Tanzania - National Maize Project

CIRCULATING COPY Report No. 897a-TA JMg10 E RETURNED TO REPORTS DESK United Republic of Tanzania Appraisal of the FILE COPY National Maize Project December 8, 1975 General Agriculture Division Eastern Africa Regional Office Not for Public Use Document of the World Bank This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization  CURRENCY EQUIVALENTS 1/ SDR 1.00 - Tanzania Shillings (TShs.) 9.66 SDR 1.00 = US $1.20 US $1.00 = Tanzania Shillings (TShs.) 8.05 TShs. 1.00 = US $0.12 GOVERNNENT AND PARASTATALS FISCAL YEAR July 1 - June 30 WEIGHTS AND MEASURES 1 hectare (ha) = 10,000 m = 2.47 acres 1 kilometer (km) = 0.62 miles 1 sq. kilometer (km2) 0.39 sq. miles = 100 ha. 1 kilogram (kg) - 2.20 pounds 1 liter (1) - 0.26 U.S. gallons 1,000 kg = 1 metric ton 0.98 long tons 1/ In October, 1975, the three East African Community currencies were pegged to the value of the Special Drawing Rights (SDR) of the International Monetary Fund. An SDR/US dollar conversion rate of 1.2 has been assumed for purposes of this appraisal, although the actual rate is floating.  ABBREVIATIONS ABEDIA = Arab Bank for Economic Development in Africa AFO = Assistant Field Officer ARTI - Agricultural Research and Training Institute AVI = Audio Visual Institute BRALUP Bureau of Resource Assessment and Land Use Planning, University of Dar-es-Salaam CAN Calcium Ammonium Nitrate CIMMYT = International Maize and Wheat Improvement Center, Mexico CUT Cooperative Union of Tanganyika DADO District Agricultural Development Officer DAP = Diammonium Phosphate DDD District Development Director DMPS = District Maize Project Supervisor EAC/MI East African Community, Management Institute FA Field Assistant FAO Food and Agriculture Organization of the United Nations GAPEX General Agricultural Products Export Corporation IITA = International Institute of Tropical Agriculture, Nigeria KILIMO = Ministry of Agriculture MDB = Marketing Development Bureau NADP National Agricultural Development Project NAPB = National Agricultural Products Board NBC = National Bank of Commerce NMC = National Milling Corporation NIMP = National Maize Project NPK = Compound Fertilizer PMO = Office of the Prime Minister and Second Vice-President PSU = Project Servicing Unit RADO = Regional Agricultural Development Officer RDD Regional Development Director RMPC = Regional Maize Product Coordinator TANU = Tanganyika African National Union TAT = Tobacco Authority of Tanzania TCA = Tanzania Cotton Authority TFA = Tanganyika Farmers Association TRDB = Tanzania Rural Development Bank TSP Triple Superphosphate UNDP = United National Development Program USAID = United States Agency for International Development VMT = Village Management Technician   TANZANIA NATIONAL MAIZE PROJECT TABLE OF CONTENTS Page No. SUMMARY AND CONCLUSIONS . ........................ . i - iii I. INTRODUCTION ...................... 1..... so ..1 I1. BACKGROUND ............................. . 2 A. General ...................................... 2 B. Regional Government and Administration ....... 2 C. Villagisation and Land Tenure ................ 3 D. The Agricultural Sector ....... ............... 4 E. The Food Situation ..... ...... ............ . 7 F. The Maize Production Program ................. 7 III. THE PROJECT AREA ..... .. ......... .. .... 8 A. Location . ........... 8 B. Land Use and Types of Farming ................ 8 C. Rural Infrastructure ...% ...... ............... 9 IV. THE PROJECT ................... 11 A. Brief Description .. ........................... 11 B. Detailed Features ....o............. ......... . 12 C. Project Costs .. .............................. 17 D. Financing ............. .. ........ ............ 19 E. Procurement and Disbursement ....o........... 20 F. Accounts and Audit ...... ..................... 21 G. Ecological and External Effects .. ............ 21 V. ORGANIZATION AND MANAGEMENT ... o .... o ............. 22 VI. PRODUCTION, MARKETING, FARMERS' BENEFITS, EFFECTS ON GOVERNMENT BUDGETS ..... ............. 25 VII. ECONOMIC BENEFITS AND JUSTIFICATION .... .......... 27 VIII. AGREEMENTS REACHED AND RECOMMENDATION ........... 28 This report is based on the findings of a joint preparation- appraisal mission which made visits to Tanzania during March, April, May and June 1975. The mission was composed of A. Denness, R. H. Clough, J. F. A. Russell, F. Stubenitsky and G. Ablasser (IDA); and P. Grosjean and A. M. Wilson (Consultants).  