,050 48 - _ 1,768 248 Radio-Telephone Hookup 200 150 - - - 350 49 Livestock Pilot Project (Table 8) ) 257 500 114 2 873 123 Agric. Trials and Training Center (Table 8) Vehicles 1/ 185 95 185 95 _ 560 78 Demonstrations (Ain= 15, Table 10) - 43 53 58 60 214 30 Aerial Photography 2/ 400 1 200 200 - - 1 800 252 Sub-Total 1,712 .038 600 155 60 5,'5%5 780 E. Technical Assistance (Table 9) Salaries & Related Fringe Benefits 230 850 715 625 325 2,745 384 Housing, Vehicles & Equipment 505 480 125 125 125 1,360 190 Training - 183 193 193 193 762 107 General Funds 205 205 205 205 205 1,025 144 Sub-Total 940 1,718 1,238 1,148 848 5,892 825 F. Project Preparation and Evaluation (Table 10) 475 2J615 2.100 409 335 Si925,02 - TOTAL 10,410 16,708 14,490 12,899 13,202 67,709 9,484 Contingencies ( Price 24,012 Physical 2,888) 26,900 3,767 TOTAL PROJECT COST 94,609 13,251 1/ For 2 lorries (@ TSh9O,400) and 8 four-wheel drive vehicles (@ TSh47,400) to supplement existing transport in meeting increased supervision and materials movement by the regional administration during the project. 2, lne region encompasses 37,000 km2, of which an estimated 2/3 (about 25,000 km2) would require aerial photography. The cost for the photography, 3 sets of prints, simple viewing devices, 3nd Aho,nt ` weeks training in Kigonma for staff in interpreting photographs, is estimated it TSh&O per ,T2 for a tmtal f I'Sh2 ,000,000(I. iOnuary 3' 197'r TANZANIA ANNEX 11 Table 2 KICOMA R-JRAL DEVELOPMENT PROJECT Page 1 A. Water Supply rI pica] PunmedSnLstemY (T Sh 000's) Preparatory Work 6.5 Pumrphouse Structure 8.8 Intake 1.0 Storage Tank 11.5 Transport 20.0 Plant Hire 6.5 Rising Main (100 m elev.; 2,000 m length) 25.8 Pum.p 28.5 Distribution System: Pipe (2,470 m) 36.9 Hydrants 1.8 38.7 Sub-total 147.3 Supervision and Other Overheads 58.0 Total 205.3 Source: Maji and mission estimates V'pical Village Using Dugwells and Handpumps i dugwell with handpump T Sh 1,510 For total village (assumes 15 wells) T Sh 22,600 Phasing of Costs: 110 villages would require water systems during the project period. It is expected that half would be by pumped systems and half by dugwells with handpumps (Annex 6 ). Per Year (years I through V) 11 villages with pumped systems (at T Sh 205,300) T Sh 2,258,300 11 villages with dugwells and handpumps (at T Sh 22,600) T Sh 248,600 Total T Sh 2,506,900 -2- ANNEXll Table 2 Page 2 Privy Program This would cover 10 percent of fanilies in project villages over the five-year project period. - Total families in project villages (135 villages at 350) 47,250 - 10 percent of families would be 4,730 - Each slab would cost an estimated T Sh 27.40 (Us$4.00) Summary - Water Supply and Privy Program Year I Year II Year III Year IV Year V Total (T Sh 000's) Water Supplies -2,507 2,507 2,507 2,507 2,507 12,535 Privy Program -- 13 26 39 52 130 Total 2,507 2,520 2,533 2,546 2,559 12,665 3. Health 1. Dispensaries and rural health centers (RHC) are the identified objects of expenditure. Dispensaries are estimated to cost T Sh 30,000/-, rural health centers, T Sh 400,000/-. 2. While Kigoma Region is not below average in respect of these facilities compared with other parts of Tanzania, the dispersion of population associated with the settlement of new villages would justify some additional health investments over the life of the project, as would improving standards for rural health care for Tanzania as a whole. 3. It is proposed to include 10 dispensaries and 3 RHC, with expected timing as follows: Year I Year II Year III Year IV Year V Totals Dispensaries Number -- 2 2 3 3 Cost -- 60,000 60,000 90,000 90,000 300,000 RHC 50,000 350,000 450,000 400,000 1,200,000 110,000 410,000 540,000 490,000 1,500,000 ANNEX 3 Table 3 TANZANIA KIGOMA RURAL DEVELOPMENT PROJECT Education Cost Estimates and Phasing 1. 60 Existing Ujamaa Villages i. Data for 73 of the Phase I and II ujamaa villages show primary schools to be distributed as follows in 1973: Grades Percent of All Schools I - VII 34.2 I - VI 8.2 I - V 4.1 I - IV 9.6 I - III 12.3 I - II) 31.5 The backlog of classes over appropriate facilities can be calculated as: Classrooms 83 Teachers' Houses 145 N.B.: -6 classrooms per I - VII complete school; with Grades I - II sharing a classroom. -7 teachers per 7-standard school; 5 until Grade VII is added. -1 house per teacher. ii. Applied pro rata to 60 villages, the backlog is: Classrooms 68 Teachers' Houses 85 The 60 villages will have received a share of the 1973/74 building program for the Region of 112 classrooms ( @ 10,000) and 64 teachers' houses ( 15,000). Assume one-half of all classrooms to be constructed in ujamaa villages and two-thirds of all teachers' houses. After application pro rata to 60 villages, the backlog in these villages is reduced to 2h classrooms and 85 teachers' houses. ANNEX .11 T-abIe 3 Page 2 Over the project period, teachers' houses would be provided up to 5/6 of the level that would provide 1 house per teacher. This allows for the joint use by bachelor teachers of a single house. For 1974, the local need for housing equals the number of teachers required, for 60 villagesequal to 229. Hence, at proposed standard, 1/6 of 229 = 38 from ba log. Hence, backlog of teachers' houses becomes85-38 * 47. iii. For rehabilitation to existing schools in 60 villages, we assume 20 percent of existing (permanent) classrooms need replacement-- i.e. 25 in total in the 60 villages. iv. Hence, total replacement/additions at beginning of project period in 60 villages would be: Classrooms: 25 + 24 = 49 Teacherst Houses: 47 v. Phasing: No. Year I Year II Year III Year IV Year V Teachers' Houses 47 20 20 7 - 15,000 each, cost 705,000 300,000 300,000 105,000 - Classrooms 24 10 10 4 - - 25 - - 10 10 5 10,000 each, cost 490,000 100,000 100,000 140,000 100,000 50,000 N.B.: Phasing of rehabilitation in line with equalizing overall building program. vi. Extensions to 60 schools over the Project period: All Grade II and above will be completed for a single stream I - VII complete primary school. 1973/74 Grade I schools, completed through Grade VI. From data for existing schools, applied pro rata to the 60, total additional classrooms will be: 137, and teachers' houses 163, less 1/6, 136. ANNEX 11 Table 3 Page 3 Phasing: Total Year I Year II Year III Year IV Year V Teachers' Houses 136 23 30 32 27 24 15,000 each 2,0o40,000 345,000 450,000 480,000 405,000 360,ooo Classrooms 137 26 34 32 26 19 10,000 each 1,370,000 260,000 340,000 320,000 260,000 190,00D N.B.: Phasing according to need. vii. Workshops and tools are to be provided as the school reaches Grade VI. There may be a few workshop buildings in the 60 villages--we suppose--but tools are virtually non-existent. Phasing: No. Year I Year II Year III Year IV Year V Workshops 58 20 12 7 10 9 '; 15,000 870,000 300,000 180,000 105,000 150,000 135,000 Tools 60 22 12 7 10 k 30,000 1,800,000 660,000 360,000 210,000 300,000 . ,00( viii. Head teacher's offices and toilet blocks: Rehabilitation is estimated for 23 offices and 35 toilet blocks: Phasing: Total Year I Year II Year III Year IV Year_ Offices 23 8 8 7 - - 9 3,000 69,000 24,000 24,000 21,000 Toilet blocks 50 30 20 - - - k 2,400 120,000 72,000 48,ooo ix. Textbooks, desks and library: see Section 3. ANEX 11 Table 3 Page 4 2. Facilities in 75 New Villages i. Under Phases I and II, about one-third were newly started schools, one-third completed schools and one-third partially completed. The new program for 75 villages is assumed to absorb 35 existing schools (from 104 existing, including 9 urban), and that 40 new schools will be started. ii. Of new schools, we assume 8 will be started each year. Hence, classroom and teacher house needs are projected as follows: Total Year I Year II Year III Year IV Year V New Schools 40 8 8 8 8 8 Teachers' Houses 74 7 7 13 22 25 15,000 1,110,000 105,000 105,000 195,000 330,000 375,000 Classrooms 88 8 8 16 24 32 @ 10,000 880,000 80,000 80,000 160,000 240,000 320,000 iii. Extension to existing (35) schools: We assume 20 are complete Grades I - VII: 10 are Grades I - VI; and 5 are Grades I - V. We further assume that on coming under the program, schools will be 30 percent deficient in classrooms and 40 percent deficient in teachers' houses. iv. Hence, for teachers' houses, total houses needed equals 35 X 7 = 245, minus 1/6 = 41, = 204; Stock on hand, 20 X 7 X 0.6 = 84, plus 10 X 5 X o.6 = 30, plus 5 X 4 X o.6 = 12; total 12. Hence, the housing deficit is 78. v. For classrooms, we have a need for 35 X 6 - 210, and stock on hand 20 X 6 X 0.7 = 84; plus 10 X 5 X 0.7; plus 5 X 4 X 0.7; total 133. Hence, the deficiency is 77 plus, for classes in need of rehabilitation, 20% of 133 = 7 , hence the total of 104 classrooms. ANNE 11 Page 5 vi. Phasing: Total Year I Year II Year III Year IV Year V Teachers' Houses 78 18 18 16 13 13 4 15,000 1,170,000 270,000 270,000 240,000 195,000 195,0)0 Classrooms 4 10,000 Additions 77 16 16 17 14 14 Rehabilitation 27 5 5 5 6 I 104 21 21 22 20 2) 1,040,000 210,000 210,000 220,000 200,000 200.0(?Y vii. Workshops and Tools: Are supplied after Grade V Phasing: Total Year I Year II Year III Year IV Year 1 Workshops 35 5 9 7 7 4 15,000 525,000 75,000 135,000 105,000 105,000 105,000 Tools -----as above---- g 30,000 1,050,000 150,000 270,000 210,000 210,000 210,000 3. Other Elements i. Desks, textbooks and library: Desks supplied to new classrooms at 20 per room; for all other classrooms at one-third this rate for rehabilitation. ii. Textbooks, sets as necessary and per classroom. Libraries to all schools. ANNEX 11 Table 3 Page 6 Phasing: Total Year I Year II Year III Year IV Year V Desks: 20 per new classroom 9 120/- 5,48o 880 1,040 1,140 1,200 1,220 Rehabilitation 2,440 420 460 460 500 600 7,920 1,300 1,500 1,600 1,700 1,820 Costs 950,400 156,000 180,000 192,000 204,000 218,400 Textbooks: s Set 7,613/-: Grade I - VI 5,544/-; I - V 3,545/-; I - VI 2,100/-; I - III 1,260/-: I - II 930/-; I 425. Total Year I Year II tear III Year IV Year V For 60 villages: 51 complete, 9 Grade I-VI 440,400 266,000 37,600 50,60o 47,500 38,700 For 75 villages: 40 complete, 8 I-V, 8 I-X7, 8 I-III, 8 I-II, 8 I 342,050 52,500 604oo 699280 76,060 83,710 732,450 11,500 9d,00 123,560- Library: Total Year I Year II Year III Year IV Year V 135 103 8 8 8 8 1,000 135,000 103,000 8,000 8,000 8,ooo 8,000 ANNEX 11 Table 3 Page 7 Education - Swmiay Costs and Phasing (T. Sh.) Year I Year II Year III Year IV Year V Total 3,595,100 3,224,700 2,897,500 2,8941,200 2,822,400 15,433,900 Teachers' Houses 1,020,000 1,125,000 1,020,000 930,000 930,000 5,025,000 -assrooms 650,000 730,000 840,00o 800,000 760,000 3,780,000 - ols 810,000 630,000 420,000 510,000 480,000 2,850,000 Workshops 375,000 315,000 210,000 255,000 240,000 1,395,000 Desks 156,000 180,000 192,000 204,000 218,400 950,400 Textbooks 318,500 98,100 U g9,900 123,600 1225400 782,500 Toilet Block 105,600 81,600 33,600 33,600 33,600 288,000 Head Teacher's 57,000 57,000 54,000 30,000 30,000 228,000 Office Library 103,000 8,ooo 8,000 8,000 8,000 135,000 Table TANZANIA KIWRA RURA1 DEVELOPMENT PROJECT Feeder Roads An average of 4 km of road upgrading to connect the village to the nearest road is provided for. The timing of the improvement would correspond with villages generating marketable surpluses and becoming eligible for credit under the project. The work would be performed by gangs consisting of seven unskilled men tli;s a leader. The unskilled labor would be on a modified self-help basis, at T Sh. 5 per day, rather than the T Sh. 11.20 per day normal govern- ment r-te. Unit cost would be T Sh. 2,527 per km. Phasing and costs would be as follows: ----------Year ------------- I II III IV V Total Number of villages 20 25 25 25 95 Kms of road 80 100 100 100 380 Tost (T Sh.) 202,000 252,000 252,000 252,000 958,ooo 3. Tsetse Clearing Assumplt-ions '. 35C family villages being settled at the rate of 15 per Sear for 5 fears. '. clearing of two acres per family in years one and two of settlement for 1:!,-h payment required. '. 'vor .ge labor input for tsetse clearing is 20-25 man days/acre. -- .y-,ent range 1972/73 T Sh. 5-15 per acre (say T Sh. 15 per aare). '. At r Sh. 30 per f_oily, cost per village = T Sh. 10,500 cost for 75 villages= T Sh.787,500 O. .viJ cost of one additional tsetse clearance survey team irnvolving the following equipment: I wheel drive vehicle T Sh. 50,000 I',ntige 3,000 J,irvey equipment 1 000 - ---------Years----- Sh. ----------------------------- 6'UtUtlarV Tse-tse Cltarinz I II III IV V Total J', oring 160.0 1]60.0 160.0 160.0 147.5 787.5 Equipment 54.o - - - - 54.o 214.0 16o.0 160.0 160.0 14?.5 841.5 TANZANIA KIGOMA MTA%L NWPTPRPJIO B. Estimated Msdium/Long Teom Credit Nee" by 'rear and Type vr ige produoTive infrastructure -Project -eur-- -------- -- 2 i i 5 Total Thrrber of Villages Rflcoiving Gredit 20 45 70 95 95 - - - - - - - - - -T 9b# -- -- ----------- - - -- - Haad aisze Shdllers at T Sh. 3750 each / 75,000 93,750 93,750 93,750 356,250 (No. of Units) (20) (25) (25) (25) (95) Mzine Mill, 20 HP engine, building and installation at T Sh. 22,800 installed 228,000 342,000 342,000 456,000 1,368,000 (No. of Units) (10) (15) (15) (20) (60) Godowrn at T 3h. 36,800 each 2/ 736,000 920,000 920,000 920,000 3,496,000 (No. of Units) (20) (25) (25) (25) (95) Fishing Nets aud Other Gear / 20,000 40.000 70.000 1231000 Totals M l^-a a 1,425 1,592.750 5.4 .250 1/ Price of Treadle Maize Sheller delivered and installed. 2/ Each 350 family village is estimated to produce about five 7-ton lorry leads of marketed surplus (i.e., 60 bags of maize, 60 bags beans or 300 bags cotton) per week in the period September 1 to November 15. Each godown would be planned to hold 3 weeks' marketed surplus, equivalent to 15 lorry loads or about 115 tons (actual weight would generally be less where cotton is stored, due to its lower density. The estimated cost includes T sh. 600 for a simple portable platform scale and T sh. 150 for a simple strongbox, locked and cemented into the ground; together with simple office facilities. i Twenty villages (about 50% of the region's fishing villages) are expected to qualify to borrow under the fisheries credit fund. The amount of credit was estimated on the following basist T ah. Beach seine net units (10 villages getting 1 unit each at T sh. 9,000) 90,000 Ca. Drying units (10 villages at T sh. 3,000) 30,000 Pressure la.ps (10 villages receiving 30 lamps each @ T sh. 110) 33,000 Improved local boats, with gear (10 villages X T sh. 10,000) 100,000 253.000 TANZANI KIGM RMI.AL DKELVMPKUT PROJECT Loand and Equity Ceitributian to [him Cooperitie Uni Project rer (T Sh) I II III IV v Ta*al A. Loans 1. Larries 1/ 158,200 429,400 361,600 610,200 1,559,400 P. Four-ubel drive vehicles 1/ 189,600 142,200 47,400 94,800 914,800 568,800 3. Spare parts 1/ 147,400 80,750 124,850 114,100 181,900 549,000 4. Office and ofice equipuent 200,000 200,000 5. Repair and maintence shop 80,000 80,000 6. Godoms g/ a. Bigaa 70,000 70,000 140,000 b. Kibondo (Section I) 150,000 550,000 700,000 c. Kibondo (Section II) - 150,000- 550,000 700.000 sub-total 737,000 1,001,150 601,65o 720,500 1,436,900 4,497,200 B. Equity Contribution 230,000 !a0,0o 450,000 90,000 - 1.200.000 Total 967,000 1,431,150 1,051,650 810,500 1,436,900 5,697,200 1/ The 7-ton lorries and lcng-wheel base four-wheel drive vehicles to be financed are estimated to cost T Sh 90,400 and T Sh 47,400 respectively. A provision for spare parts equal to 25% of the cost of vehicles has been made. 2/ The Kigoma godown is 5alculated at 2,500 ma in sie for a capacity of 2,000 tons. The Kibondo gcdown is calculated at 6,250 m1 in size for a capacity of 5,000 tons. Both assme unit costs of T Sh 280 per ton capacity. Further information is contained in Annex 14. _ ANEX J Table 7 TANZANIA KIGOMA RURAL DEVELOPHNT PROJECT Rural Training Centers (TSh 000's) 1 New Center for Kigoma District: Total Classrooms 30 Workshops 310 Stores and multipurpose area 39 Kitchen 33 Administration block 36 Dormitories 117 Toilets, showers, laundry 24 Staff housings Staff Houses (3) 240 Supporting staff houses (3) 79 319 TOTAL COST 908 2. Rehabilitation and Expansion of Existing Centers at Kibondo and Kasulu: Extensions (at each site): Dormitories 117 Toilets, showers, laundry 24 Classroom 60 Extension of mess 22 Recreation block 38 Water tank and tower 11 Teachers' houses 51 Tools, equipment and reference books 72 Rehabilitation of existing facilities 35 TOTAL, each 430 TOTAL, for two schools 860 TOTAL of 1 & 2 1,768 TAZNWITA KIGO(A RURAL DEVELOPMENT PROJECT Capital Cost of Training and Trials Center and Pilot Livestock Project (Tah) No. of Net Area Total Net Total Gross Coat Fer Construction Furnishing Cost Total Furnishing Construction Spaces Per Space Area Area. ft Cost Per Space Cost Total Cost (ft. T 7 ?(ft.Z (ft.2) Classroom (25 students) 1 500 500 550 60 33,000 5,000 5,000 38,000 Workshop and Laboratory 2 - 1,000 1,200 56 67,200 5,000 10,000 77,200 Storage 3 - - - - - - 83,000 Kitchen and Kitchen Store 1 - 400 440 70 30,800 - 700 31,500 Office 1 - 400 440 60 26,400 - 5,000 31,400 Boarding Accommodation (25 students) 25 60 1,500 2,000 60 120,000 800 20,000 140,000 Ablutions 1 3D6 171 342 70 23,940 - - 23,940 Staff Housing Grade A 2 - - - - - - 160,000 Grade B 4 _ _---0 Subtotal Construction 745,040 Equipment Tractor 33,000 Plow 4,000 Cultivator 5,000 Trailer 10,000 Vehicle 28,000 Fencing 9,000 Other Equipment 3.000 Subtotal Equipment 92,000 Livestock In calf Improved Boran Heifers (x 30) 30,000 Improved Boran Breeding Bulls (x 6 over project) 6.000 Subtotal Livestock 36,000 Total Capital 873,000 A)WX 11 Table 9 TANZAN IA EIOOMA RURAL DEVELOPMENT P9DJECT Technical Assistance (TSh 000's) Unit Costs T Sh 000 Year I Tear II Year III Year IV Year V Total A. Salaries and related rringe benefits Financial Controller 180 30 180 180 150 50 590 Operations Manager 180 30 180 180 150 50 590 Credit and Training Superv,ision 180 30 180 180 150 50 590 Land-use planner 180 30 180 45 45 45 345 Sr. Agr. Training Officer 45 45 45- 45 45 45 225 Trials and demonstrations Officer 45 45 45 45 45 45 225 Supporting staft (10) 3.6 20 40 4o 40 40 180 sub-total 230 850 715 625 325 2,745 B. Hou-sing. Veh _1_s and msc . Housing: ( 4) 150 300 300 600 (2) 80 80 80 160 (95) / 5 1/ 100 125 125 125 475 Vehicles ( 2) 47.4 95 95 Office and photo mosaies equip. 