ANNEXES 1. Ujamaa and the Cooperative Sector 2. Extension and Training 3. Technology and Potential for Maize Production 4. The National Milling Corporation (NMC) 5. The Food Crops Sector 6. The National Maize Program 7. Terms of Reference for Project Staff and Consultants 8. Project Cost Tables 9. The Tanzania Rural Development Bank 10. Maize Markets and Prices 11. Farmer and Government Budgets and Benefits 12. Schedule of Estimated Disbursement of IDA Funds 13. Economic Analysis ('harts World Bank-8466(R) Government Organization World Bank-8522(R) The Ministry of Agriculture World Bank-15244 Implementation Schedule Maps IBRD1O?99 Agricultural Research and Training Institutions IBRD11872 Regions and Districts IBRD2803R Mean Annual Rainfall IBRD28;4R Potential Land Use  TANZANIA NATIONAL MAIZE PROJECT SUMMARY AND CONCLUSIONS (i) Tanzania places high priority on agricultural and rural develop- ment to close the gap between urban incomes and those of 90% of the pop- ulation who live in rural areas and work mainly in traditional, near-sub- sistence agriculture. The development of collective and self-help activi- ties through villages is emphasized. Because of poor rainfall and the dis- ruptive effects of villagization, food production, especially the main staple maize, fell drastically in 1973 and 1974. Grain imports costing about US$160 million were required during 1974 and 1975. The Project, over a four-year development period, would assist Tanzania in increasing maize production to replace imports, and would lay the foundation for the long-term develop- ment of food crop production through improved extension services and the introduction to farmers of modern inputs and better cultural practices. (ii) The Project which would improve and expand the maize production program started by Government in 1973 would provide input and exten- sion packages designed to increase maize production in about 950 villages located in the best maize growing areas in 13 regions of Tanzania. It would provide villages with subsidized maize production inputs and support maize production demonstrations and research trials to further improve maize pro- duction practices throughout the Project area. Additional trucks would be provided to facilitate transport of maize and farm inputs within the 13 re- gions. Improved storage for maize and farm inputs would be established within Project villages and in regional centers. Extension services would be strengthened through staff training, provision of adequate transport and extension aids, mobile film units and additional support for the Ministry of Agriculture's newspaper. (iii) A Project Servicing Unit would be established within the Ministry of Agriculture. This unit would be responsible for input procurement and disbursement, financial control, monitoring Project progress, preparing annual work programs, and provision of technical assistance to regional authorities responsible for Project implementation. A Maize Production Inputs Account would be established in the Ministry of Agriculture, which would be used to finance the purchase of maize production inputs and to subsid- ize their use when necessary. Initially, the subsidy on maize production inputs would be 75%, but Government plans to progressively reduce this as efficiency of input use and village incomes increase. The Project would also provide consultancy services to prepare investment proposals for ex- panding and modernizing national grain storage and milling facilities; establishing a national grain reserve; and for the preparation of a national agricultural development program.  - ii - (iv) The Project would be implemented over a four-^rear period by regional authorities, who would be responsible for selecting the best villages within their regions, preparing village work programs, and ensuring the timely delivery of inputs and their efficient use. The Project design incorporates a flexible approach to Project implementation whereby Project villages, and storage and transport requirements would be determined during implementation and incorporated in regional annual work programs. These programs would be approved by the Project Servicing Unit and incorporated into an annual national work program which would require the approval of IDA. Policy issues would be considered by a Project Steering Committee comprising representatives from the Ministries of Agriculture, Finance and Planning, the Prime Minister's Office, and the Commissioner of Cooperatives. (v) The Project would cost US$38.1 million; its foreign exchange component would be US$24.4 million. The proposed IDA Credit of US$18.0 million would cover 47% of the Project's total cost. Thirteen percent of Project costs would be covered by a loan from the Arab Bank for Economic Development in Africa. The remaining 40 percent of Project costs would be covered by farmers (11%) and the Government of Tanzania (29%). Government's share includes the