30 30 sub-total 505 480 125 125 125 1 ,360 C. Training.Costs Maintenance on trials training center 28 28 28 28 112 Maintenance of vehicles 25 25 25 25 100 Student maintenance (agr. field assist.) Z/ TShl5/day 100 100 100 100 400 Student maintenance- bookeeper 30 40 40 40 150 sub-total 183 193 193 193 762 D. General 205 205 205 205 205 1,025 … TOTALS 940 1,718 1,238 1,148 848 5,892 Footnotes ] Materials for houses of agricultural field assistants at village level; actual constrtiction assumed to be by self-help. 25 students per year for 250 days per year. 25 students per year for 100 days each. NOTE: Figures in parenthesis indicate members of units. ANNEX 11 Table 10 TANZANIA KIGOMA.RURAL DEVELOPMENT PROJECT Project Preparation and Evaluation Years - T. Sh. Unit Salary 'I II III IV V Program Coordinator 280,000 100,000 280,000 280,000 180,000 - Field Chief 280,000 20,000 280,000 260,000 - - Agriculturalist 250,000 250,c00 250,000 - - Engineer - 250,000 250,000 - - Consultants 20,000 50,000 400,000 300,000 50,000 - (month) Evaluation Specialist - 20,000 80,000 40,0 40,000 60,000 Transport - International 80,000 250,000 190,000 20,000 10,000 Field Operations 2 Four-Wheel Drive Vehicles 20,000 120,000 110,000 - HQ Car + Driver 1,000 30,000 30,000 - Consultant Subsistence 6,000 35,000 185,000 60,000 10,000 15.,0oo (month) Housing Allowance 50,000 20,000 240,000 140,000 - - (.year) HQ Office Rental, Secretarial ass't. 60,000 100,000 40,000 - and supplies Local Survey Costs 50,000 150,000 150,000 100,000 250,000 475,000 2,615,000 2,100,000 4IK,OOO 335,000 Total T. Sh. - 5,925,000 US$ 830.000 ANNEX 12 TANZANIA KIGOMA RURAL DEVELOPMENT PROJECT Estimated Schedule of IDA Disbursements 1/ End of Disbursed Cumulative IDA Fiscal Year Quarter During Quarter Amount Disbursed Balance of Credit --------------------(US$ 00) -------- 1974/75 1 - 10,000 2 - 10,000 3 - - 10,000 4 340 340 9,660 1975/76 1 600 940 9,060 2 600 1,540 8,460 3 600 2,140 7,860 4 430 2,570 7,430 1976/77 1 500 3,070 6,930 2 500 3,570 6,430 3 500 4,070 5,930 4 510 4,580 5,420 1977/78 1 510 5,090 4,910 2 510 5,600 4,400 3 500 6,100 3,900 4 510 6,610 3,390 1978/79 1 600 7,210 2,790 2 600 7,810 2,190 3 600 8,410 1,590 4 590 9,000 1,000 1979/80 1 280 9,280 720 2 280 9,560 440 3 280 9,840 160 4 160 10,000 1/ Allows for expected delays between expenditure and IDA disbursement. ANNEX 13 Page 1 TANZANIA KIGOMA RURAL DEVELOPMENT PROJECT Procedures for Selection of Villages A. Background 1.. In Kigoma region, the cycle of low productivity/poverty, low productivity has been particularly difficult to break. Agriculture, the mainstav of over 90% of the population' has been characterized by shifting cultivation generally at or near subsi,tence level. Because farms were small, scattered and periodically shifting (as fertility declined), providing im- proved techniques was difficult and expensive. Moreover, communications within the region have been poor, resulting in additional difficulties in the provision of extension, inputs or marketing services. Similarly, because the costs of providing social services to scattered and difficult to reach families is high, the level of those services provided to the rural population has remained low. 2. The national policy of bringing people together in ujamaa villages (cooperative arrangements stressing self-help and communal organization) is a means of breaking the low productivity-poverty cycle associated with traditional agriculture and simultaneously of improving the rural population's access to social services. By bringing people together into villages, Government hopes to be able to provide improved agricultural technology and inputs and social services at acceptable costs per capita. 3. Starting in July 1972, Government launched broad program "Operations" in three regions, including Kigoma, to provide resettlement of much of the rural population into ujamaa villages. By late 1973 "Operation Kigoma" had resulted in the establishment of 76 ujamaa villages into which about 24,000 families-roughly 25°% of the rural population-had been settled. Government assisted in transporting villagers' belongings and building materials. During the first season after settlement, the villagers built their own houses and cleared and planted single acre house plots. In subsequent seasons villagers cleared or will clear village block farms for crops such as maize, cotton, groundnuts and beans, on the basis of one acre per household per block farm. The village itself should graduate to a multi-purpose production cooperative, with corporate status, which would have usufructuary rights to the land and be the borrowinR-for-inputs vehicle. 4. Parallel discussions concerning the siting and timing of villages :o ie es tablished under Operation Kigoma are held throughout the regional administration and Tanzania' s only political party the Tanganyika African National Union (TANU). Decisions concerning which villages should be started and which should receive particular investments culminate in the annual budget Au4NEX 13 Page 2 process (Annex 4). In about February of each year the region's proposals are forwarded to the Treasury in Dar es Salaam for consideration in the budget for the fiscal year starting July. The decisions concerning particular sites (to be started in a particular year) take into account the resolve of villagers to form ujamaa villages and such specific technical factors as soils, available water, existinR settlement and accessibility. Unfortunately, sufficient atten- tion has not always been paid to these technical factors and poor, or at least sub-optimal, siting resulted in some cases. 3. Villaees started under Operation Kigoma vary in size from fewer than 100 families to almost 1,000; the average is between 300 and 500. At any oiven site some factors xœork toward favoring a larger village size while others favor a smaller size. Generally, factors favoring a larger village are economies of scale in water supply systems (especially pumped systems), education (the possibility of eventually having a full primary school), dis- pensarv and support personnel (generally the agricultural field assistant and village bookkeeper). Where dugwells are used for water there is little advantage to a larger village size; for pumped systems, larger numbers of families result in a lower per family cost (Annex 6). The Tanzanian system of having a separate teacher for each school standard (grade) favors having a larger village in order to have enough children of school age to justify a complete school of seven standards of 45 students each. A village of 350 families generally can suipport a full seven standard (i.e. 7 grades) primary school. The collection and credit activities of the regional Cooperative Union and TRDB are also facilitated by fewer larger villages rather than a larger number of smaller ones. 6. On the other hand, the limited management capability of village leadership and the need for village cohesiveness (joint debt-burden and some communal work) and possible transition difficulties, argue for smaller villages. These factors, plus the fact that the economies of scale noted above become much less significant when there are more than 400 families, imply that very large villages (say, above 500 families) are probably questionable in most cases. 7. Under the project the process of site selection would be strengthened by preparing, for each project village, whether already in existence or to be started, a Village Site Feasibility Report (VSFR). Technical officers at the district level, supported as necessary by regional officers, would be respons- ible for preparing the VSFR. A study team would generally consist of special- ists in agriculture, public works, land planning, water livestock and finance, supported as necessary by specialists in education, forestry and health. A pro-forma outline of a VSFR is attached as Appendix A. The preparation pro- cedures and content of the VSFR are descrlbed in the following paragraphs. ANNEX 13 Page 3 B. Selection Procedures for New Villages 8. Examination of Aerial Photographs. During the first few months of the project implementation period, aerial phtographs would be taken of those parts of the region where village development under the project might be expected (report para 4.02). A complete set of prints (of scale 1:12,500) would be kept in the regional Lands Office. In analyzing a proposed site the evaluation team would first examine the relevant aerial photographs. The land planner and his staff would take the lead in interpreting the photographs and preparing sketch maps as necessary. This examination would provide: (a) a reasonably accurate plotting of existing (if any) settlement, infrastructure, roads, trails and fields; (b) a preliminary indication of the location and extent of land with soils and topography suitable for sustained cultivation under the project; (c) in cases where a water supply system has not been developed, a preliminary indication of possible water sources and their likely ease or difficulty of development; and, (d) an indication of special features such as erosion hazards. This step would enable the team to: (i) determine, with little time investment, those proposed villages that are obviously unsuitable; and (ii) for those sites meriting furtheT. study, locate the site precisely and indicate matters that are likely to be significant, thereby facilitating the field examination. 9. Field Examination. A field visit by the study team would be an integral part of the preparation of the V.SFR for each project village (whether existing or new). The team would begin by assembling the data required to complete the questionnaire on the existing features (Appendix A attached). To assist in the analysis of the water supply component, the water engineer in those areas where a system of previously tried surface wells do not exist, would have arranged, where appropriate, for the prior sinking, at representative locations in the proposed site, of at least two dug wells, each to a depth of at least 5 meters. Also, to assist in analysis, a sketch map of the site would be prepared showing existing roads, settlement, water sources, buildings (godowns, etc.) and cultivated fields, and currently unclutivated areas having fertility adequate for development under the project. 10. Definition of Village Site or Sites. For any proposed village number of sites may be possible depending on terrain, existing settlement and structures, the location and amount of fertile land and the type and location of water supply. The team would not be restricted, at a particular location, to considering only the original proposal it was requested to ANNEX 13 Page i examine. Where several different sites could be readily identified, each would be sketched, its carrying capacity and costs calculated and its merits compared with alternative sites. Different sites would result in greater or lesser fertile land available for farming (depending on whether the house plots are located on the scarce good arable land or on the less-favored land). The sketch of each site would show the location of proposed house plots, water supply, roads and infrasturcture. 11. Carrying Capacity. In general, subject to general considerations listed in paras 5 and 6, the number of additional families that a particular site could support would depend on how much good arable land existed within easy walking distance of the village (usually within 3 km of the edge of the proposed settlement, depending on village layout and topography). It is expected that each family would generally cultivate 2.8 ha (7 acres) of good arable land, in addition to his 0.4 ha (1 acre) house plot, built up over several years. This would enable each family, at full development, in each year to operate 1.6 ha (4 acres) and have 1.2 ha (3 acres) in fallow; each 0.4 ha (1 acre) of a family's holdings (excluding house plot) would be cultivated 4 years out of 7 and lie fallow for the remainder (see Annex 3). 12. Cost Estimates. Cost estimates for investment costs and annual recurrent costs for each scheme would be prepared by the relevant functional specialists, taking into account recent experience and expected price changes. For each plan the water engineer would prepare a preliminary least-cost water system, taking into account the results of the dug-wells (par& 9 above). 13. Evaluation of Alternative Schemes. The several sites would have different costs and, perhaps, carrying capacities. The team would compare the alternatives on the basis of the present value of the initial investment and annual recurrent costs of each. For each alternative, the present value costs would be the sum of the original investment cost plus 7 times the annual recurring costs. 1/ In general the plan having the lowest present value cost per family at full development would be selected. 2/ 14. Recommendation. The team would make a signed recommendation cover- ing: (a) whether any further development was warranted at the proposed site; (b) which site is preferable (generally the one having lowest present value cost per family); and, (c) the maximum sustainable size of the village in terms of supportable number of families. I/ This is a short-cut approximation for a present-value calculation using a 15% discount factor. It assumes that public investments in Kigoma have an opportunity cost of 15%. 2/ The criterion of lowest present value cost per family may not give the theoretically optimal solution in all cases. It does not, for example, treat explicitly the tradeoff between reducing the cost of water supply by locating the village closer to the water source, at the expense of villagers having to walk farther to block farms. However, as long as the distance to block farms is reasonable (say within 3 km) this criterion should give reasonably good results plus having the important advantage of simplicity. ANNEX 13 Page 5 C. Selection Procedures for Existing Villages 15. The purposes in carrying out a VSFR exercise for already-existing villages for which investments or credits are proposed under the project are (a) to ensure that these villages would be viable for existing and proposed populations; and (b) to permit a detailed examination of the alternative ways of carrying out the proposed investments (e.g., water supply). 16. The same form of VSFR could be used for both new and existing villages. In the pro-forma VSFR (Appendix-A attached), special care would be required in the calculation of "carrying capacity" under C-2, to ensure that existing families would have sufficient land at full development of the village block farms (at 2.8 ha per family). AaNNEX 13 APPENDIX A Page 1 TANZANIA KIGOMA RURAL DEVELOPMENT PROJECT PRO FOPMA VILLAGE SITE FEASIBILITY REPORT (VSFR) General '4ame of village District I-ocation Distance by road from district headquarters A. Examination of Aerial Photographs 1. Code number of relevant aerial photograph 2. Sketch map of existing features. This should be traced from the aerial photograph and should outline all relevant existing features including location of existing cultivated areas, areas of good fertility (not currently cultivated), and, if any, schools, dispensaries, settlement, roads, trails, godowns or other important features. The sketch map becomes "Exhibit 1" to the VSFR. 3. (Preliminary) estimate of arable land (in hectares and acres) not presently in cultivation and suitable for development under the project. Area within 3 km of edge of proposed village Area within 5 km of edge of proposed village 4. Description of existing or possible water sources: type, location. 5. Description of any special features or problems. 6. 1'reliminary conclusion (basedi on examination of aerial photographs). I[ere indicate whether the proposed site is suitable for further study and field visit. If the judgement is that the site is obviously unsuitable (e.g. because of insufficient arable land) state reasons. ANNEX 13 APPENDIX A Page 2 B. Detailed Description of Existing Features at Site Proposed for Village Investments under IDA-Supported Project (to be filled in at field visit stage) l. Existing Settlement (if any) a. Number of families b. Principal crops currently grown c. Numbers of livestock (by types) d. Existing cooperative organization (if any): - most recent balance sheet (or equivalent) - bank balances or other savings - value of transactions by year for 5 years - principal activities - names of officers, directors C. Names of village leaders, length of service. 2. Existing Infrastructure (if any) a. Schools: - number of classrooms - number of students by standard - number of teachers and teachers' houses - location in relation to existing settlement - qualitative assessment of condition of buildings, furniture and textbooks - adult education: numbers of classes, subjects, students; average attendance b. Water Supply: - nature of existing source (whether stream, well, spring, etc.) - distance in meters from center of settlement - difference in elevation from source to principal area of village and to site of possible storage tank - results of previous tests of water quality (if any) - results of present tests of water quality - results from dugwells (where applicable): - depth at which water struck - assessment of quantity, preferably in dry season (draw-down and recovery time) - assessment of quality ANNEX 13 :APPENDIX A Page 3 c. Health Facilities: - existing structure, equipment and stocks - qualitative assessment of condition of establishment - existing staff - length of time established - distance in meters from present settlement - during past year: - number of patients - principal ailments - numbers of lectures given - other activities d. Marketing Facilities: - location and nature of nearest buying point for crops - distance from buying point in meters (or kilometers) - location and nature of nearest source of agricultural inputs - distance from input source in km - location and nature of nearest source of consumer goods - distance in bIn from nearest consumer goods source e. Roads: - distance to nearest main road - condition of access road; special features - whether Passable during rainy season f. Surrounding Area: - population within 10 km - population within 25 km 3. Agriculture a. Existing situation: - topography - soil type - cropping sequence - area under crops (if any), by crop during each of last three seasons - crop production (tons) by crop during each of last three seasons (where available) - quantity marketed during last three crop seasons (where available) ANNEX 13 APPENDIX A Page 4 - quantity of fertilizers, insecticides and improved seed used in each of last three years - numbers of cattle, if any, with ownership pattern - disease or pest problems - area of forest; principal species b. Potential: - area within 3 km of the proposed village having soils and topography suitable for agricultural development under the project. C. Definition of Village Site or Sites 1. Village Plan or Plans At a particular location, often several alternative sites can be identified. Each will have a different development cost, annual recurrent costs, and perhaps, carrying capacity, dependinp, on length of roads, water-supply configurations, location of household sites and block farms, etc. A separate sketch plan will be prepared for each alternative site identified. Each will be attached to the VSFR as Exhibit 2 (Scheme A, Scheme B, etc.). 