subsidy portion of the cost of farm input requirements for the first year (US$6.4 million or 17% of Project costs) which would be financed from the proceeds of an existing USAID loan. Funds provided under the Project for construction of storage and purchase of trucks would be channelled through the Tanzania Rural Development Bank (TRDB). TRDB would borrow funds from Government at 4 percent interest and onlend at 8-1/2 percent. The terms would be set out in a subsidiary Loan agreement to be agreed by IDA. (vi) Requirements for fertilizers and insecticides from the second year onwards, and Project requirements for vehicles, village store construction materials and the contracts for the construction of regional and NMC storage would be procured through internat:ional competitive bidding, in accordance with IDA guidelines. Farm input requirements for the first year would be procured through the normal Goverrment of Tanzania procedures. Local manufacturers would be allowed a pireference margin of 15 percent or the existing rate of duty, whichever is lower; local civil works contractors would be allowed a preference margin of 7-1/2 percent. Orders for small items such as office equipment costing less than US$120,000 and for maize seed would be placed in accordance with normal Government procedures. (vii) Project incremental maize production would be about 195,000 tons annually at full development after 1982. At the 1975 producer price of TShs. 0.75/kg this would be worth TShs. 146 million (US$18.1 million). 'lost Project produced maize would replace imports or be absorbed into the proposed national grain reserve. Small export surpluses would be produced only in good seasons and absorbed in border trade. The Project would, therefore, not generate significant export earnings, although the saving in foreign ex- change expenditures through reducedl maize imports would be substantial. The incomes of about 380,000 participat:ing families who are among the poorer sections of the rural community in Tanzania, would be increased. At full development the average increase in family income has been estimated at US$42 annually or about 20% above current levels.  - iii - (viii) The economic rate of return of the Project is estimated to be 37 percent. The rate of return is high, partly because the Project would produce maize as a substitute for expensive imports, and also because the Project would concentrate on improving the efficiency with which resources presently devoted to maize production are used. (ix) The Project has considerable risks on account of the unproven pro- ductive nature of the villages where the Project would be implemented, and uncertainties regarding farmers' willingness to purchase and use fertilizers and other farm inputs. Quality of management is also critical for the success of the Project. However, the sensitivity analysis indicates that even with the very adverse assumption that incremental output were 30 percent below expectation, the Project would still yield an economic rate of return of 16 percent. (x) Nine IDA Credits totalling US$91.1 million and two IBRD loans for US$30 million have been approved so far for agricultural and rural development projects in Tanzania. These include projects concerned with credit, livestock, tea, tobacco, cotton, sugar, cashewnuts and one regional development project. Implementation of those projects has generally been slow. (xi) The Project is suitable for an IDA Credit of US$18.0 million on standard terms.   TANZANIA NATIONAL MAIZE PROJECT 1. INTRODUCTION 1.01 Tanzania's rural development strategy emphasizes the need to increase agricultural production by mobilizing the productive capacity, and improving the living standards of the estimated 13 million people who live in the rural areas, and who are almost wholly engaged in traditional, near subsistence agriculture. To implement this strategy, Government has taken three important steps: the regional decentralization of its admin- istrative organization; improved coordination between Government and the Tanganyika African National Union (TANU), Tanzania's only officially recog- nized party; and the creation of development or ujamaa villages (para. 2.08). Although problems have arisen, especially in implementing decentralization and villagisation policies, and the end result of the latter remains un- clear, rural people are now able to participate more fully in the decision making process, and are being more effectively mobilized for food production and other aspects of national development. Unfortunately, Government's in- tensification of its villagization program coincided with two successive years of poor rainfall. This situation, coupled with the inevitable dis- ruption of production caused by villagization, necessitated the import of large quantities of grain, particularly maize. The proposed Project would assist Tanzania to increase its maize production and become less dependent on imports. 1.02 Previous Bank Group assistance to the agriculture sector in Tanzania has included IDA Credits totalling US$91.1 million and IBRD Loans for US$30 million for nine agricultural and one rural development project. Agricultural projects include one for agricultural credit, three for live- stock, and one each for tobacco, tea, cotton, sugar and cashewnuts. A rural development project has been approved for the Kigoma region. Credits have also been provided for agricultural training (1971), and feeder road development (1972). 1.03 Project implementation has been slow in most cases, but there has been some improvement recently. In general, management, organization and staffing have been continuing problems. All small-holder projects such as the cotton, tobacco, tea and rural development projects have been affected by the sudden acceleration of the villagization program in 1973 and 1974, requiring a major reorientation of these projects and causing significant delays. Under the completed credit project, a need was found for improved supervision of loan records and input distribution. The first livestock project was reasonably successful, but the second is progressing more slowly than anticipated due to delays in preparation and approval of ranch plans, cost overruns in construction of slaughterhouses, and changes in project management. The tea project has encountered difficulties due to  -2- inadequate services and lack of effective management control. The Geita cotton project is experiencing problems due to villagization, national emphasis on maize rather than cotton, and weak management. The Kilombero sugar project is proceeding well, although some delays are expected due to slow delivery of construction material and equipment. The cashew project and the Kigoma rural development project have only recently become effective. 1.04 The proposed Project was identified in the Bank's 1974 Agricul- tural and Rural Development Sector Study, and the credit application was prepared by the Government, assisted by the IBRD Resident Mission in East Africa (RMEA). This report is based on the findings of joint preparation- appraisal missions which made visits to Tanzania during March-June 1975. The missions were composed of A. Denness, R. H. Clough, J. F. A. Russell, F. Stubenitsky and G. Ablasser, (IDA) and P. Grosjean and A. M. Wilson (Consultants). II. BACKGROUND A. General 2 2.01 Tanzania with a total area of about 880,000 km is2sparsely pop- ulated with an average population density of about 16 per km . The pop- ulation of 14.3 million is increasing at about 3% a year. Average per capita GNP for 1973 is estimated at. US$110 per annum, but in many rural areas the per capita income does not exceed US$40 per year. Over 90 per- cent of the people are dependent on agriculture and 40 percent of GNP is derived from this sector. The real growth rate of the economy was about 4-1/2% per annum between 1969 and 1974. Actual growth has been behind the Second Five-Year Plan (1969-1974) targets, mainly due to failure of the main agricultural crops to reach plan projections and recent adverse changes in the country's terms of trade. B. Regional Government and Administration 2.02 A major change in Government structure occurred in 1972, with the decentralization of a wide range of governmental responsibilities (Chart World Bank-8466(R)). In particular, regions were authorized to plan and im- plement their own development progreams and allowed a high degree of autonomy in regional administration. 2.03 The regional political hea.d is the Regional Commissioner, with the rank of cabinet minister. He is a member of the National Assembly and TANU's National Executive Committee. His counterpart at the district level is the Area Commissioner who is the TANU district secretary, chairman of the District Development and Planning Committee and a member of the District Development Council. All commissioners are appointed by the President.  - 3 - 2.04 The regional administration is headed by the Regional Develop- ment Director (RDD) who is assisted by a Regional Planning Officer and the heads of the ten regional functional departments. Heads of these depart- ments report directly to the RDD, who in turn reports to central government through the Prime Minister's Office (PMO). At the district level the admin- istration is headed by a District Development Director (DDD). He is assisted by a District Planning Officer and ten district functional managers. All important development decisions are made by a District Development and Plan- ning Committee, chaired by the Area Commissioner, and have to be approved by the District Development Council, chaired by the TANU district chairman. While field staff are answerable to the DDD through the heads of their res- pective departments, they receive technical guidance from the staff at the regional level and through them from the Ministry of Agriculture (KILIMO). The organizational structure of KILIMO is shown in Chart World Bank-8522(R). 2.05 The new administrative arrangements ensure improved interaction between the political structure and the civil service, and enable villagers, through che basic TANU unit (the ten-family cell) to play a role in deter- mining the development process. However, problems exist. Planned staff strengths have not been achieved, particularly at the lowest level, and staff quality remains low. Project implementation capacity varies con- siderably between regions and districts, and regional performance is affec- ted by the interpretation of and relative emphasis given to aspects of national policy by regional officials. C. Villagization and Land Tenure Organization of Production 2.06 Until the advent of Government's villagization program most pro- duction was concentrated on smallholdings which were cultivated by hand and produced family subsistence and cash crops. Most holdings (over 80%) were less than 2 ha, and only 2.5% were 5 ha or larger. Plots tended to be irregular but fragmentation was a serious problem only in the densely populated areas of Kilimanjaro, Arusha and Mbeya regions. Ujamaa and Villagization 2.07 The Arusha Declaration of 1967 and the TANU guidelines of 1971 reaffirmed Tanzania's commitment to a socialist society and identified the ujamaa village as a means of effecting self-reliance and a community ap- proach to development. In addition to political advantages the ujamaa village was seen as the vehicle for providing economies of scale in de- livering inputs and social services, which could improve the lives of rural people. The ultimate objective is that ujamaa villages should be- come multi-purpose cooperative societies with authority to own land and to borrow. Government has provided considerable assistance to newly de- veloped ujamaa villages, mainly by transporting villagers' effects, providing  - 4 - building materials for houses, water supply schemes and other social services; and by encouraging the development of productive activities. (Annex 1) 2.08 By early 1975 Government estimated that about 65% of the rural population, almost 9 million people, were living in ujamaa-type villages throughout Tanzania. Completion of the national villagization program is expected by the end of 1976. The precise pattern of development within each ujamaa village has been left to villages and regional authorities to decide. Consequently, the extent of collective activity varies among villages and the precise workings of ujamaa are still evolving. During 1974 and 1975 emphasis has shifted from the establishment of ujamaa coopera- tives to the formation of planned or development villages. A development village is any planned or existing village which is registered with the Registrar of Cooperatives and has thus attained cooperative society status with borrowing powers. Only if in such a village production is organized primarily on a communal basis and the village has achieved economic viability can it be declared an ujamaa village, according to a recent Government directive. D. The Agricultural Sector Foreign Trade 2.09 Agriculture and livestock exports in 1974 were valued at TShs. 2,050 million. Cotton and sisal were the muajor foreign exchange earners, each ac- counting for about 23% of the total value of agricultural exports. Other important crops were coffee and cashew, with export shares of 17% and 10%, respectively. Imports of agricultural commodities (food, live animals and animal fats and oils) reached a tota:L of TShs. 1,138 million (US$159 million) in 1974, equivalent to 21% of total imports. Important smallholder cash crops are cotton, cashewnuts, tobacco, coffee, maize and rice. Agricultural Services 2.10 Extension. The agricultural extension service is organized re- gionally and coordinated by KILIMO in Dar-es-Salaam. About 2,500 field staff are engaged in crop work, and a further 800 in animal husbandry and animal health programs. On average one officer is responsible for 900- 1,000 farmers (Annex 2). 