2. Calculation of "Carrying Capacity" (for each identified site) (1) Total area of suitably fertile land convenient to proposed or existing village site (say 3 km) ha ha (2) Less: area required for new godowns, roads, schools, house-plots, etc. (3) Less area required to allow 2.8 ha (7 acres) for each existing family (4) Net new accessible cultivable area (5) Area per family required for block farm (generally about-' 2.8 ha or 7 acres) ANNEX 13 APPENDIX A Page 5 Maximum number of additional families that site can sustain under this particular site (4) + (5) families. Note: In the case of some existing villages, the figure for (4) mav be zero (indicating no further settlement is feasible) or negative (indicating that the site already has more families than it can carry an(d thought should be given to encouraging some families to move elsewhere). In either case further investment in the site would not be warranted. 3. Cost Estimates (for each identified site) (a) Investment Costs Land clearing (number of hectares ) Access roads (if any) Schools: Classrooms; number Teachers' houses Tools Furniture Textbooks Rehabilitation Codown House for field assistant Dispensary or other health facility Water supply (according to type of system recommended, taking into account lift and distance) Other required public investnent (e.g. cattle dips, erosion control, etc.) Total cost. (b) Recurrent Costs (average, per year) T Sh Access road maintenance Education Teachers' salaries Textbooks Tools replacement Maintenance on buildings Other (specify) Dispensary or other health facility Salaries Medicines and other supplies Building maintenance ANNIEX 13 APPENDIX A Page 6 Water supply Salaries Fuel Maintenance Other Other Staff - Salaries and other expenses Agricultural field officer or assistant Cooperative and ujamaa assistant Village bookkeeper Other (specify) Total, Recurrent Costs D. Evaluation of Alternative Schemes S C H E M E S A B D E Etc. l. Estimated investment cost 2. Estimated annual recurrent cost 3. Present value of recurrent cost (five times number 2) 1/ 4. Total present value of costs (1 + 3) 5. Number of families expected to settle during first two years 6. Present value of costs per family (4 * 5) 7. Numer of families at full development 8. Present value of costs per family at full development (4 + 7) E. Recommendation 1. Does the team recommend the site as suitable for any further development? Yes No ~. Recommended scheme (from,D above): 3. Maximum additional number of families site can sustain: 4. Signed: 1/ Implicitly assumes opportunity cost of public funds of 15Z per year. ANNEX 14 Page 1 TANZANIA KIGOMA RURAL DEVELOPMENT PROJECT Trading and Lending Procedures Introduction 1. Under the project, the Kigoma Cooperative Union would be responsible for provision of agricultural inputs and the marketing of surplus produce generated by project villages. 1/ To complement these trading activities, it would also serve as a lending channel to project villages, supplying credit for the provision of agricultural inputs and cash advances for the purchase of marketable surplus. In addition, it would act as the onlending agent to provide credit for the economic infrastructure of its member villages. 2. The proposed increase of the Union's operations, relative to exist- ing activities, is enormous. At present, the Union provides no inputs and handles negligible amounts of surplus outputs. It owns no storage or office space and has only two lorries. Its staff is small and inexperienced. In contrast, under the project total sales of marketable surplus are projected to reach T Sh 27 million by Year 5 (PY5) and T Sh 40 million by Year 12 (PY12) and even larger amounts of credit -- for seasonal inputs and crop purchases-- would also be required (see Table 1). Successful implementation of such operations would require significant strengthening and expansion of the Union. By the final year of the project (PYS) it would have to own and operate dis- trict and regional godowns, valued at about T Sh 1.5 million, 17 lorries and 12 four-wheel drive vehicles and repair and maintenance facilities to service them, together estimated at about T Sh 2.1 million, and new offices, expected to cost some T Sh 200,000. Debt repayment on this capital investment in the same year would amount to approximately T Sh 1 million. In addition, staff numbers would increase to 81, including new posts of operations manager, financial controller, and credit and training supervisor and personnel pro- vided for the motor vehicle fleet. 3. At the level of the primary society, the volume of inputs handled would increase to T Sh 114,000 in Year 5 and T Sh 118,000 in Year 8, and the sale of marketable surplus to T Sh 294,000 in Year 5 and T Sh 332,000 in Year 8 of village development (see Table 10). Credit needs would expand in a similar manner. It is projected that a godown, valued at T Sh 36,800 would be constructed to meet storage requirements of each creditworthy village. A maize mill and maize sheller, estimated at T Sh 22,800 and T Sh 3,750 respec- tively, would also be provided to each creditworthy village. In addition, a bookeeper and godown attendant, as well as seasonal labor, would be employed during the harvest season to oversee the primary society's transactions. 1/ All villages eligible for credit under the project would already have been registered as multipurpose primary cooperative societies. ANNEX 14 Page 2 Chronology 4. Procedures for the provision of credit for agricultural inputs would begin in December, almost one year before the inputs would be applied in the field. The Tanzanian Rural Development Bank (TRDB) would solicit input requirements from the regions. Under the project, central responsibility for the appraisal of primary societies would fall to the Union, whose society cre4it officers (SCO) would each be responsible for the appraisal and super- vision of about 12 societies and would be distributed in Union branch offices throughout the three districts. In close cooperation with representatives of KILIMO, the Ujamaa and Cooperative Development Division, and TRDB the Union would evaluate the credit needs of its member primary societies and submit a request by February, when TRDB would enter into formal loan agree- ment with the Union for the amount of inputs indicated. In turn TRDB would place orders with appropriate suppliers in March for all purchases for which it is charged procurement, following preliminary tenders the previous December. In practice, TRDB would combine its procurement needs for inputs under the project with its procurement of similar items under other IDA projects. The agricultural inputs to be provided would include fertilizers (DAP, CAN, and SSP), pesticides (DDT, Endosulfan), and ULV sprayers and accompanying batteries (see Table 2). The suppliers would then provide for delivery of inputs to railhead Kigoma from August to September, according to the terms of the contract signed with TRDB. A Union representative would be at railhead to verify the receipt of goods. The Union would move the inputs to its regional godown in Kigoma town, where they would be made into village consignments and subsequently be delivered by Union transport to the respective societies. Efforts would be made to coordinate the delivery or inputs with the collection of outputs so as to minimize transport costs. 5. Upon receipt of the agricultural inputs, a representative of the primary society, normally the bookkeeper, would sign a document, which along with the invoice and other documentation would represent a contractual loan agreement between the society and the Union. The cost of the inputs to the society, and ultimately to the individual farmer, would include the per unit costs of insurance, handling, transport, interest charges and Union margin estimated at the beginning of the season (see Table 4). 2/ The appro- priate amounts of inputs, as previously calculated for each crop by the village agricultural field assistant (APA) and agreed to by the village development committee (or agriculture subcommittee), would then be distributed to each farmer involved and the full cost of the inputs provided would be debited to the farmer's account, as maintained by the society's bookkeeper. Finally, the farmer would apply the inputs, following the advice of the A"A. 1/ See Chart I. 2/ Only interest charges on the Union's buying price of the inputs, not its selling price, would be attributed to the society; any financing costs represented by the margin between buying and selling prices would be borne by the Union. ANNEX 14 Page 3 6. In general, starting in May, approximately one month before the principal purchase period, the village committee and AFA would estimate the yields by crop by month in order to submit a request through the society's bookkeeper to the Union for regular cash advances for the purchase of crops from farmers. The Union, drawing upon the same appraisal capabilities used in evaluating agricultural input needs, would evaluate the estimates and subsequently request a line of credit, at 6-1/2% interest, from the National Bank of Commerce (NBC) 1/ for the aggregate amount, together with other needs for working capital such as produce inventories. The Union would assume the full costs of this interest charge, unless it could be demonstrated that member societies had purposely withheld remaining cash balances beyond the end of the calendar year, at which point such societies would be charged all additional interest. 7. The appropriate amount of cash would be advanced to the respective societies about every three wieeks during the approximately six months of harvest, at the time of the regular visit of the society credit officer (SCO). The SCO would use the occasion to review the society's books and inspect the godown. Cash would be disbursed for the following three weeks only after the SCO had verified that all was in order and that the amount requested was appropriate. The cash distributed by the SCOs would be transported in secured four-wheel drive vehicles and would be kept in small safes in the individual villages. The safe could be opened only with two keys, one held by the bookkeeper and one by the village secretary, both of whom would have to be present to open the safe at the beginning of the day and to close it at the end. The cash contained in the safe would be counted on both occasions and checked against the society's books. The operational costs incurred in providing the services of the SCO (as high as T Sh 1,000 per visit) would be covered by the UJnion's margin. 8. At time of harvest, the farmer would deliver his marketable pro- duce to the society godown, where it would be graded, weighed and bagged, under the supervision of the godown attendant and bookkeeper. He would be credited the amount of his produce at the Union's buying price, as recorded by the bookkeeper. A portion of the value of the agricultural inputs provided him at the beginning of the season, which included the costs of Union service and interest charges, would be deducted and he would be paid on the spot the balance owed him from the cash advanced by the Union. It is suggested that a single system of repayment of input credit be introduced (e.g., that 50% of the value of the farmer's produce be deducted from each delivery to repay input credit until the full debt has been repaid, after which further deliveries would be paid the full value) to assure full repayment of the society's debt for seasonal inputs. Once delivered to the society godown, the produce would become the property of the society. The society would have produce in its godown whose value would be sufficient to repay the TRDB credit, for which 1/ Under Tanzanian practice TRDB lends for farm inputs and NBC for crop purchases, thereby involving two institutions in the provision of short- term credit. AMUE 14 Page 4 the farmers would already have been assessed, and to repay the NBC cash advance used in the purchase of the produce. The society would be responsible for assuring security of storage, such that all produce for which the farmer was paid remained intact in the village godown for collection by Union lorries. 9. The Union would be responsible for the collection of the agricultural produce from the society's godown and delivery to the Union's or parastatals' godowns. Ybre than four weeks' produce would not normally be allowed to accumulate in the village godown, and in principle, collections would be made several times a week at full village development. All produce would normally be collected by the onset of the rains in mid-November. Although the society would not receive a storage fee, the mounting interest charges assumed by the Union on the NBC cash advance should be an incentive for the Union to collect the produce promptly from the primary societies. In addition, part of the contractual agreement signed when the inputs were supplied would be an obligation by the Union to collect all marketable surplus before the rainy season. 10. At time of collection, a receipt for produce collected would be given to the society's representative, and repayment of the input credit and cash advance would be considered completed for an amount equivalent to that of produce collected. The proportion of each kind of credit to be represented by the produce would again be calculated by a simple formula requiring 50% initial deductions to repay TRDB's input credit. Cash advances accordingly would only have been supplied for the net value of crops plrchased (i.e., total produce anticipated during the three-week period valued at producer prices minus a specified percentage to repay input credits). Such a system-of cash advances on a net rather than a gross basis (where repayment of input debt would be made separately and afterwards) would (a) introduce less total cash and, therefore, risk into the system, (b) provide for parallel rather than subsequent repayment of TRDB's input credit, and (c) involve fewer physical '-ransactions. Therefore, the sub-borrower's responsibility would end for .X> equivalent of input credit and cash advance. It would then remain to the'<-(on to repay its debts to TRDB and NBC. A primary society levy, which has be>-. applied traditionally to cover the societies' expenses (see Table 4), woula 'e imposed on the produce collected during the month. 11. The produce would be delivered variously to either the Union's district or regional godown, the godown of the National Milling Corporation (NMC), or one of the ginneries of the Tanzanian Cotton Authority (TCA). All cotton, representing about 602 of the value of total marketable sur- plus, would be transmitted directly to TCA custody. All beans, maize, and groundnuts collected during the 15-week peak of harvest in Kibondo district would be delivered to the Union's godown that is to be constructed in Kibondo. The remaining beans and maize, and the groundnuts from Kasulu and Kigoma districts would be delivered respectively to the NMC's and Union's godowns in Kigoma town. Inventories in the Union's Kigoma godown would be expected to turn over at least three times during the harvest season. Once the rainy season began and the Union's lorries were no longer required to collect oroduce from the primary societies, the produce stored in the Union's Kibondo godowns would be transferred to NMC's godown in Kigoma town. ANNEX 14 Page 5 12. The parastatal would be invoiced the value of any direct delivery, at the Union's selling price which would include transport, storage, financial and handling costs, as well as the Union's and society's levies (see Table 5). Panment would be made directly to the Union's NBC account, where deduction for NBC's cash advances would take place automatically. Likewise, the TRDB loan would be repaid as the Union was paid by the parastatals for the produce delivered with an automatic transfer of accounts being arranged by NBC and TRDB. The groundnuts retained in Union godowns eventually would be sold either to GAPEX (The General Agricultural Products Export Corporation) and delivered to railhead or marketed by the Union in the region, following authorization from GAPEX. It is roughly estimated that 25% of groundnuts handled by the Union would be marketed within the region (see Table 3). Transport and Storage 13. Critical to the Union's trading activities is the transport and storage system, currently non-existent in any form, which must be built up under the project. The efficient handling of marketable surplus and agricul- tural inputs is essential. Several considerations weigh heavily in the design of an optimum system, with the foremost consideration the balance of storage to transport capacity where, in the case of the project, the extremes would range from no village godowns and a large fleet of vehicles for daily collec- tions of produce to large village godowns with small transport capacity for delivery to market. In striking a satisfactory combination, several factors merit particular attention and require critical assumptions: (a) costs, both capital and operating, particularly considering the life of the capital good and vehicle fuel and maintenance operating costs, (b) benefits, parti- cularly relating to off-season use, and (c) level of operations, particularly the godown's rate of turnover and the lorry's rate of use. The capital investment can provide various levels of service, depending on the assumptions made on, e.g., the percentage of downtime and the hours in use for vehicles (factors, in turn, affected by double shifting or more efficient handling) or the turnover of inventory for godowns. The design of a transport/storage system requires the critical review of these and other factors, including different combinations of both godowns and lorries. Hence, the system pro- posed within the project is merely indicative and makes no pretense to be an optimum solution for a complex situation still some years into the future. 