2.11 So far the extension services have been ineffective in introducing improved techniques to farmers. Major reasons have been the lack of adequate staff, inadequate training in the more valuable techniques, and poor manage- ment. The latter is reflected in a lack of work programing and logistic support, frequent transfers, and problems with the administration of pay and allowance. Refresher courses are infrequent and a recent analysis revealed that, on average, an extension worker receives additional training only once during his entire service.  - 5 - 2.12 Research. Agricultural research is organized in twelve major centers, each handling several crops or livestock. (Map IBRD10799) Priorities are set at an annual conference under the joint chairmanship of KILIMO's Directors of Crop Development and Livestock Development. Maize research is carried out at the Ilonga (Morogoro) and Ukiriguru (Mwanza) stations, and at the Agricultural Research and Training Institute in Mbeya. The Ilonga and Ukiriguru stations have developed several successful maize composites, of which the IC-A and UC-A are the best known and most widely used. Maize research is being supported by a grant from US-AID under which maize breeders and agronomists, including two seconded from the International Wheat and Maize Improvement Center (CIMMYT), are continuing to improve exist- ing composites and develop new composites and hybrids. Trials to improve maize production practices and to test the use of fertilizers have been carried out in all major maize producing areas of the country. The results show clearly that maize yields can be significantly increased by the timely planting of improved varieties, effective weed and insect pest control and fertilizer use (Annex 3). Further work is required, however, for a more detailed analysis of existing experimental results and to refine the maize fertilizer and variety recommendations for some ecological zones and for varying levels of maize husbandry. Also, in view of the soil acidifying effects of continuous applications of sulphate of ammonia and its high trans- portation cost per unit of nitrogen, trials are required to determine optimum methods for replacing it with more concentrated and less acidifying nitrogenous fertilizer such as diammonium phosphate and calcium nitrate (paras. 4.02 and 4.15). Farm Mechanization 2.13 The use of oxen, tractors and improved implements has developed only slowly in Tanzania and programs to increase it have generally been un- successful. Tractor utilization has been poor, due to frequent breakdowns and poor maintenance, and trained oxen made available to farmers on a grant basis have often remained unused or underutilized. Trained instructors have been in short supply and the use of improved low-cost implements has not been widely demonstrated. At present, ox cultivation is common only in Shinyanga region, but some use is also made in other western and northern areas. Government's villagisation program and the urgent need to increase crop production have created renewed interest in accelerating the use of mechanization, especially oxen, to improve yields and expand cultivated areas. However, past problems associated with the introduction of ox culti- vation justify no more than a gradual or pilot approach limited to those areas where oxen have traditionally been used, and ox training centers already exist (para. 4.18). Agricultural Marketing and Pricing 2.14 Major traded crops are marketed through state-owned or controlled marketing boards or corporations. The National Milling Corporation (NMC) is responsible for purchasing, processing and marketing for internal consump- tion mai,z' rice and wheat (Annex 4). Regional cooperative unions and  -6- primary societies act as agents of the parastatals collecting marketed crops and selling processed products. Unions obtain working capital from the National Bank of Commerce (NBC), and make cash payments to producers through member primary cooperative societies. 2.15 Producer prices and parastatals' buying and selling prices are established by the Economic Committee of the Cabinet. The former are an- nouced in advance of a planting season and are uniform throughout the country. Parastatals' buying prices at regional centers vary and allow for regional differences in handling and collection costs. Selling prices are uniform throughout the country. 2.16 Prices of production inputs are also established centrally but because parastatals operate various input subsidy and credit schemes, prices for selected items vary throughout the country. An input subsidization policy was originally adopted for cash crops only in order to provide in- centives to farmers to adopt improved practices involving the use of pur- chased inputs. Such a policy has had the effect of lowering farmers' resistance to purchasing inputs largely as a result of the lower required cash outlays and, consequently, a diminished financing risk in case of crop failure. Subsidies have also reduced the need for credit to finance such inputs which is an important factor in view of farmers' liquidity problems and a poorly developed rural credit system. The cost of subsidizing inputs for cash crops has traditionally been recovered through the price mechanism and through taxes. 