14. One godown would be provided for each credit-worthy village under the project financed by a TRDB loan channeled through the Union, to be constructed in the second year of village development. It would be valued at approximately T Sh 36,800, including T Sh 3,500 for a small safe and weighing scale. The capacity has been calculated to accommodate one month's market- able surplus in a village's fifth year of development, or about 115 tons. Total turnover in PY 5 is estimated to be almost 6,000 bags for a typical 350 family village. The godown would include a storage building (144 m3) '74t¢ t- corrugated iron siding and concrete floor, adjoining concrete platform, security fencing and small office. ANNEX 14 Page 6 15. The Union is projected to have two godowns, one in Kibondo district with a capacity of about 5,000 tons (6,250 m3) and costing approximately T Sh 1,400,000 and one in Kigoma town of about 500 tons (650 m3) and costing about T Sh 140,000. Total turnover for each is projected in PY 9 to be roughly 7,500 and 3,000 tons respectively. The district godown would be constructed in two equal sections, during PY 1 and 2 and PY 4 and 5, and the regional godown would be constructed during PY 1 and 2. In projecting these godown requirements, it is assumed that TCA would construct its own godowns to accommodate all cotton delivered directly from the village prior to ginning, and that the 10,000 ton NKC godown currently under construction would accom- modate all maize and beans delivered by the Union to Kigoma town. Moreover, it is assumed that all marketable surplus of beans, maize and groundnuts produced in Kibondo during peak season would be stored in the Union district godown until the beginning of the rainy season. 16. The volume of goods to be stored and transported under the project is estimated to involve considerable handling costs. Handling costs of market- able surplus are roughly estimated to be T Sh 8.70 per ton at the society level and T Sh 5.80 at the Union level. Because the volume of agricultural inputs would be as little as 10% that of marketed outputs, and because many of the inputs would hopefully be unloaded as outputs were loaded, the handling costs of inputs would be relatively insignificant. 17. Handling of marketable surplus, at the society level, involves the weighing, bagging and stacking of produce when received, and weighing and loading of bags when collected by Union lorries. In addition to the bookkeeper hired on a regular basis, a godown attendant would be hired full-time for five months to supervise the loading, unloading, weighing, counting and bagging of agricultural inputs and produce to and from the village godown. The attendant would be assisted by seasonal labor as necessary (see Table 10). 18. At both the Union's district and regional godowns, handling would involve the unloading, weighing, stacking and loading of agricultural inputs. In addition to the permanent storekeeper who would weigh or count the material handled for each godown, wage labor would be required. 19. The requirements for transport are expected to grow substantially under the project. The number of lorry loads (seven-ton lorries) of market- able surplus is expected to grow to 5,600 in project PY 5 and to 7,200 in PY 12. Lorry requirements during the harvest season are estimated to be 40 vehicles in PY 5 and 57 vehicles in PY 12 (see Table 7a). 20. Transport is expected to represent the most significant source of income and of expenditure to the Union. The Union's fleet of lorries is projected to grow to 17 by PY 5 and 26 by PY 12, and by PY 8 to require a maintenance and repair shop manned by 3 mechanics and a total transport staff of 46. Total running costs are estimated to be about T Sh .68 million ANNEX 1 4 Page 7 in PY 5 and T Sh 1.04 million in PY 12. Total investment in the fleet of lorries, including the repair aka maintenance shop, spares,-and replacements every five.years,'is e'xpeeted.to'be T Sh 7.3-million over the t2-year period. -Incoime derlvd from the Union lorries is txpedt6d to equal T 9h. 1.8 miiin in PY.5 and..T Sh 2.? millo;n in PY 12 whieieud to,traffspolt agricultural inputs and mAtketai,le surplus, baed on a rate ofg T-Sh-Lh35 per'revenue ton/ mile revenue (see Table 7b). The transport reveitue would be recovered in the price the Union swells the produce it handles to the parastatals. Income derived from the hire of Union Morries to the vGoverament- during: the off-season is ?roJected to be apVr6!xbwte1ytT Sh-5OO00,O in PY,S andT- ShtOOO6eO in c: PY 12, ' 21. Despite the relatively important part that the Union's fleet of lorries would play in the overallioperations of,'the,Union,-substantiaa numbers af' additional lorries would still have to be hired during',pba h4rvest season. I1nPY S end PY'12 respectively 28eand-31 vehicles would be4hi1in*&durivg peAk havest season by the Union* either from Gowvernment or' prit#UC sources. Lending Procedures 22. Four kinds of credit are to be provided under the project: (a) credit for the provision of seasonal inputs to the primary-societies( (b) -redit to the Union for working capital-including cash-advances'to sotieties fot the purchese of marketable surplus; (c} credit for investment in village economic infrastructureg (d) credit for investment in Untoun-.s,building and equipment (see ,able 9). The credit might be viewed alternatS&ely #s that for on-lending to member village4 and as that for the Union's-own ac og-t. Both NBC and TRDB would be sources of credit, the former for cash ad--nces for crop purchases and the latter for the remainder, in accordance %42-La. national policy. It is proposed; however, that TRDB would be responsible f9r supervising the total Lending program to and through the Unaon in order to assure integrated, eomprehensLve and continuous evaluation and oontrol of. the-Union's financial statusa, 23. Credit arrangements under the project would rely heavily on both the Union and TRDB. The Union would assume primary responsibility for all trading and lending operations involving project villages. The reasons that this central role be accorded the Union are several: (1) With three marketing boards, two lending institutions, and approximately 100 widely dispersed and relatively inaccessible primary societies involved, it is preferable tv conce,trate all contacts at the village level in one body (The pWion), to simplify logistics an4dminimize costs; (2) To undertake any of its assigned functions effectively, the t1nion must be significaptly strengthened, including a network of personnel operating-regularly at %he village level. and the infrastructure to service member societies. This can onlv be justified economically if the Union assumes respon- sibility for the full range of credit and trading service-s to be provided; A?ls'EX 14 Page 8 (3) one institution intimately involved witlh the project Imst assume full responsibility for all credit provided to project villages and full responsibility for the financial viability of such villages. Because NBC would provide approximately 75Z of working capital requirements from non-project sources, and the volume of project-related TRDB credit is thus relatively insignificant, only the Union is satisfactorily placed to undertake such comprehensive responsibilities; and (4) The proposed system uses existing institutions and procedures and, in particular, supports the Government's efforts to build up the country's cooperative structurea. 24. Thus all loans to villages for crop inputs, cash advances, aiid economic infrastructure would be channeled through the Union. The Union would develop its own appraisal capability and would undertake independcent evaluation of the creditworthiness of member societies. The Uhion would assume responsibility, including full repayment obligations,,for &l' dGel;s it incurred. It would pass on to its member societies interest chargas on seasonal inputs and economic infrastructure but would pay itself interest charges on cash advances for crop purchases (recoveriug credit costs of cash advances from its margin on trading those items). The Union would assume full responsibility for loan recovery from primary societies, and the HUcieties idl turn would assume full responsibility for loans to their member farmers. In both cases a fund to provide for bad debt, and ultimately supply working capital requirements, should be established from income derived from trading activities. 25. TRDB, for its part, would have no means for direct appraisal of individual primary societies, but rather would supervise and control closely the credit appraisal and onlending activities of the Union. TRDB would carefully monitor the Union's overall operations, including cash advances, inventories and other borrowings, with which TRDB would not be directly involved but which would be critical to the Union's ability to make timely repayments to TRDB. Total cash advances to primary societies through the Union for crop purchase would represent several times the value of total credit for agricultural inputs. Nonetheless, the ability to withhold new loans for inputs at the village level until outstanding crop loans had been fully repaid, and at the Union level to withhold new loans for capital investment or call prematurely old loans if the Union borrowed indiscriminately, would allow TRDB to enforce proper credit discipline in the region. Such a system would: (a) Force a single financial itisitution (TRDB), actiug indirectly as IDA's agent, to monitor continually the Union's total financial activities (involving crop loans, cash advances and inventories); ANNEX 14 Page 9 (b) Force credit discipline by requiring TRDB in effect to approve the crop loan requests of individual societies to the Union; and (c) Introduce a sense of financial accountability at all levels of borrowing. 26. Different appraisal procedures would be required for each type of credit. Credit appraisal for agricultural inputs would be the primary respon- sibility of the Union, under the overall supervision of the regional staff of TRDB. Individual primary societies would be requested to submit preliminary requests in about December for seasonal inputs to be provided almost one year later. Each village, under the primary responsibility of its development committee and with the active assistance of its agricultural field assistant (AFA), would assess its needs and forward an application to the Union. The Union's society credit officers (SCOs) would then appraise the individual requests. 27. The basic factors requiring review in each appraisal would be (a) technical; (b) human; and (c) financial. Specific preconditions for the first input credit would be the installation of a village bookkeeper and agricultural field assistant and a satisfactory agricultural performance during the previous season. In addition, the following types of questions should be posed: (a) Do the crop projections seem reasonable in light of past experience either in the village itself or under similar ecological conditions elsehwere? What are the projected yields by acre and by unit of input? Are they sufficient to induce the individual farmer to use the inputs, and to use them efficiently, and to repay with a satisfactory margin on initial investment?; (b) Are the Chairman and his committee honest men of integrity, capable of strong and good leadership and receiving full village support? Are the AFA and bookkeeper adequately trained and competent to perform the services with which they have been charged? Are the villagers fully satisfied with the past performance and level of services of the society? How stable is the village, and what is the turnover of population? What percentage of past production has been channelled through the Union, and what percentage appears to be consumed locally or directed through non- authorized channels?; and (c) Have all past debts been paid off, and have they been paid off promptly? What is the current level of the society's savings and what have been the uses of past savings? What ANNEX 14 Page 10 are the society's assets and do they appear to be satisfac- torily operated and maintained? What has been the overall income of the entire village in past years?. 28. Finally, on the basis of the appraisal by the SCO in close coopera- tion with the appropriate Government authorities, and under the supervision of TRDB's regional staff, the Union would fix the credit needs of its member societies and subsequently submit a credit application to TRDB. Credit appraisal for subsequent seasons would be based most heavily on the agricul- tural performance and debt repayment of the previous season. 29. The advancement of cash for village crop purchase would require two stages of credit appraisal. First, a month before crop purchase, an assessment would be made of the total credit requirements for the purchase of crops during the full harvest season. Second, during the five months of crop purchase and collection, independent and regular evaluations of each society's position would be made each time additional cash was advanced to the society. 30. The appraisal of credit needs for cash advances would constitute an updating of the appraisal for input credit conducted earlier. NBC, or possibly TRDB on behalf of NBC, in close cooperation with KILIMO would survey the standing crops of a sample of villages to estimate the extent to which actual yields would meet the projections calculated at the time of appraisal of input requirements. These estimates would form the basis for assessing the total credit required for cash advances during the season. TRDB would be informed and asked to evaluate the working capital arrangements, or other forms of indebtedness, to be undertaken by the Union. Following agreement of both lending institutions, and in consultation with the Government autho- rities concerned, an appropriate line of credit would be opened. 31. To advance cash during the harvest season, the SCO would begin, before harvest time, by arranging with each member society a schedule of disbursements. He would then visit each society every three weeks or so to supply the cash required for crop purchases. Between each visit the society bookkeeper would be expected to maintain weekly cash return forms, indicating, inter alia, crop finance received and use of crop finance by crop, grade and quantity. The SCO would review these returns, the society's books, and cash on hand during his visit. In addition, he would inspect the quality and quantity of produce contained in the society's godowns. He would pay particular attention to checking that the society"s cash on hand and inventory of crops purchased minus collections by Union lorries equalled total cash advances. He would also periodically check, inter alia, the accuracy of the society's scale and the grading of produce. Only after the SCO had verified that the previous cash advance had been properly disbursed, that it corresponded with the produce in the godowns or collected by the Union lorries, and that all was properly reflected in the society's books, would ANNEX 14 Page t 1 the amount of the next cash advance be considered. Agreement would theni be reached with the society authorities on whether the scheduled disbursement was appropriate or whether modifications were required; thereafter, the cash would be advanced accordingly. If irregularities were found in any of the society's financial statements, rather than withold credit for crop purchase, which in a monopolistic state-controlled marketing system would be politically; impracticable and unpalatable, the Union or Government would temporarily replace the village bookkeeper with one of its own staff. 32. The SCOB would be required to submit weekly reports summarizing the financial position of villages visited and total cash advances during the week. On the basis of their reports, the Union's financial controllers would prepare weekly aggregate statements on its working capital position, which would include an indication of its total short-term borrowings, cash advances to societies, inventories, and deliveries to and receipts from parastatals. This statement would serve to guide Union's management on its on-lending operations and to inform TRDB and NBC of the Union's weekly cash flow. Specifically, it would provide TRDB the tool required to undertake the credit supervisory and control functions to which it would be assigned. It would enable TRDB to monitor several critical aspects of the Union's operations by: (a) examining the difference between short-term borrowings and cash advances, it could determine whether the Union was accumulating excessive cash; (b) comparing the Union's inventories and deliveries to para- statals with cash advances to societies, it could assess the efficiency of the Union's transport system and the timing of collections; and (c) comparing deliveries to and receipts from parastatals, it could determine whether the Union is being repaid on schedule in order to retire its debt. Thus, the weekly statements would provide the indicators necessary to assure that rapidly increasing Itnion indebtedness was matched by comparable increases in Union inventories and/or accounts receivables from parastatals. In addition to the statement, TRDB would monitor a sample of villages, by re- viewing the SCO's weekly reports and by undertaking periodic field visits. 33. Credit for the economic infrastructure of member primary societies ,;ould also be channeleld through the Union. The credit appraisal techniques would be similar to those used for agricultural inputs and cash advances. Primary responsibility would again fall to the Union's SCOs because the TRDB branch office would not have sufficient staff to appraise the several hundred credit applications; but because the investments are relatively large and long-term, TRDB's regional staff would play a more active and direct role. Terms of the loans would be as follows: ANNEX 14 Page 12 Interest Repayment Item Unit Cost (T Sh) Charge Period Mize sheller 3,750 7.5% 4 years Maize mills 22,800 7.5Z 5 years Fishing gear various 7.5% 2 years Village godowns 36,800 7.5% 15 years (including 2 years grace period) (See Table 9). 34. It is expected that the economic lives of the sheller, mills, fishing gear, and godowns would be 6, 8, 2 and 15 years respectively, and that the shellers and mills would be replaced at the end of their economic lives, per- haps by further borrowings. The project would finance only the initial invest- ments through PY 5. 35. The Union would receive credit for capital investment on its own account, in addition to credit for on-lending to member societies. Total capital investment would amount to about T Sh 1.4 million in PY 5. Any credit appraisal would begin by assessing the general financial position of the Union, both present and historical, including, inter alia, liquidity, debt-equity ratio, total revenue and expenditure and earnings. In addition, the integrity, experience, and competence of management would be reviewed. In particular, throughput and margins that might be assumed for the period of the loan would require examination. Revenue to service the indebtedness created by Union investment would derive primarily from the margins obtained on the handling of agricultural inputs and produce. Throughput would be a function of projected area under cultivation the percentage of the region's marketable surplus handled by the Union and projected yields, by crop. It would, in effect, represent the aggregate of estimates made in appraising input credit for individual villages (less subsistence requirements and any marketing by primary societies outside official channels). The margins would be estimated on the basis of existing margins. An assessment would first be made of the revenue accruing to the Union from its margins in the past and of the adequacy of the revenue to meet all operating expenses and to service all debt obliga- tions. Where revenue appeared inadequate, modifications could be introduced in light of the average margins assumed by cooperative unions elsewhere in Tanzania or proposed increases under consideration by the Ujamaa and Coope- rative Development Department. Following thorough appraisal, credit would then be provided on the following basis: ANNEX 14 Page 13 Items Interest Charges Repayment Period Vehicles 8.5% 36 months Buildings 7.5% 10 years (1 year grace period) (see Table 8). 36. Vehicles would be considered to have three year economic lives, after which they would be replaced under regular TRDB credit arrangements. Only buildings and additional vehicles acquired through PY 5 would be financed under the project. 37. The marketable surplus anticipated and the procedures proposed under the project would require the delivery of large amounts of cash to the villages. The average village, for example, in its fifth year of development would receive about T Sh 35,000 (US$5,000) in cash every 3 weeks. Attempts must be made during project implementation, as experience is gained, to reduce this figure, and thereby the attendant risk. The periodic visits to villages by the SCO's could perhaps be increased in order to reduce the cash required per visit. Alter- natively, some sort of savings scheme could be introduced to tap the accumula- ting cash of individual farmers, resulting from sale of marketable surplus. Or nossibly a mobile banking system (e.g., weekly visits to villages) could be developed, replacing with notes and cheques the large amounts of cash r-errently proposed to pay individual farmers. 38. Attention must also be paid to consumption as well as production needs during project implementation. As disposable income increases under the project, the spending patterns and needs of villagers should be identified and the appropriate goods and services should be made available. Othetwise, v'llagers' incentive to meet the production levels anticipated under the project would be undermined. Pricing and Costing 1! 39. The farm crops to be produced under the project require the involve- ment of three marketing boards: The National Milling Corporation (NMC) for maize and beans, the Tanzanian Cotton Atithority (TCA) for cotton, and the General Agricultural Products Export Corporation (GAPEX) for groundnuts. These marketing boards, together with KILIMO, the Prime Minister's Office, and other Government Offices directly concerned, annually fix prices for each crop for the subsequent season. Two sets of prices are of particular significance to the regional cooperative unions: the producer price and the into-store price. The producer price represents that price paid the Xi See Table 5. ANNEX 14 Page 14 farmer and has been fixed for the 1973/74 season at a uniform level through- out the country for maize, beans, and cotton, 1/ although it varies with the regional cooperative unions' wholesaling costs for groundnuts. Following certain assumptions on the quality of produce generated under the project, producer prices per bag for maize, beans, cotton, and groundnuts are estimated under the project to be T Sh 32, T Sh 76, T Sh 48.6, and T Sh 124 respectively. The into-store price represents a theoretical price paid by the appropriate marketing board when produce delivery is made to its stores in Dar es Salaam. 40. To the producer price are added the intermediate costs of bagging, shirinkage, cooperative levies and transport and financial charges. However, the costs assumed by the cooperatives for handling, storage and marketing are excluded and are instead reimbursed by the marketing boards through separate accounts. Into-store prices have been fixed for the 1973/74 season for maize, lint cotton, and groundnuts. Although an into-store price has also been recommended for beans, the regional cooperative unions are free to sell any beans they handle outside the authorized marketing board if they can obtain a higher price. In summary, therefore, the cooperative union is still permitted to fix, in accordance with its own costs, the producer price of groundnuts and the into-store price of beans, if a more favorable buying price can be found. For both maize and cotton, however, both the producer into-store price are fixed by the appropriate parastatal, yielding a standard margin for all regional cooperative unions which might or might not be adequate to cover their individual costs. Both the National Milling cooperation (NMC) and the Tanzania Cotton Authority (TCA) are aware of this problem. NMC is currently reviewing the cooperative unions' demand with the intention to increase the into-store price for 1973/74 maize to reflect more accurately the unions' actual costings. Likewise, TCA claims to reimburse the unions for any costs not fully covered by the into-stores price if pro- perly Justified and documented. 41. The wholesaling costs per ton of the regional cooperative unions are fixed at the beginning of the agricultural season. Certain costs, if assumed by the union, are stipulated at specific rates by the appropriate parastatals. These include handling, storage, insecticide treatment, and marketing. The remaining costs are estimated by the unions' management in light of the previous season's experience, subsequently recommended by the annual meeting of the cooperative union, and ultimately approved by the Regional Development Director in his capacity as Assistant Registrar of Cooperatives. These costs include local transport, bags and twine, interest charges, insurance, shrinkage, and society and union levies. The society levy is designed to cover all costs assumed by the primary society, such as overheads, storage, and handling charges, as well as a provision for bad debt. The union levy, on the other hand, is intended to cover only overhead and provision for bad debt. 1/ Some variation has been introduced for maize, where prices differ between regions designated priority and non-priority. 2tA'WEX 14A Page 15 42. Recent efforts to standardize the cooperative margins by fixing both producer and into-store prices nationwide arises from the substantial cost variances that have existed heretofore among different regional coopera- tive unions. Fcr examDle, society levies for maize are estimated to vary as much as 400% shrinkage estimates for maize as much as 300%, cost of bags and twine for maize as much as 66% I/ On the other hand, the effective introduction of fixed margins does not allow for the differences in variable costs that obviously exist among regions. For the time being, therefore, the movement toward fixed prices, both producer and into-store, proceeds, but with the growing realization that some flexibility will have to be retained to reflect regional differences. 43. Finally, Table 5 refers to the Union's sellingprice, which repre- sents the actual value received from the appropriate marketing board for produce delivered. It constitutes the aggregate of all costs incurred by the union per ton of produce, from the purchase at the primary society through the variet"r of wholesaling activities to the delivery to the parastatal. The selling price also represents the total revenue accruing to the union for any ton of produce. It is to be distinguished from the into-store price, a theoretical value which does not reflect the full costs incurred by the union, which includes extra-regional transport costs not incurred by the Union, and which is used by the marketing boards for accounting purposes. 1/ An Appraisal of Marketing Costs for Crops Handled by NAPB for Use in (nov'r-mer,t s 1973/74 P'ice Review, P.M. Newhouse, Dar es Salaam 1972, pp. 6'-67. TANZANIA KIGOMA RURAL DEVELOPMENT PROJECT KIGOMA Cooperative Union Trading and Lending Cycles deliveries (AU advanCeS TRDB Z> UNION . \s s o o X A G A~~~~~~~GPEXS Pr ~ons Cr0 NBCM TCA ~ ~ ~ ~ ~ ~ ~~~: WorldNBank-8728C,^8X Q ~ ~ ~ ~~~~~Wol ak82 TASiZAS L-A KIGSC% RP,AL DEVEXLoPSEWr PROJECT KIGONA COOPERATVE UNION Annual Cash Plo, 1/ Project Years TSh (000's) I. Trading Activities A. Receipts 7YI PY2 PY3 PY4 PYs PY6 PY7 FY8 PY9 PYIO PYl1 FY12 1. Sale of Crops 2/ 3,442 9,993 17,887 26,536 31,417 33,848 35,586 36,898 38,010 39,115 40,265 2. Sale of Inputs I/ 630 2,481 5,047 7,706 9,474 10,172 10,353 10,354 10,354 10.354 10,354 3. Hiring out Vehicles 4/ 56 84 306 473 557 613 641 668 696 724 724 Total 4,128 12,558 23,240 34,715 41,448 44,633 46,580 47,920 49,060 50,193 5,343 B. Disbursements 1. Purchase of Crops -t 2,765 8,150 14,702 21,930 26,178 25,326 29,855 30,978 31,914 32,833 33,795 2. Purchase of Inputs 61 527 2,073 4,221 6,443 7,925 8 515 8,669 8,824 8.824 8,824 8,824 3. Operating Costs 7/ 134 886 1,876 2,950 4,160 4,670 4,978 5,198 5,389 5,568 5,740 5,873 4. Interest on Cash Advance _/ 22 58 99 145 170 185 199 209 218 228 237 5. Interest on Seasonal Inputs 9/ 45 176 359 548 674 724 737 750 750 750 750 6. Debt Service on Union Investment 101 99 283 576 784 987 1,155 1,299 1,232 1,062 923 785 646 7. All-.ce for Bad Debt. 11/ - 26 104 211 322 396 426 433 441 441 441 441 11 Total 233 4,554 13,013 23,326 34,535 41,168 44,453 46,323 47,653 48,638 49,601 50,566 Surplus (Deficit) (233) (426) (455) ( 86) 180 280 180 257 267 422 592 777 II. On-Lending Activities 12/ A. Advances to Villages 1. Tf&redtitcrfor Village 1,059 1,396 1,426 1,613 631 636 380 409 669 724 837 2. TRDB Seasonal Credit 630 2,481 5,047 7,706 9,474 10,173 10,353 10,534 10,534 10,534 10,534 3. NBC Cash Advances for Crop Purchase 2,135 5,670 9,654 14,225 16,705 18,153 19,502 20,444 21,380 ZZ,300 Z3,262 Total 3,824 9,547 16,127 23,544 26,810 28,962 30,235 31,387 32,583 33,558 34,633 B. Union Debt Service Obligations 1. TgDB Credit for Village Infra- structure 118 314 559 870 1,098 1,207 1,185 1,154 1,159 1,145 1,180 2. TRDB Seasonal Credit 571 2,250 4,580 6,991 8,599 9,239 9,406 9,574 9,574 9,574 9,574 3. NBC Cash Advances for Crop Purchase 2,158 5,728 9,753 14,370 16,876 18,339 19,702 20,654 21,599 22,528 23,500 Total 2,847 8,292 14,892 22,231 26,573 28,785 30,293 31,382 32,332 33,247 34,254 1/ The following calculations include only disbursements and receipts relative to the project, although a relatively minor amount of the Union's operations will continue independent of the project. The project year is considered to be the Government's fiscal year, from July I to June 30. Although the fiscal year of the Kigona Cooperative Union is Klay I to April 30, in the interests of simplicity and consistency, all calculations are based on the Government's fiscal year. 2/ Source is Table 3. 3/ Source is Table 2. 4/ Source is Table 7. 5/ Source is Table 3 a X 6/ Source is Table 2. 7/ Source is Table b, f,' flrleed frun Table 9. 9/ Derived from Tables 2 and 9. 10/ Source is Table 8. 11/ Esti-,ted at 5% of vredit advanced for c-sa-onal i,puta. i2s able q IoU lhs fii rrr.-e. . 1 TANZ1AN KIOIA SIALA V11,MlnlT PROJECT KlOCClA cOorcATm UNION trbsue ed 5Sale of Articulto-I Imp.ts Protect Years rT2 FY3 nY4 Pf5 P76 ?Y7 nTa PY9 PlO 1YII PY12 A. ,t-bs.. of IPfoi volt cIne Volu Colt Volv_ Cos Volt Cost Volo.V Coat Vol. Cast VOI. Cost Volt_ Coit Vol Cost V.1e Cost Vo.v Coot Unitt Costs IteA- tlit T 9h 1. Du Tot. 140d.02 115 161,900 449 632,200 933 1,313,700 1,416 1,993,750 1,774 2.497,850 1,949 2,744,250 2.010 2,830.100 2,071 2,916,000 2,071 2.916.000 M 071 2.916,000 2,071 2,916,O0O 2. CAN TOo 1118.99 115 128,700 449 502,450 933 1,044,000 1,416 1,54.3500 1,774 1,985,100 1,949 2,180,940 2,010 2,249,130 2,071 2,317.450 2,071 2,317.450 2071 2,317,650 2,071 2.317,450 3. n1 DrT ti.. 1.98 21.000 41,600 56,000 110,900 91,000 180,200 126,000 249,5w 133,000 263,350 133,000 263,350 133.000 263.350 133 000 263.350 133.000 263,350 133,000 263.350 133,000 263,350 4. .ULV Spr-ys ...h 133. 10 350 46,600 1,225 163,050 2,100 279.500 2.97S 395,950 31125 442,550 3 325 442,550 3.325 4421550 3.325 442.550 3.325 442,550 3.325 442. 550 3,325 442,550 5. oatteiy sets of 16 23.70 350 8.300 1,575 37,350 3,325 78,800 5,075 120,3w 6.300 149,300 6.650 157,600 6.650 157.600 6,650 157.8w 6,650 157.600 6.650 127,600 6,650 157,600 6. Endo-ulf- Ittrs 13.28 10,50w 139,450 47,250 627,500 99,750 1,324,700 152,250 2,021,900 189.000 2,509.900 199,500 2,649.350 199,500 2,61;:,88 199,500 2.649,350 199,500 2,649,350 199,500 2,649,350 199.500 2,649,350 7. SSP Ton 772.12 - 100- - too 1 77.200 100 77,200 100 77,230 100 too 77,200 100 77.200 100 77.200 100 77.200 Total 526,550 2,073,450 4.220,900 6,443.100 7,925,250 8,515,200 8,669,300 8,823,500 8,823,500 9,823,500 8,823,500 8. Sale of InouC un0t Co ts 0t_e U0,1 T Sh 1. WA? Too. 1636.80 115 188,250 449 734.900 933 1.527,150 1,416 2,317,700 1,774 2.903,700 1,949 3.190,100 2.010 3,290.000 2,071 3,389,800 2.1 3.389.0 2.073 2.3898 2.071 3,369,800 2. CAN Too. 1322.09 115 152.050 449 593.600 933 1.233.500 1.416 1.072.100 1,774 2.345,400 0,949 2.576,750 2,010 2,657.400 2,071 2.738.030 2.071 2.738,030 2.071 2.739.050 2,071 2.738.050 3. 35 DDT 1g.. 2.45 21,000 51,450 56,000 137,200 91,00 222,950 126,000 308,700 133,000 325,850 133,000 325,850 133,000 325,950 133,000 325,850 133,000 325.850 133,000 325,850 133,000 325,850 4. ULV Sprayer Osoh 162.80 350 57,000 1,225 199,450 2,100 341,900 2,975 484.350 3,325 541,300 3,325 541,300 3.325 541,300 3,325 541,300 3,325 541,300 3.325 541,300 3,325 541,300 S. Bettery sts of 16 28.88 350 10,100 1,575 45,500 3,325 96,050 5,075 146,550 6,300 181,950 6,650 192,050 6,650 192,050 6,650 192,050 6,650 19Z.050 6,650 192,050 6,650 192,050 6. Eado-olf.. lit-r 16.30 10,500 171,150 47,250 770,200 99,750 1,625,900 152,250 2,481,700 189,000 3,080,700 199,200 3,251,850 199,500 3,251,850 199,500 3,251,850 199.500 3,251.850 199,500 3,221,950 199,5w 3,251 850 7. 301 Too. 946.25 - - - 100 94,600 100 94,600 100 94,600 100 94,600 100 94,600 100 94,600 100 94,600 100 94,600 Total 630,000 2,480,850 5,047,450 7,705.700 9,473,500 10,172.500 10,353,050 10,533,500 10,332,500 10,533,300 L,533,500 DI Derld fro A.... 15, Table 9 nd frms Table 4 of this a.-x. 2/ unit co00s d-"fa Oro. Table 4 o8 thIs .ae.. elF__ " t _g*s 7 n2 n 9 64 Ffvrs m 96 67 WI 9 nso ml ml .a. Valo "s* V.1. SlW eg ve *al_ WP Volvo le Volvo _ g vol.6* SW WO1. Iw 1.1.. SW Vlao lag ",se sag Ylose a .ta 16. 569 529.560 39.838 1,274.630 65,535 2.097.130 91.92,9 l,%1.740 9,147 3,172,703 106.965 3,356.200 11-.100 3,525.062 116,7 3.715,465 122,62 3,9"2. t12.7M7 4. 5 132,241 6,21,920 b. lac- 11.479 872,44 25,484 1, 398.990 3,750 ,020,950 51,49 4.065,410 52,726 4,007,085 52,017 3,9f3.24C 51,297 32.89,570 50.70D 3,V59,335 50,65 3.0h9.630 51.0n 1 3 ,8,070 12.012 3,59)2.,7 , Ctttoo 20,4Al 13953,280 105,289 4,927,620 197,724 9,305,020 295,115 1,945,670 359,43 17,077,700 37,7e0 18,019.175 392,89 1,16.615 06,195 19,417,495 419,185 20,127,760 41.761 20, 0 44,715 21,5, 4. r-Lta - - Z,336 276,560 7,.5 977.225 15,6023 1920.63, 23,717 2,926.61 79,156 3,614.60 31,417 3,95,501 3Z,339 4,012.425 33.330 4,11*2. 36,130 4,2,153,5 Total 57.450 2,765,300 170,630 9,130.450 310,350 14,701.700 448,50) 21,130.350 526.900 726,176.300 3.0.400 2.325.600 613,300 29.855,150 605.15 30,077.0 62i.900 31,913.70 661.90 523,630 "63,=5. 21,793,Ze Tos VOl"e 2,. Valr. Te V.1, Too. Val". T-. Yal, 7.. 9.l To.e Vol" Ie.. 9.8. Twe V.1o. T_ Vole T_ 9*e1 * p,.lao 1,429 773,400 3.44 1,06t,250 s,662 3,064,250 7,943 4,290,750 5,5S6 4,635.00 9,062 4,904.930 9,517 5.190.401 O4,096 5,450.550 16."Sf 5,736,Gf0 81,919 5,971306 11,426 6,26*3,71 b. R- 1,012 1.036,9S0 2,266 2,316,400 1,So6 3,141,650 4,718 4,927,500 4,680 4.685.450 4,S9 4.,791.700 4,525 4,729,0"0 4,479 4,677,650 4,4,66 4,6644400 4,03 4,702,950 4,51 4,790,454 C. Cottan 1,07 1,611.200 3,28 5.74. 200 7,189 10,825,200 10,730 16.157.250 13,019 19.679,030 13,772 Z0.7377.90 14,256 21,5111.50 1,769 22,219.150 15,242 22.951.400 1C,69 21,639.550 161.70 24.30,400 d C-ournd-t. - - - 206 33S.200 706 1,1521200 1.376 2.25.650 2,092 3.414,150 2,572 4,117,$00 2,771 4,522,29 2054 46.637,750 2,040 4.7",to00 3.028 4.941.700 T°tol 3.511 3.441,550 9,515 9 9939,840 16.I56 I1,"7.300 24.097 26,535,700 27,641 31.417,300 "r S1, 100 30.o 3s55 1 32,105 36.6. 33,159 316,0 ,,55 314.II ,.110.96 35.211 4,8440.906 1/ 5.6t td that .rk.tbl. p,du of t--1a, otton, gOaaata, o4 bo Int ba8t buha .ltlet fer M1M2, 68.6.76 7 124 p. bag a tttns d w1 i to ael rt rb. aeprtte netet-l er a.xtd t S 541.28. 1,5015 90. 1,632.00 _1d 0 a 47 par tCo r-.pttlTly 1lffar..- rap,nt. .11 1.elsalla o t. ornt t7 the polary *otAtL- aM ..t e. Diff.tra- to volum r.efltt prod-. Abriobag. 1l8r to Tblo 5 * f to lotivo of et Iltwg pvt.. ANN= 14 TANZANIA XIGMiA l.URh DIVKLOPEIIEU PROCT KIlGI COOPhATIIVE UNIOSI Unit Costs of IholesaliigAicultural InDuts DA? CAN DDT ULV Batteries hdosulf an SSP Sprorer (tons) (tns (t (each) (Sets of 16) (litrs) (tons) A. rost FOB [iguma 1,408.02 1,118.99 1.98 133-10 23.70 13.28 772.12 B. Wholesl Cgsts 1. Transport Y 67.50 67.50 .01 .30 .02 .02 67.50 2. Insurance2/ 1.60 1.30 .02 .16 .03 .02 1.C0) 3. Interest 3/ 119.68 95.11 .18 1.i44 2.05 1.13 656o3 4. Handling i 10.00 10.00 .01 1.50 .20 .25 10.00 5. Union Levy / 30.00 30.00 .25 16.30 2.90 1.60 30.00 sub-total 228.78 203.91 .47 29.70 5.20 3.02 174.13 C. Cost to Society 1,636.80 1,322.90 2.45 162.80 28.90 16.30 946.25 1/ To calculate transport costs of agricultural inputs, several of which have no _eaningful weight equivalent, a comn measure of a bag Of maize has been used. Assumed that about 60 begs, or approximately the equivalent of 5.5 tons, can be carried by a 7-ton lorry whose costs are TSh 1.35 per revenue ton dle. Roughly estimated that 10 baga of SSP, DAP, and CAN equal 1 ton, and that 1 maism bag is equal to 90 kg of DDT, 4 ULV sprayers, 50 sets of 16 batteries, 60 liters of endosafan. Estimated that inputs must be transported, on the average, 50 miles, and although soe of the lorris sWhuld be able to retrn with marketable onrplus, transport costs have asumed responsibility for the full round-trip oosts. 2/ Cost of insuring input* dhile in the possession of the Union ie estimated to be TSh 2 per TSh 2,000 value insured at sell4ng price to society. ]/ Interest charges are estimated to be 8.5%1 and loan is expected to be outstanding for 1 year. Wz Handling charges derived fro handling charges of a mai;e bag, based on the equivalent volumse cited in footnote 1. 0/ Union levy calculated to be Tab .03/kg for the fertilizers and 10% of the selling price for other inputs. ANNEX 14 TArWIA Table 5 KIGOQA UBRAL DRKI IFIT PiCJET Kisea Cooperative Union Urnt Costs of Wlolesalin Marketable Surelus / miie Cotton Oroundnut Sean3 Rate (5i/ Sb/ToY Ba(Tsb) TSb/Tor / Bate (TSh) lSh/Tor. Rute (TShJ TSh/Tor// A. Producer Frice 32 per bag 352.00 b8.6 per bag 1,336.50 124 per bag 1,346.00 76 per bag 836.00 a. Society Lt / .03 per kg 30.00 0.066 per kg. 66.oo .03 per kg. 30.00 .03 per kg. 30.00 C. Cooperatives' Wholeseline Costs 1. Local tranaport 1.35 0 50 mi. 67.50 1.35 0 50 id. 67.50 1.35 0 50 mi. 67.50 1.35 0 50 mi. 67._0 2. Bags and twie 4/ b.65 per bag 51.00 4.65 per bag 51.00 b.65 per bag 5;.00 3. Bank inteet S 6.5% for 3 so. 5.70 6.5% for 5 weeks 9.biO 6.5% for 3 so. 21.80 6.5% for 3 mo. 13.5C 4. Cash Insurance / 2.0 per 2,000 .40 2.0 per 2,000 1.80 2.0 per 2,000 1.30 2.0 per 2,000 8.bO 5. Inventory insmrance W 3.0 per 2,0DO .50 3.0 per 2,0O0 2.60 3.0 per 2,000 2.0ro 3.0 per 2,000 1.30 6. Handling charges / - - 5.80 - ?. Storage § - - - _ 1.25/bag/week 56.10 - S. Shrinkage 2/ S% 1.410 - - 2% 26.90 2% 16.70 9. Marketing 39/ - - - - 1.00 per bag 3.60 - - 10. Union levy / .02 per kg 20.00 0.22 per kg. 22.00 .02 per kg. 20.00 .02 per kg. 20.OC sub-total 159.20 103.30 256.00 178.40 D. Union's Selling Price la/ 541.20 1,505.80 1,632.00 1,044.40 E. Trwnweort to Dur as Soa_m IV 750/ton 37.50 1250/ton 93.75 75.0/ton 37.50 F. Total Into-Store Price W/ 578.70 1,662.25 1,081.90 3/ The rates are oelculatad on avere annual voluse of marketable surplua handld by Union during first 3.2 years. They do r.ot reflect the actual unit coot of the previous yer, aa they should in practice, and therefore do not refleot decraing unit costs that would normally oocur as wolim hadled increased. 3/ Iach ton of matis, ootton, grouTndute and beans calculated to have 11, 27.5,11 and 1 bags respectively. 3/ Costa incurred by tha primary society for handling, storage and shrinkage are not included under the separate categories and are only compensated through the general socistJ levy. W Transport oosts based on estimate of Table 7 , pro rated per ton of produce. Because denait3, or ,otton is les thar that of maize, beans or groundnts, both trport and handling charges wil be higher per ton. Howver, because distne ootton is to be transported is lose than that for other orope, no diffrentiation by crop is side. Average distance calculated acording to infomation containd in Table 7 . All locl transport costs borne by Ouio. / Cost asms new bag osts TSh 4.65 and ne usad for maie, beas and groundnuts are returned for the following seasoc while virtually all used for cotton ar rturnsd. Costa of begs win be borne by Union for mAise, bena and groundnut., but costs of baga for cotton, roughly estimated to be Tl 1.00/bag to account for wstap eand increasing number required, are taken up by TCA. r/ Both bank interest and insurance for cash advance for crop purchase considered to be borne by Union, for an average 1 month for cotton, 3 montha tbr other crops. Period firm receipt of fasmers' produce until papsAnt by appropriate parastatal expected to be 3 montbs for mtses, beans and grouninute, and 1 sonth for cotton. Although for 1/3 of this period produce is expected to remain in village godown, all inventory insurance charges asumed to be borne by Union. H/ Handling charges include only costs of Union at both of its godowns for unloading, weighing, rebagging of 10% of the turnover and stacking when produce arrives end weighing and loading of produce to be delivered to the respective paratatal, based on PYS as typical year. Union assmes handling charges ony for goudnuts, as all other crops handled by relevant paraatals at their respective godo s. / Rates as determined by the apropriat- parastal. Because NHC regional godowns under construction and storage facilities to be provided by TGl at local ginnery can be assumed to be used respectively for all mainze beans nd cotton, Union would be reaponsible only for storage of groundnutes for which it would be reimbursed by OA1. Oroundnuta are estimated to rm_dn in the nion's reional godowne an average of lionth. All storage costs borne by primary society are again eVpected to be covered by societ levy. 2/ Full costa borne by Union. 32/ Marketing costs wben incurred by Union re reimbursed by appropriate parastatal at preecribed rates. Groundouts are only cocnodity stored, nd therfore Marketed by the Union. UOdon is assumed to market 25% of the groundnuts it handles. / Levies taken frm The Lint and Seed Marketing Board directives for 1972/73 for otton nd fro. existing rates in Kigoma Region for mgie and groodnute. The existing rate of ShO.05/Ag for beans was considered unjustifld and calculated instead in line uitk the maize and groundout rates. 3 The Union's selling price reprmenta the price paid the Union by the approPriate parastatal when ownership of the produce 76 transferred to the parstatal. rS Cost of transport of produce frm Kigo, to Dar es Blsam is responsibility of appropriate parastatal. Such transport costs are included in calculation of into-store price, andre a adjusted here to refloet that an estimated 50% of maise and beana and 75% of groundnute is expected to be transported outede the region. The railroad rate of Sh7.50 per 100 kg igma-Dar es Salem for maize hae alec been used for bean, wherea the transport rate for grwdmts is ShI2.5 per 100 kg. 3WUnion reimbursed for maiso and beans by WEC and for groundnutg by GAM for any handlin, storage, isciietetnn n aktn cOtes cu Instplttb tePrpi,epttV ^ oae insecticide treatment and narketint cot nurred by the Union at rates stipulated by the aPPropriate parastatals. None of these are included in the "into-store price', which is a teoreticeL value that includes All other costs ast med by the Union plus the ligoma-Dr as Salae_ transport costs. For cotton, howver, no into-sftore price has bean Indicated, bcnse the procdure. proposed under te prOjact - whereby TCA instead of Union asninues responsiwiity for internediate handing , storage and gin-in - do not conform to normal practicea.dieeyrne the cocept of into-store price of little value.an l-brd , ~ ~ ~ C 9.4 fri I- 1- p_ -_ A_'. t Zo e-, a n*[¢ "' ." r a'i. 2 a3E $Xv '*i tii t f t1t11 0 Oh 2 * * - -. ; N," 0 _3 > 3 * 33 Ca. n C,~ N ** g Ct~a, - r~~~~c _ i E ~ .~ g. .i . .'g - '' iF go§g§ , ~ ~ ~ - n: w a'. o~ x 3s1.Ws Wef_- K 3 f t*e og * . o . _1. "MW . a,,,X *~ s *-F o . 'I<0s c" Kt 3 aai g 4?2,. th O Bc ' 4 N- W eo a.' ,_ g__N8 1- ' ' ~' '' * 2 * s a '*_ Sco8S S8 S S.g.. "8 a.'4U>o r. - _ s I - . of Lb O 41 9- HI na S8S888888888 .4 t X r c _s " n X:4 G ft r * r '44 _ a * O C 4*~~~~~~~~~~~~~ p1 _t. 30 0N -N oKr "wr -N >So 3.10 ~ ~~~~~~~~~~~~~~~~~~ S. 8~ 8a 8 8 8 8o 3 'a'C _A N :3 ^t1 : I~~~~~~ ~ ~~~~~~~~~~ ;SA g g 8o8oa8o8 8 g 3:iY §.L TAN ZANI A KIGOMA RURAL DEVELOPMENT PROJECT KICOMA COOPERATIVE UNION Buildup of Motor Vehicles Categories Project Years A. Lorry FYI PY2 PY3 PT4 PY5 FY6 PY7 Ff8 PY9 FY10 PII PY12 1. Lorry loads of 1/ marketable produce - 720 1,945 3,375 4,910 5,630 6,010 6,290 6,535 6,750 6,960 7,170 2. Lorries required V 5 16 27 40 45 48 50 52 54 56 57 3. Union lorries required a. Total fleet3/ 2 3 7 11 17 20 22 23 24 25 26 26 -b. Increental 4/ 0 1 4 4 6 3 2 1 1 1 1 0 c. Replaced 5/ 0 1 1 0 1 4 4 6 3 2 1 1 d. Total credit needa 0 2 5 4 7 7 6 7 4 3 2 1 4. Non-Union lorries required a. Quantity 2 9 16 23 25 26 27 28 29 30 31 b. Coats (TSh) 6/ 84,400 379,800 75,200 970,600 1,055,000 1,097,200 1,139,400 1,181,600 1,223,800 1,266,000 1,308,200 B. Four-wheel drive wt0sl vehicle a. Total fleet 5 8 9 10 12 12 12 12 12 12 12 12 b. Incremental 4 3 1 1 2 0 0 0 0 0 0 0 c. Replaced 0 0 0 1 0 4 3 1 2 2 4 3 d. Total credit needs 4 3 1 2 2 4 3 1 2 2 4 3 1/ Estimated that 60 bags of maize, beans, or groundnuts, or 120 bags of cotton can be carried in one lorry-load, yielding an average 5 tons/lerry load. See Table 3 for the volume of marketable surplus requiring transport. 2/ Lorry requirements calculated on the basis of projected average lorry trips per week. Average time for one-way trip estimated to be- Kasulu: village to regional godown 3 village to Kasulu ginnery 2 Kigoma: village to regional godown 1.5 or Kigoma ginnery 1.5 Kibondo: village to district godown 1.5 village to Kasulu ginnery 3 2.5 average hours (@ 20MPH - 50 miles, average one-way trip). 3/ Calculated that Union will own slightly less than half the vehicles needed to meet its lorry requirements. 4/ Only vehicles incremental to total fleet through PY5 are to be financed by the project. Incremental vehicles beyond PY5 and vehicles replaced will be subject to TRDB credit on normal terms. 5/ Lorries and 4-wheel drive vehicles assumed to have respectively 3-year and 5-year ec,noominc lives, after whtch they, anid spare parts representing 2tY of their original value, are replaced, under regular TBDB financing. Lorries assumed to have residual value equalling 25% of original value. 6/ Cost of the hire of non-Union lorries asaured t., .e 3511 35 revenue ton mile - the Sane rate L ;on attrihutte te its --, lor-e,, - an 'i 7-ton lorry hired for harvest season (Shl .35 /rcvenue i.n mille S toiis x 50 mi I- E ninr1 4 KIGOa& RURAL D EfM PROJKCT KIOWA COOPDT IVE UNIW Union' a Lorrv Account 6/ Pro1ect Years PF2 pm3 P4i PY6 FY7 Ts PY9 PYlO FYI1 PY12 A. Revenue 1. Union transport inome 1/ a. Agricultural inputs I 16,050 62,500 129,550 203,200 252,450 276,300 284,550 292,800 292,800 292,800 292,800 b. Marketable surplus 3/ 242,450 656,450 1,139,050 1,656,300 1,899,500 2,026,950 2,122,250 2,205,400 2,278,050 2,348,500 2,419,400 2. Hiring out of Union lorries 4/ 55,700 83,550 306,350 473,450 557,000 612,700 640,550 668,400 696,250 724,130 724,100 Total 314,200 802,500 1,574,950 2,332,950 2,708,950 2,915,950 3,047,350 3,166,600 3,267,100 3,365,400 3,436,300 B. Expeznditures 1. Vehicle operating costs 5/ 255,300 595,700 936,100 1,446,700 1,702,000 1,872,200 1,957,300 2,042,400 2,127,500 2,212,600 2,212,600 Net M Income58,900 206,800 638,850 886,250 1,006,950 1,043,750 1,090,050 1,124,200 1,139,600 1,152,800 1,223,700 Footnotes 1/ Rate of Sh67.50/ton allowed Union for transport of inputs and outputs in the ,mrgin between its buying and selling price. 2/ Derived from Tables 2 and 4. 3/ Derived from Tables 3 and 5. 4/ See Table 7, footnote 1. 5/ See Table 7, footnote 2. 6/ This is a "dummy" operating statemnt of Union's lorry operations, reflecting total income and expenditure accruing to Union lorries, and not cash-flow. This table reflects full accounting of Union operations, including revenue for transport services included in produce selling prices and including debt repayment. These calculations are not used as inputs for other tables. TAMMIA KIM^A IUdML D0WUh98 Pp J2CT atlo Ceseretitv Ouiso Capita1 Tave.tMn.t nd Rpev-nt 7 re ecr e arrs tet PfYI n2 7m PY4 FfS 7Y6 PY7 pYs PY9 PY10 PYIY W12 A. Recetipts 1. Lorries 158,200 429,400 361.600 610,200 542,400 452.008 497,200 293,600 226.0W0 1538,200 67,800 2. Four whel drive vehicles 189,600 142,200 47,400 94.800 94,600 189,600 142,200 47,400 94, 00 94, 00 189,600 142,200 3. Spare pert. 47.400 60,750 124,650 114,100 181,900 205,600 171,150 170,050 114,100 915300 92.600 58,150 4. Office nd offi.e equitpmnt 200,000 5. kpair od aeinteae hop 80.000 6. Ki4_ Sodo- 700.00 70,000 7. Kibodo godoem - Section 1 150.000 550,000 S. Kibondo godos. - Section 11 150,000 550,000 Total 737,000 1. OI 150 601.650 720,500 1,4366900 937,600 765.350 714,650 502,700 412.300 40.400 268,150 8. Disburasmets 1. Lorries I/ 61,950 230,100 371,700 548,650 592,850 628,250 564,050 486.750 398.250 265.500 177 000 2. Four sheet drive vehicles 1/ 74,250 129.950 148.500 111.350 92.750 1486450 167,050 148,500 111,350 92,750 14U,450 167,050 3. Spr pe rts 11 186550 50,1S 0 99,050 125,200 164,800 If,.400 218,700 214,100 178,300 147,150 116,800 94,850 4. Office nd office equipsnt 2/ 6,500 31,350 31,350 31,350 31.350 31,350 31,350 31,350 31t350 31,350 5. RepeSr nd meintenonce hop 2/ 2,600 12,550 12,550 12,550 12.550 12,550 12,550 12,550 12,550 12.550 6. Kigsm godown 2/ 2.250 13.200 21,90 1,900 21,90 00,9O 219900 21,900 21,900 21,900 10.950 7. Kibondo godo.m - Sectioc 1 2/ 4,050 41,350 109,750 109,750 109.750 109,750 109.750 109,750 109.750 109.750 86,250 8. ibondo godorv - Section 11 2/ 4,850 41,350 109,750 109,750 109,750 109,750 109,750 109.750 Total 99,300 283,100 576,100 763.800 9866.00 1154.600 1,299,300 1.231,950 1,061,700 923,450 784,700 645,650 1/ Cost of * 7-ton lorries celculted es Sh 90,400 *nd of 4-whel drive vehicles as Sh 47,400; econotmi lifes estimted to be 3 yeerr for lorris and 5 ye.rs for 4-vheel 4rive vehiclo.; lorriss .seused Co have r-sidual value equal to 25% of purchase price. Repayment period celculated a. 3 years interest hargeo s 8.5% on lorries, 4-heel drive vehicles end spere parts. Asumed that spare ports equsl to 25% of vehicle value -re bought at time of vehicle p.rchase. Only increeset.l vehicles during the first five years are financed by the project. See Table 7e for schedule of vehicle bWild p. / Godons, office, and oiotenance shop all calculated to have repayment period of 19 ye.s,. with one year grace period, and to h.e- interest O-rge of 7 5%. I.nerst 1s paid during the Srtce period ton-tuctio, nd therefore disburse-nts, of e-ch are spread over 2 yers. Kign__ gododn calculttd at 650e in ite for capacitv oa 500 on.. The ihondo godown calculated at 6750 n o he built in 2 -ton for capaclt* f 5000 ton. BoCh asse unit co.t$ of TSh 280 per ton capacity. TAIlZA2II KG1A0 RURAL DEVEULOPsT PROJECT P1G09A CODPVIATIVE 35108 "I n72 ) P74 j PY6 pYT pyg ?Te Fy0 rno F2 A. Advances to llies I/ 1. Vtl_z t iefratructur 2I a Villep R eadi mis. sheller 75,000 93,750 93,750 93,750 75,000 93,730 3,o50 93,750 75,000 b. VIllae - N stIle 228,000 342,000 342,000 456,000 456,000 342,000 228,000 342,000 342,000 Villap - Godown. 736,000 920,000 910,000 970,000 d. Village - Fhing nt nd ear 20,000 40,000 70,000 143,000 175,300 218,800 286.700 315,300 346,800 381.600 419,700 sub-total 1,.09,000 1.345.750 1.4253750 1.612,750 631,300 635.800 380.450 409,050 668,s30 723.600 836,700 2. Credit for soaaonal .ts. 3/ 630,000 2,480,850 5,047,450 7,705,700 9,471,500 10,172,500 10,353,050 10,533,500 10,533,500 10,533,500 10,533,500 3. CAmh_aYangL for cr55 pd tchse6 4 2,t35,300 5,670,000 9.654,250 14,224,650 16.704,850 18,153,100 19,502.100 20,444,400 21,380,200 22.299,600 23)261,700 Total 3,824,300 9,546,600 16,127,450 2,A43,100 26,809,650 28,961.400 30,235,600 31,386.950 32,882,250 33,556,700 34,631,900 8. Uiota Dbt Slrvina Oblizatteoa 1. Village sconoi intraatr,ats rs a. Villa - Rad Miss sheltr 22,400 50,400 78,400 106,400 84,000 78,400 78,400 780400 106,400 84,000 78,400 b. Village Mairs sills 56,400 140,900 225.400 338.100 450.800 478,900 394,400 309,900 253,600 225,400 225,400 c. Villge - Codo-ns 27,600 89,700 194,100 307,300 386.000 430,200 430,200 430;200 430,200 43u,2-I.t 433,200 d. Village - Fihlg -t and gear 11,150 33,450 61,300 118,650 177,300 219,500 281,550 335,300 368,750 405,650 446,250 sob-total 117,550 314,450 559,200 870,450 1,098,1D0 1.207,000 1.184,550 1,153,800 1.158,950 1.145.250 1.180,250 2. S*asonl Credit 571,300 2,249,700 4,579,700 6,990,750 8,598,900 9,239,000 9,406,200 9,573,500 9,573,500 9.573,500 9,573,500 3. Cash ancaso r croP purchases 2,157,700 5,728,100 9,753,2f0 14,370,450 16,876,050 18,339,150 19,702,000 20,654,000 21,599,350 223528,150 23,500,150 Totfl 2,946,530 9,292,250 14,892,100 22,231,650 26,573,050 28,785,150 30,292,750 31,381,300 32,330,NO0 33,246,900 34,253,900 1/ T-ble indicates total asah flow during esch year and not aer.ly at ye*r end, obey both credit for seasonal inputs and csh advances for crop purchase should have bhan repid. On ledtig astivitis of Union rsquirs thot cash rhsnge hands 4 times (a) bank to Union, (b) Uion to village, (c) village to Unoc, sod (d) Ulon to Booh For purpossa of this sash flow a.alveIs, only step (b) end (e) have been shown, the first as a di.ebuvssnnt by the Union and tho second as a receIpt fron the cilags. P-coipte fom, villfat nill itcluds not coly cepeytnt of prinipol bhut alec irtor..t, cith the soception of sash ad-asce. EI tEtinted that averae livas of nine hellter, -sein mill, godown, end fishing equ4ipoasnt re 5 years. A years, 15 yema and 3 years -esp-cti-ly, sod asaoed they are .11 r-placsd accordigly,. fieced by * r-gs1- credit frls TIDB. Only the origoal cosimc isfrastauetsre to be finsncd under the project Although s crsdit provision has bees ads for replacsot inv-t t in fishing equipnt tI expected to grow at 10% after PY5. Unit costs of siss sheller, mine sll, and godewn usad to bh Sh3750, 22900, and 36800 rspetvely Coat of fishiog squtpaect astisatad at Sh9000 for b..ch esiee u.it., Sh3000 foe dryiog urit., ShlO fur pr s..re laqss, and 81,O1O fer iLqproed local boat. Iterest CharmI Rtepar_st Paritd leais ehellsr 7,52 4 years Naio sill 7.52 5 years village godpw 7.5% 15 years (iIcluding 2 yesar gran. period) 3/ Credit for seasonal tpots provided on FOR value, railhed Kigos and interest charges assesed accordingly. Table 2 indic-tis uno *n and valu of ipuis at ligos rallhed open which istareat charges are assessed. Estiasted th.t a11 titere-t chagre on seso.al ipurt are re--ponibility of recipiL-t vIllae and eount to no sore ths 1 year at 8.57 per wr, aIthough actual repaysast of both Jeterest aod priocipl sill ost likely set be code until following fiscal year. 4/Cash advecc, provided th-ough Uiion by NbC for parchase of crops, i-volces interest cherge of 6. 5 to be borne by U.i,on Ctush sd--oes e7stised to be ont-a-dinig for, 3 cooks for corcon and 3 cynths te or er r crops (verage iotareat rats of .01025). Cash advance is equal to csop purchased ciou- tho credit Ior seasocel inputs. AkSNEX 14 Table 10 T~ANZANI KIUGA EUSAL DEVELOPMNT PROJECT Czu?0rh,Lw of Prima society (tlodel350-feiily villege) 2 3 4 5 Villete Years 7 8 9 10 A. Heceipts 2/ 49 1. Bwrewings for inp.t ptzrhases 14,900 77,850 113,700 113,700 117,900 117,900 117,900 117,900 117,900 2. BDrtwin* farcp purchae.s 3l 88,350 156,850 155,200 180,500 193, Bo 204, 9t! 214,0'( 223,000 233,300 3. Society levy k' 7,360 12,380 13,750 14,500 15,220 15,790 16,290 16,790 17,290 Total 145,610 2147,080 282,650 308,700 326,920 338,590 348,190 357,690 368,490 B. Disbursements 1. Crop purchases 5/ 138,250 234,700 268,900 294,200 311,700 322,800 331,900 340,900 351,200 2. Salaries and wages 2,850 3,200 4,000 4,150 4,200 4,250 4,280 4,300 h,320 3. Audit and supervision 220 370 410 440 460 470 490 5S0 520 Godown debt repayent / 1,380 2,760 4,530 4,530 4,530 4,530 4,530 4,530 4,530 5. Maintenance and repair 9/ 100 100 150 150 200 200 250 250 Total 142,700 241,130 277,940 303,470 321,040 332,250 341,400 350,480 360,820 Surplu 2,910 5,950 4,710 5,230 5,880 6,340 6,790 7,210 7,670) Cusmmulative 8,860 13,570 18,800 24,680 31,020 37,810 45,020 52,690 1/ Investment in and income from a maise sheiler and maize mill to be purchased by each village in Years 2-3 have not been includec here, as they are assumed to be self-liquidating. 2/ Cost of sasonal inputs to village includes margin and services provided by Union (see Table 5). Although TRDB is assumed to provide credit only on the FOR Kigoma price, it is expected that Union, rather than village, will bear any credit costs that might be incurred on the difference of its buying and selling price of irnuts until the village's marketable surplus is purchased. Note that seasonal inputs here include only those provided on credit, not aTnr that might be provided by Ooverrment on subsidy. Village input requirements are derived from Annex 3, Table 9. j/ Cash advance represents the difference between the crops sold and seasonal inputs received by the society. Intereat charges an. insurance costs are borne by Union, unless the society retains balances after harvest it should return. W Village levy was calculated on the following basis: Years of Village Development 1 2 3 4 5 6 7 8 9 10 a. Cotton 1. volume (bags) 1,472 3,424 3,766 3,928 4,074 4,229 4,356 4,486 4,621 ii. levy (TSh 2.4/bag) 3,533 8,21B 9,038 9,427 9,778 10,150 10,454 10,766 11,090 b. Maize, beans, nuts l. volume (bags) 1,401 1,524 1,724 1,859 1,990 2,065 2,138 2,205 2,271 ii. levy (TSh 2.73/bag) 3,825 4,161 4,707 5,075 5,438 5,637 5,837 6,020 6,200 Total levy (TSh) 7,358 12,379 13,745 14,502 f5,216 15,787 16,291 16,786 17,290 Volume of marketable surplus is derived frcm Annex 3, Table 7 nd the levy rates from Table 5: / Derived frm Annex 3, Table 7 W Salariea and wages are calculated on the following basis: Category Years of Village DevelopMent ( TSh) 1 2 3 h4 5 6 7 8 9 10 a. Bookkeeper 1,400 1,400 2,100 2,100 2,100 2,100 2,100 2,100 2,100 b. Godown attendant 1,000 1,000 1,000 1,000 1,000 1,000 1,000 1,000 1,000 c. Seasonal labor 450 850 950 1,050 1,100 1,150 1,180 1,200 1,220 Total 2,850 3,250 4,050 4,150 4,200 4,250 4,280 4,300 4,320 Bookkeeper's salary is expected to increase from Sh 200h/o. to Sh 300/mo. once he has completed his apprenticeship with the UCDA. The UCDA, while in the village during years 1-3, is expected to be on the ooverrsient's, not the village's, budget. The salary of the godown attendant in Sh 200h/o. for 5 months. Secretary, as an elected official, is considered to be unpaid. Seasonal labor calculated on following basis: in year 4, assumed to be roughly 60 lorry loads per vilage, which would take 6 men, working at Sh 0.75/hour, 1.3 hours ech to unload and load, reeulting in a total of Sh 350/year. In addition, to assist in the bagging of marketable produce, a task insolving Spproximataly 50 bags per day during harvest of year 4, seasonal labor to be hired for the 5 months ot harvest at Sb 5.00/day, totalling Sh 600. Total seaonal labor in yer 4 is therefore Sh 950. Seasonal labor for the remaining years of village developoent is calculated similarly on the basis of produce handled. j Audit and supervision costa repreaset 3% of income derived from society levy. §/ Terms of the TRDB credit for godowns are asmmed to be 15 year repayment with 2 years grace and 7.5% interest charges. Godown's original cost is TSh 36,8D0, including storage building with currugated iron siding and concrete floor, adjoining concrete platform, security fencing and al.l office, as wel as Sh 3,000 for a mall safe and weighing scales. 2/ Repair and maintenance is on godown invsetst. ANNEX 15 TANZANIA KIGr(KA RURAL DEVELOPMNT PROJECT Crops Budgets and Production Models TANZANIA KIGOMA RURAL DEVEI0PMENT PROJECT Maize Acre Budgets Actual Incremental Traditional stage I Stage II Stage III Stage I Stage II Stage III Yield in lb shelled grain/ac/- 535 675 1,020 1,700 140 345 68o Gross Return/ac at Tsh 0.16/lb/2 85.60 108.00 163.20 272.00 22.40 55.20 108.80 Cost of Inputs A. Materials Seed at 12 lb/ac traditional & Stage I, 20 lb/ac for Stages II & III /3 2.88 3.36 5.60 5.60 0.48 2.24 T 5% dust at 4kg & Tsh 4.90/ac/4 -.- -.- 4.90 4.90 -.- 4.90 Calcium Ammonium Nitrate at 56 lb7-c 15 - - 33.05 -.- 33.05 Di Ammonium Fhosphate at 56 lb/ac /5 -.- -. - 40.92 -.- -.- 40.92 Tools, including shelling charge /Z7 3.70 4.85 6.75 10.68 1 1.90 3.93 Sub-total Materials 6.5 8,21 17.25 95.15 1.63 9.04 77.90 B. Labor (nan-day) /7 Land preparation 25 27 30 30 2 3 - Planting 1 1 2 2 - 1 - Thinnirg & weedinC 12 14 15 17 2 1 2 Dusting - - 3 3 - 3 - Fertilizer application - - - 2 - - - 2 Harvesting 6 7 8 10 1 1 2 Shelling /8 1 1 2 2 - 1 - Sub-total Labor 50 60 ;r 10 lNet Return excluding Labor (Tsh) 79.02 99.79 145.95 176.85 20.77 46.16 30.90 .Return to Labor (Tsh/nan-day) 1.76 2.00 2.43 2.63 4.15 4.62 5,15 /1 Iission estimates. 77 Based on 1973 official farmgate price. 73 Traditional based on local variety at Tsh 0.24/lb, Stages I, II and III based on Ukirigura Composite A at Tsh 0.2X/lb. 7 Recommended application rate at farmgate price incorporating transport, insurance, handling and 10% Union levr -r.l'nEy, t Cvi1C.L'i ;: 4> is teN ;,.V2 -se d t"r se .'i t 5 A1NIKEX 'Table Page 1 TANZANIA KIGOMA RAIJL DEVEWRIL-WPR }¶OJECT 50 ftmi V e Production Model at Full Develo ent House Ploti ladi ti to 1. Acreage of Cp - ' Mixed 350 - - - - Maize - - 18 166 166 3,5 ikeans _ _ 18 166 166 30Ci ['utton1 - 55 - - 295 3- 1;roundnuts - 75 - J 100 i Production (lb.) /6 Mixed -- - - Mlaize 105,000 - 12,150 169,320 282,200 568,6 7 Beans 70,000 - 9,000 83,000 99,600 261,60- Seed Cotton - 19,250 - - 2952000 3 14,22 3roundnuts _ 26,250 - - 70,000 96,25:- ;ubsistence (lb/yr) /8 li)xed: assorted house plot crops, unquantified Naize: at 200 lb/head x 5 = 1,000 lb/faxrily + 20 lb for seed 357,000 Beans: at. 80 lb/head x 5 = 400 lb/family * 35 lb for seed 152,250 2roeundnut.s: at 20 lb/head x 5 = 100 lb/amily + 30 lb for seed 45L,500 srre t3'; ' e Surp (lb) i.e. 2 - 3 Raize i7.34 tons) 211,670 5eans: h 1.50 t3Tns) 109,350 Seed :. )ott n: k125 tons) 268,750 .rour1flv.r.:sI (226 VatQ") 50'750 U'ashi VIalu of M4arketab.le Surplus (Tsh) /9 ;iMe: (!. Tsh) 0.1/Lb 33,867 deans: at, Tsh 0.38/lb 41,553 :c,r,d ,ot,ttojs at Tsh 4681.6o/acre 168,560 :)Lnfld?uL:;: at official market prices assuming sale of best nuts 31,4j65 (Tslh 0.62/1b) Totald ash Value, (Tsh) 275 ,445 Fertilizer insecticide Sprayer Tools, etc. To ,al L esona IiLut /10 ,'in! ze 12,279 1,627 - 2,981 16,887 Z'a s 1.75 175 t.rounruits 35,660 28,851 6,508 618 71,637 total 4i,731 - - 88 J419, Average Liross Cash Income/Family (5 ; 350) (Tsh) 7P7 Ave.age ' Cash Income Per Family (5 - 6 s 350) (Tsh) 520 Tzable Page 2 iootnotea for table: /1 House Plot cropped under traditional practices witli unaizo, beans, cassava, bananas, sorghuvi, groundnuts and assorted vegetables cuch as okra. /2 Troditional ixiplies local seed, cultural practices and. no fertillz',v. /3 Stage I irlplie3 iciproved naize -eed grown as sole crop but incorpo±at."L-, lew other rcco:=lended practiceo. Beans would be late planted with this crop (April/';:a) ,atvzring in trio dry season. /x StPtCe II inmplles use of itmproved seed of naize and cotton, inproveci ad:-oromic pracUices aiid insecticide as appropriate. Beans would e3 g,'otn in aasociation with iala;Za as described above. 4/c Siage III Impplies use of fertilizer in addition to othler im!iprov-e practices descr'bed above where recommendations ex.st. /6 Toduction `i,:ires derived froin acre budgets. /7 See footnote A yield estinatocs can ronly be mnenin6fi.:lly Iiade fo. .,iae a td 0t.ans, at 30 and 210 lb, raspectively per acre. /. M ss-i on estimatet. '9 At rrices used in acre budgets, i.e. farmgate. 1.0 Cost of seasonal imputs derived from acre budgets. Seed costs deducted .under subsistence section as income foregone. Note: Despite national adcption of the i,ietric systen, actual usage Iin the reg&o- is still wit!: acres, pounds and feet. Accordingly, the calculati.cns ½.r this araiox hn;'e used these units. TANZANIA KIGOGI URUAL, DVELONaNT PROJECT c-e. ..ad Production uIld Up by C-ops and 'ea- Aer.ge B.350 Po.ily OiltIg. Vill.g. Y.-e 2 Vtllage Year 3~ Vill.g. Year 4 vollae Y-a S Villas. Y-ea 6 Vill.os Year 7 Villa.. Year '(illa. Y-a 9 VIll.ge Y(Ca 10 OtIlag Year 11 Vlae oa-1~2 Acres lb Arre lb Acres lb Arres lb Acre! lb Acr*s lb Acres lb Acres lb Acres lb Acrez lb Acres lb H-us Plot /I 350 105,000 350 105,000 350 105,000 350 105,000 350 103.000 350 103,000 330 105,000 350 103 000 350) /3 3501 350) stage 1 87.5 59,063 39 26.325 25 16,875 18 12,050 is 12,150 18 12,150 19 12,150 10 12.150 10) 18) 10) stage 11 175 170,S00 195 198.900 175 178.500 168 169,320 129 131,380 92 93,840 55 56.100 19 19,300 19) 19) 19) stage ITT 87.3 148.5 118t 197.200 130 2500 168 8,0 203 345.100 240 40,0 277 470.900 313 532,100 LI) ____ 313) ____ 313) ___ Tt1Total 991,313 700 527.425 700 555.375 709 568.670 700 593.830 700 818,990 700 644,150 700 864,630 6881,609 709,350 730,630 H...e Plot /I 350 70,000 350 70,000 330 70.000 350 70.000 330 70.000 350 70,000 350 70.000 350 70.000 350) /3 350) 350) Stage I ~~~~~ ~~~~~87.3 43,730 39 19,500 25 12,300 18 9,000 1s 9.000 is 9.00'0 19 9,000 18 9,000 10) 10) 10) Stage 11 ~~~~ ~~~~~1)5 87,300 195 97,3000 175 87.300 166 83.000 129 46,000 92 46,000 53 21,500 19 9.0 1)1)1) Stage 111 ~~~~~~~~~87.5 52i.500 116 6960 150 90.000 166 99,60 20) 121i500 240 1400 2 7 7 16,0 31 187.800 313) ____ 33 ___ 313) ___ Tona 700 203.750 700 256,600 700 260.000 700 261,600 700 265.300 700 269,000 700 277,700 700 276,300 204.589 293,127 301,920 SEE COTTOO ()Traditi-Ia /2 87.5 30.625 117 40,950 75 28.250 50 19.250 37 12.950 10 8,300 18) /3 18) 10) 181 18) (8 dnaurd 87.5 .250 233 233.000 2375 275.000 295 295.00 313 313.000 332 332.000 337) _____ 332) _____ 337 3321 _____ 332) ____ Tota 175 117,765 350 273.950 350 301.250 350 314,250 339 325,950 350 338,300 350) 340,449 350) 38,902 330) 369.670 350) 380,760 350) 392,182 (a) Tr-ditjo...l /2 - - 157.5 55.125 75 26,250r 35 12.250 35) /3 35) 35) 35) 35) 33, (b3 Ad--cod -- 17.5 J.2L250 100 70.000 140 9!,00 14-0) ___ 140) ____ 1401 ___ 140) _____ 140) _____ 140) - Tot.1- 175 67,373 175 96.250 175 110,250 113.558 11,2410.472 2406127,810 1161 /1 Mi..d -rppiog Tcoe cr house plot 9 /2 Bl1ok fare -Iti-ati oc /3 Fr-d-tr- .-as-md to gro at TY p-r -m.at,no)rfn',m, d -.ptitrdcduder tie pro Jeti -rr_pl.ed. Source: '(able Nate Sa-pite natiol1 adopct- of tIe yerr syte-. o,,aI,s I the region Ic still with a-e., pouds and f-, Scr.p',tecl,la)n n1, c...e. or -ud - sruits T': .. .... t~~~~~~8 T': !2UM~~~~~~~~~~~~~~~~~~~~41 20 ,808 8.0 8 924 A,88.88 330 289223252108812~ ~~~~~~~~~~~~~~~~~~~M. 011.8 022 4O7j...j,,,,,,,,,,,, 11.2 5.o 061. ..3 62885.s 5208..87 22885808.....085 ...*...2..11...........o.o2 41~~~~~~~~~~~~~~~~~~~~~~/ 0/ 8/ 2/ ......L88252 1.8118 2841301.s8.so-od,O. ,sl.3,..78...o6828 188286008l3001 ....02.881278020802, 19109 -402583188628.0-80228 4o8 o.28o.P.s02,S,28 42800288o38 .858 S.S128..885to 858 7o(o/1 2,241112.832.0 2.01101 2650 35.7 0,8332 2.00185 3,11 9.9 8,9 3,2 6880 2088 302j6010 28,2 0,0 2,21 8,2 2382 68883 883,338,32330 2803712 5,8 T,, 71 ., ..1..........0 02.3880 0,2 3,3 710 007 0.83 0.0 801 310 3,80 0.3 023 2030.00 2,2 2,806028,0 828 21.9 0808,0 1.33 38 0 00 01 2912,3 0 0 00 00.noO.0000000.o0100009090 0''0~no'.000.0000000nOnOOOonOoOoOOOOnA.06000.06..6n 10...06.0.16. ~0On0 000 ~ 0000 no nnoooono0000~0 06000090000 'o 000000O~0/i0n . . 0 . . . ..... ..o ............00016T60066000000000n0 0 ...... ...... ... ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ oo000n000n00000 ...... ...... ~ ~ nond0nnooo dnoooon 0o0 0 OooonnOO,0,o 000010006000n/q. 'noon 000000 069000 'non; 600000 000.066 0,0061~~~~~~~~~~~~~~~~~~~~~~~~~~~.... .. ooo'6 000'O 0000 0009 000'! 0,0'0 6060 6000 00,0 0600 0/0600006~~~~~~~~~~~~~~~~~~~~7M 0000 0000 000 mon 000; 010 oOo 000 000 li! Ono 0000 000! 010 0010 60o 000 gon 000000 000 OOO 90on 009 OOo 00'0 0SO-- -- non; 'nn, onos non oooo oo soo 00,1 6000 oq'o 0000 0016 0o69 009n 060 oOo ,.0'oo 060 000 0000 7900 066 6I, oV0 0-0 6660 no..-...- 0660 0060- 6600 6000 on00 0000 oo'oOOOnO 0000 0'00 000! nOnn- 60, 6600- 0060 0000 - Ooo 0000'9000 6090 -0909 0009 -Lon,000 014299 69C91,99099 910410 25 - - 20,626 611,950 19,4CC~~~~~~~~~~~Mn 01.0 4C0 9,5 9,59 9665 0, 60,409 9009 9904,975 9295 ,0617 5.9 1,0619,900 14,9 1,091,90 36,994 4,199,.. 09096 9,91 25 - - - - - - 20.C6C 611 690 20 600 014140 091 631 95C 26.456 946 605 29 245 904.47s 09.092 ~~~~~~~ 31.750 1.016.19-4 5 -1-009.309 220.426T 0 060.000 95 1,59 Si3O 9491.9169 559 ,0419 1,2 421,424,4 912,9 149 ,662031,90992,45 16,1 305,9 1219 ,2921 2,00 4.9,66 19129 ,9120 ,2942 199,9 25 - . 4,941 ,000,50 14,29 1,07,290 1,040 1060,00 10,16 1.099,19 19. 1,090505 15495 1.921050 15.952 1,010.30-19,2913996951 19921 1,21,959 15,919 1,999,90 11005- 10496.41 95 - - 4,99 1090550 14.69 ,26,90 1,09 1.60.11 3,4 1,00,25 9,90 4090599 19,99 .12.40 1,29 1,10,50.3,19 96,9169,59I4096914 92.14 0,56,91 291 - - - - - - 14 349 1 090 999 21 119 16 90 14 01 06.0 .,jI& 66 2.009129- _951 10095 1 45 10190 2 0 12010 1 I 9 95 0 ____ 99 1146 9940 2,9 ,5,9 970 50090 5!9 4!4o O,9 ,0,95 43 32520 5.97 3,-9,500 5790 C,599 M'3 51959 o: ~ ~3o3 1,95,90 4-990,146 931,0 25 99.901 ~~~~~~~~~~~~~~~1.0410 9,1 ,2,0 401 Cll99 9.9 .9,5 4 ,04,90 95,19 5,001,950 10917910 522,C 111,199 5 1949,49 015,523 5.546,05 119,9 ,1,0 7.9 46,709.21 25 24 4.-~~~~~~- ,90 94,95 60 9,615 ,2 9245 ,6 1.911.90 0,406 0,941,90 2,95 1.9,05 ,11 12900 5,9 ,9,0 - .W M,2956 20.59 ,55 90,2 15,603 j1,990,00 91.1 2,526,965 -0415 0,14,90 5041 3,704,9 9155 ,91,1 30,930 4.102,955 -4, 330_2,154,75 -4.28119.1 9.0 224.99 9~~~~0 609 9991 9109992999.. 9191,2 1969991 6919. 2991 0911991.479161 99 3OC 10, 211: 9649.91919144 290 14.1:1"61 :~I:' 9199~1 2,91916. 92929990 .29964191 99 90. 91.9919.9.0-, 2199910 09191291 IC 90, 9919199 1~ 910111911221091,, 99907.919D 09919. 41119102991y, tC91 ,.2611.9190. IC 40C~~~~~~~~~~~~~~~~~~~~~~ 499919199109 9.912 92919,91 991962.o::4 ~ ,~'~: . :' TANZANIA KIGCNA RURAL DEVFLOPMENT PROJECT Credit and Input Needs by Volume and Cost L INPUT Unit Cost Village Year 2 Village Year 3 Village Year 4 Village Year 5 Total Village Needs Unit Tah Volume Cost Tab Volume Cost Tah Volume Cost Tsh Volume Cost Tsh Volume Cost Tsh Diamsonium Phosphate (DAP) tons 1,636.80 (a) on Maize 44 72,019 132 216,057 264 432,115 396 648,172 836 1,368,363 (b) on Cotton 71 116.213 317 518,866 ( 669 1,095.019 1,020 1.b69.536 2,07/ 3 399 634 Sub-Total DAP 115 188,232 449 734,923 933 1,527,134 1,416 2,317,708 2,913 4,767,997 Calcium Ammonium Nitrate (CAN) tons 1,322.09 (a) on Maize 44 58,172 132 174,516 264 349,032 396 523,548 836 1,105,268 (b) on Cotton 71 93,868 317 419.102 669 884.478 1,020 1,348,532 2077 2745.980 Sub-Total CAN 115 152,040 449 593,618 933 1,233,510 1,416 1,872,080 2,913 3,851,248 DDT 57 Dust kg 2.45 21,000 51,450 56,000 137,200 91,000 222,950 126,000 308,700 294,000 720,300 ULV Sprav Sets No. 162.80 350 56,980 1,225 199,430 2,100 341,880 2,975 484,330 6,650 1,082,620 Battery Sets No. 28.88 350 10,108 1,575 45,486 3,325 96,026 5,075 146,566 10,325 298,186 Endosulfan liter 16.30 10,500 171,150 47,250 770,175 99,750 1,625,925 152,250 2,481,675 309,750 5,048,925 Single Superphosphate tons 946.24 - - - - - - 100 94,624 100 94,624 Credit Needs by Year 629,960 2,480,832 5,047,425 7,705,683 Incremental Credit Requirement 629,960 1,850,872 2,566,593 2,658,258 /1 Derived from Tables 1 through 5. V- rANZANIA KItOMA RURAL DEVELOPMENT PROJECT Volume and Cost of Input Needs for Demonstrationsi/ Total Village Year 2 Villaje Year 3 Village Year 4 Village Year 5 Demonstrations Item n= Unit Cost Vol. C= y Vol. Cost Vol. Cost Vol. Cost Vol. c Q (Tsh) (Tsh) (Tsh) (Tsh) - (Tsh) (TCh DAP ton 1,636.80 5 8,184 7 11,458 7 11,458 7 11,458 26 42,558 CAN ton 1,322.09 5 6,610 7 9,255 7 9,255 7 9,255 26 34,375 5% DDT kg 2.45 400 980 500 1,225 500 1,225 500 1,225 1,900 4,655 ULV Sprayer ea. 162.80 50 8,140 60 9,768 60 9,768 60 9,768 230 37,4" Battery Sets ea. 28.88 50 1,444 60 1,733 60 1,733 60 1,733 230 6,613 Endosulfan liter 16.30 1,080 17,604 1,200 19,560 1,200 19,560 1,200 19,560 4,680 76,284 SSP ton 946.24 - - - - 5 4,731 7 6,624 12 11,355 Total 42,962 52,999 57,730 59,623 2T11 1/ Assumes an overall coverage of five acres of demonstrations of recommended practices for each project crop per village. C g (D ( om ANNEX 16 Page 1 TANZANIA KIGOMA RURAL DEVELOPMENT PROJECT Economic Analysis of the Project Introduction 1. Conceptually, the ujamaa village may be thought of as a system through which "inputs" -- most clearly in the form of technical knowledge, accounting and management systems on one side, and physical inputs like human labor, fertilizer and insecticide on the other -- are, together with credit, combined to produce additional output, accruing largely in the form of cash income for villagers. 2. The project covers the investment requirements of 135 ujamaa villages at different stages of development. Hence the investment needed in particular villages during the disbursement period varies from social infrastructure and set-up costs in villages just becoming established, to mainly agricultural credit and services for already settled villages. For villages at the begin- ning stage, the rate of return to project investment would depend critically on government willingness to complete the program and provide the necessary services beyond the disbursement period. 3. Under these conditions, the assessment of the viability of the proposed project has been based on a model of the costs and benefits associated with investments in a typical or representative village, beginning from the establishment of the village and taking a horizon for estimation purposes of 15 years. The analysis is based on a village of 350 families which for simpli- city is assumed to remain constant in size over the 15 year horizon. In addition to the economic rate of return, measuring the worth of the project to Tanzania as a whole, there is considerable interest in the impact on the Kigoma villagers who are the direct beneficiaries. The analysis begins from an assessment of the latter, economy-wide considerations being dealt with thereafter. The Impact on Village Income and Employment 4. Incremental production benefits to villages under the project are calculated from the farm budgets (Annex 15), including additional production for subsistence as well as marketed surplus. Benefits are net of recurrent input costs, in the form of fertilizer, insecticides sprayers and interest on seasonal credit. The traditional or previllage farm is assumed to account for the same cultivated area per family as under the project (3.5 acres), of which 2.4 acres would be required for subsistence needs. The remaining 1.1 t acres is valued on the basis of the traditional maize/bean intercropping (see Annex 15). It is assumed that without the project, production per family would have remained at the same level over the project horizon. ANNEX 16 Page 2 5. Faml1y agricultural income would increase from an estimated T Sh 580 under traditional settlement and farming patterns to T Sh 1,130 per family during the eighth year following settlement of the ujamaa village. Total agricultural income would thus increase by 95Z, while cash income (income surplus to subsistence needs) would increase more than threefold. A sall allowance for off-farm productive activities (casual wage labor, beekeeping, charcoal-burning, etc.) should be made in arriving at total family income. It is unlikely that such an allowance would raise income to more than the equivalent of US$20 per capita under existing (pre-project) circwnstances. Agricultural benefits under the project would therefore increase per capita incomes from US$20 per capita to US$35 per capita. 6. Any direct benefits from social services provided in the village -- i.e., education, water supply and health care facilities-would be additional to expected income from agriculture. Valuation of these benefits is however difficult. There does appear to be a strong demand for these services, and it is argued elsewhere that without the early prospect of such services becoming available, it is doubtful that villagers could be attracted into village settlements, or would remain through the initial period of settlement and land clearance, particularly in the tsetse fly areas. But the apparently strong demand for services on the part of villagers is not unaffected by the consideration that under present financing arrangements the benefit-cost ratio to them is practically infinite, since capital and recurrent costs are entirely borne by government. 7. The services are expensive to provide: capital costs are approx- imately T Sh 350,000 per village, and annual recurrent costs, approximately T Sh 80,000 (including an allowance for overhead supervision and management); allowing for capital recovery over a 15 year period and interest at 10%, the full annual capital and recurrent cost is T Sh 130,000, i.e., T Sh 370 per family or T Sh 75 per capita. The full cost of providing these services in relation to village income from production would thus be 29%. If direct benefits were equal to full costs, "income" from social services would account for 23% of total income, If direct benefits were only equal to recurrent operating costs, "income" from social services would be 15% of total income. On the basis of this assumption, the project would double total village income to US$45. 1/ 8. On the traditional farm, labor input per family per annum is calcu- lated at about 240 days; by Year 8 under the project, this would increase (because of the greater labor intensity per acre under improved husbandry) to about 300 days per annum (see Annex 3). But in contrast to the traditional farming cycle, heavily biased towards a maize/bean inter-cropping, labor demands under the project are spread more evenly over the calendar year (see Annex 3, Table 2). Thus despite higher overall usage of labor, labor 1/ Allowing a small margin in the pre-project situation for limited access to health and education facilities. ANNEX 16 Page 3 usage in the peak months of activity is very likely to be about the same under the project as for the traditional farm of equivalent size. This is an important consideration, since it is often the case under traditional cropping patterns and husbandry, that scarcity of family labor during the season of peak activity effectively checks overall expansion that would otherwise be possible given plentiful labor during the slack season. 9. Another influence on the cost benefit analysis and income/employment impact at village level concerns the cropping mix over the life of the project. The assumption that the cropping mix remains constant over the horizon of the project is conservative, since it denies to the village benefits that may be obtainable by modifying the cropping pattern. For example, on the basis of the structure of prices and costs assumed for the project, it could be expected, as a village gains in confidence and sophistication of operations, that a switch towards cotton would occur in respect of marketed outputs. Such a switch could increase net family income by over T Sh 100 at full develop- ment without increasing labor input. Market Prospects for Project Outputs 10. Important considerations affecting future market prospects are: Mi) trends in world and domestic markets; (ii) government protection and promotion policies; and (iii) trends in costs of collection, marketing and distribution of inputs and outputs. But because a substantial share of village production is not marketed beyond the village, the effects of market prospects are less than would be the case otherwise. At full development, over 60% of production of beans and maize is not marketed, and over 40% of production of groundnuts. 11. Prospects for cotton are by far the most important market considera- tion since cotton sales would account for more than 60% of projected agricul- tural cash income at full development. About 85% of the crop is exported; the 1973 export prices after accounting for prices to producers, distribution and marketing costs, were enough to provide for a substantial surplus (partly taken by government in the form of a 10% ad valorem export tax). 12. DDmestic production of maize averaged 710,000 tons per annum over the 5 year period 1968-72. Only about one quarter of the crop is marketed, but commercial sales appear to be growing quite rapidly, possibly at 5% or 10,000 tons per annum. The cash market is supplied through unofficial channels as well as by the authorized parastatal agency and sales through the parastatal probably account for only about one half of the total. Imports, also controlled by the parastatal, have become increasingly important totalling 200,000 tons over the four years ending 1972 (including 135,000 tons in 1972). These were partly offset by marginal export sales totalling 78,000 tons over the same period. It seems likely that the parastatal agency would need to sell an additional 150,000 tons of domestically produced maize in ten years' time, to displace marginal imports and accomdate the growing cash market. Under the project, by Year 10, 15,000 tons of maize should be marketed or about 10% of this requirement. ANNEX 16 Page 4 13. At T Sh 1,100 per ton in 1973 for imports, the comparison with the Kigoma supply price of about T Sh 650 per ton is very favorable. But on the basis of self-sufficiency, other regions in Tanzania may be able to supply additional maize to the major Dar es Salaam market at costs marginally below those for Kigoma. This is difficult to assess because full costs are not properly accounted for by the various unions and societies; however, high transport charges within Kigoma region, added to rail freight of T Sh 75 per ton Kigoma-Dar es Salaam, tends to favor some other regions. By the same token of geographical isolation however, Kigoma producers have a decided advantage with respect to prospective additional sales within the region. The (small) urban population is growing quite rapidly, and with the secondary expansion of regional income created by the project itself, additional sales within the region might account for a third of the marketable surplus of maize produced under the project by Year 10. To the extent that marginal supplies for other markets are at a cost disadvantage relative to supplies from other regions, there is a case for switching production to other crops; however, it is probably premature to assume that additional supplies from other regions will be fact materialize. 14. Supplies of groundnuts and beans are at the present time almost entirely absorbed by the domestic market, despite reasonably favorable export prospects. Growth of the domestic market may absorb 80,000 tons in the case of pulses over the next ten years and 20,000 tons in the case of ground- nuts. Incremental marketed production by Year 10 of-the project would be 6,000 tons of beans and 4,000 tons of groundnuts--a relatively small proportion of the projected incremental requirements of the Tanzanian market. As in the case of maize, a large share of marketed produce is now sold outside the official marketing channel and this seems likely to continue. 15. With marketing and dis'ribution costs already high, ranging over 80% of the producer price in thS case of low unit value maize, a potentially serious problem would arise if further substantial increases in internal marketing and distribution costs occur. Within the region, where the bulk of these costs are incurred, provision has been included under the project for a reorganization of the Regional Union, accompanied by provision of technical assistance, equipment and other facilities to ensure cost-effective operations. However, the major non-regional cost for marketed crops is rail freight and some increase from current low levels may be expected within the horizon of the project. 16. The likely behavior of internal prices and costs generally are in part determined by overall prospects for the Tanzanian economy. So far as can be predicted at this point in time, the most important longer term consideration here would appear to be Tanzania's extremely weak energy y position and lack of mineral resources. Given a balance of payments already fairly heavily buttressed with licensing restrictions on the import side, the likelihood is for greater incentives to be provided by government, favoring the production of traded commodities (exports or import substitutes) such as maize and cotton to be produced under the project. ANNEX 16 Page 5 Economic Rate of Return Estimates 17. Prices for inputs and outputs reflect the situation as at January 1974, with adjustment for expected trends on the basis of commodity pri ce forecasts relating to 1980 and 1985; a shadow rate of exchange of T Sh 10 Lo US$1:00 in respect of foreign exchange receipts and outlays was used; and domestic costs are calculated net of indirect taxes and subsidies. On the basis of these assumptions, the economic rate of return to the project would be 22%. 18. There are indications that social service expenditures should be treated as a cost of production, necessarily incurred because the benefits of additional income through improved technical practices and marketing facilities in agriculture cannot be achieved except where people live in villages. There are great difficulties in reaching scattered homesteads, as is testified by the dismal developmental record of Kigoma region during the long scattered homestead period. Two possible assumptions would seem to be: (a) that social services just 'repay' their full operating and maintenance costs through direct benefits to villagers (i.e., repay recurrent and capital cost at zero interest); and (b) that direct benefits should bear some plausible relationship to other forms of income. Under assumption (a) the overall rate of return would be 18%. If direct benefits just repay recurrent costs incurred in providing social services they would amount to 15% of total village income at full development (see para 7); on this assumption, the overall rate of return would be 12%. Neither of these assumptions is very satisfactory; on the other hand, nor is an assumption that production benefits would be unaffected by the provision of these services. 19. For the rate of return, the least favorable type of outcome would be where all inputs, social and economic, were supplied on schedule, but yields are less than predicted. If gross yields are, on average, 10% lower than expected the rate of return would be 17% rather than 22%; with a 20% loss, the return would be 5%, and if a 25% loss occurs the project becomes economically worthless. A second, and not unlikely, type of failure would be where certain villages fail entirely and fall apart, while others perform according to expectations. This type of failure is, however, likely to be much less expensive than the first type (because some inputs are saved in the process) though it could be more disruptive socially. Thus, if 30% of project villages fail, the rate of return to the project could still be as hiigh as 15%. Government Recurrent Expenditures and their Recovery 20. Recurrent expenses of government in project villages at full development would be about T Sh 80,000 per annum. This includes the operation *1 ~!iiatenance of a primary school (at T Sh 50,000 per annum), a village water supply (T Sh 6,500 per annum) and a share of rural medical facilities (T Sli 10,000 per annum), together with the salary of a village agricultural assistant and other minor items. Supervision and overhead staff servicing ANNEX 16 Page 6 for these facilities at district and regional level might add an additional 20% to 25% to the in-village costs, bringing the total to about T Sh 100,000 per annum. The 135 villages included under the project therefore, commit government to expenditures of about T Sh 13.5 million per annum in the region. The total recurrent fbudget for Kigoma region for 1973/74 was about T Sh 26 million. 21. It is difficult to estimate the recurrent costs of servicing other ujamaa village development schemes, but it is unlikely that these would be below those estimated for Kigoma region. Government policy is for Tanzania's entire rural population of about 13 million people to become resident in ujamaa villages and to enjoy free water, primary education and basic health services. Numbers reportedly living in ujamaa villages might now number some 2 million and are increasing rapidly. Some of those in ujamaa villages now nave schools, water supplies and health services, as do some resident in more developed areas not in villages. For a majority, however, perhaps 10 of the 13 million total, the situation would not be unlike that of Kigoma region, calling for extensive social and economic investments likely to generate substantial demands on the recurrent budget. At Kigoma levels of cost, supplying a population of 10 million with currently prescribed levels of services would generate a recurrent cost outlay of roughly T Sh 570 million per annum, or more than US$80 million. Total government recurrent expenditures for 1973/74 would probably be less than US$300 million of which less than one half would be for social and economic services. 22. The increase in current expenditure required for the provision of social services associated with the ujamaa program is therefore a large one. Perhaps more significant, assuming that the program goes ahead rapidly, a correspondingly large increase in tax revenues (government savings being already very small) would be required, which revenues already approach 20% of domestic product and include relatively heavy taxation of the small modern industrial and urban sector. 23. To a limited extent, development of village agriculture may provide additional tax revenue to government directly, so partly offsetting the re- current costs. In Kigoma region, for example, a major crop would be cotton which currently attracts an export tax of 10% ad valorem, and roughly T Sh 30,000 per annum might be recovered through this and other tax proceeds. The amount of tax generated through this route, however, depends on the crops grown since some crops (non-food exports) are much more easily taxed than others. Assuming that the program will go ahead at a rapid pace, it is hard to avoid the conclusion that some form of additional taxation of villages, to support the services being provided to them is ultimately required. The issue of cost recovery would be an important subject of study for the proposed technical assistance team (Annex 10). KIGOMA RURAL DEVELOPMENT PROJECT OVERALL GOVERNMENT ORGANIZATION TEE PREIDENT _ MINISTRY OF [LM OF |DEVELOPMENT TREASURY LI | PRYOF FP LANRIN M NL MINISTER OF MINISTRY OF MINISTRY OF MINISTRY OF minuto, of State lot IITRO IITYO MINISTRY OF OF IN STER MINISTRY OF~MIISTY O AGRICULTURE COMMERCE AND NAiNAsEAT NDCNSRCTO AT"ER ANDLN NATURAL RESOUR- m ent Authoriv 1( =y REGIO AL NoLticAl A t _MN REG ON Y OF DUERVEL OPMENT IAL COMRS SIONER U | PLANNING j M FINANCE ENDEELLE I INUSR I 3 EI3 I j< DEEOMN REORE Development Manage- ment Authority DISTRICT ~~~~~~~~~~~~Political AuthorityAE DEVELOPENT co MssoF.rP DIRECTOR J? AGRICULTURE COMMERCE AND EDUCATION HEALTH CONSTRUCTION WATER LAND RERIONAL INDUSTRY Jo I I IrDEVELOPMENT CO SSIONRES De.elopm.t Ma..ge- I s ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~~~~~~~~~~~Wfy an-5E IBRD 10941 ',,<' ' ake 5.0 --.i 03 r --0'r<.yfS¢rlE¢S 32° V'ctoroa O!, -; MARCH 1974 AWASJ4tA r k < . - K < N Y A, . -- t ,j to~~~~~~~~~~~~~~~~~~~~~~~~~~~A, PROJECT Kigoma2r AREA 3 aee TANZANIA Tong,, e'ka \ eAop = I oTbr "'''~~~~~~a es \Salaam\ ,.K; ::.:. ,\ ZA NI La'ke .;4E. - - <: \< < ~~~~~~Kigoma Rural Development Project ,.-,,, . -,_t,.,,>,, ! - ~PROJECT AREA -~ ~~~ * 03'u roject area- Kigomo Region - . /eyq ?,.;'- >3q-t-\ ---R~~~~~~~~~~~~~~egional boundaries z;~~~~~ . D ...\.......;.......Dstrict boundaries - i-tL t * \ < \ A District headquarters / MMain roads6° Kigoma~~~~~~~~~~~~~~~~~~~~~Rvr ,,7., ! __ International oundaories -2 - < \>X= ) = Fly front, 1973 -- 5>. - .1 \ i < < < - Tsetse infested area ....... Sleingtsict ondressedmcaa . - ..> t-t/. ' -d';.,, ,/'"'', ,-/0 20 40 60 80 100 '.' < < ,. . ~ ~~\ . / K ILO ER \- 0 ,i,r, 40 ,q60 / M0;32 ______________________________________ ~~~~~~~~IBRD 10942 Victoria ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ 'ro~b f~MARCH 1974 PR'OJECT- AREA TANZAN I A 1 Lake TO KolhaMa take 7,A& AB I A., Nyasa 41 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~4- * Mohyovu- / ~6 Tanganfly'4a K'g:j o mtly Rnin able Developm Riers Pojc __ I. Popu~~~~~~~~~~~~~~~~~~~Latonehr Rest' -- negnional boundaries Miom~Uban populatidn 100 20e40le 300, MILES 3270CD rb n op laio