2.17 Under the National Maize Program, inputs were provided free in 1973 and 1974 since most participating farmers had only very limited financial resources and all previous attempts to institute a credit system for small- scale food crop production had failed. Only a small fraction of the maize input subsidy could be recovered through the price structure because of the high cost of marketing and government's efforts to keep the rise in the cost of basic foods to a minimum.. This subsidy thus became a growing burden on the general budget. While continued subsidization of certain farm inputs remains justified, subsidy levels and methods of cost recovery have to be reviewed with the objective of gradually reducing the cost to the Treasury and eventually eliminating the subsidies; the introduction in 1975 of a cash payment by farmers of 25% of input costs represents a step in that direction (para. 6.04). Government is aware of the need for a consistent input pricing policy and the importance of keeping subsidy costs at manage- able levels and the Marketing Development Bureau of KILIMO is carrying out studies and assisting in the formulation of a viable long-run policy (Annex 5). 2.18 Farm inputs can be obtained relatively easily in the agricul- turally more developed areas of Tanzania, which include Mwanza, Mbeya, Iringa, Arusha and Kilimanjaro. Small and large-scale farmers and estates in these regions are usually able to purchase their requirements directly from organizations like the Tanganyika Farmers Association (TFA). Similarly in those areas where cash crops (like coffee, tobacco, tea and cotton) have been developed, farmers are usually well served by the crop authorities and  -7- cooperative societies. Outside these areas farmers experience great dif- ficulty in obtaining inputs, and the performance of cooperative unions and primary societies varies enormously. Scarcity of skilled manpower to maintain accounts and manage and supervise union and cooperative affairs, especially transport and produce handling, remains a major problem. The cooperative movement, however, is receiving substantial assistance to over- come these problems under a project supported by the Nordic group of coun- tries. Further support would be provided under the Fifth Education Project for which IDA financing has been requested. E. The Food Situation 2.19 Major food crops in Tanzania are maize, millet and sorghum, paddy, potatoes, cassava, bananas and beans. These are produced predominantly by smallholders, many of whom live in newly established villages and use most of their production for subsistence. Agricultural output has stagnated in recent years, with real growth in GDP in the agricultural sector falling behind population growth. Similarly, food crop production has failed to keep pace with the expanding population, and Tanzania has become increas- ingly dependent on imports of maize, rice and wheat. 2.20 The deteriorating food balance position was seriously aggravated by the poor harvests obtained in 1973 and 1974, and in 1974 some 440,000 tons of grain (291,000 tons of maize) were imported, at a cost of about TShs. 758 million (US$106 million). Maize, the traditional staple, is par- ticularly important since it is the only food crop marketed commercially on a national scale, and in the event of a general reduction in food crop pro- duction, unsatisfied demand for other staple foods increases demand for maize. Further grain imports are required in 1975 and these are expected to total about 300,000 tons valued at TShs. 390 million (US$55 million), of which maize would be 247,000 tons (Annex 5). F. The Maize Production Program 2.21 The food crop failures of 1973 and 1974 have emphasized the need to increase domestic food crop production and to establish strategic grain reserves. First among the national objectives is a significant increase in the output of maize. To achieve this, Government initiated in 1973, and con- siderably expanded in 1974, a National Maize Production Program. This program, executed by regional authorities, consisted of the free distribution of inputs, mainly fertilizer, seed and insecticides to farmers in newly-formed villages. In 1974 the program was extended to 13 regions, and appears to have reached approximately 25,000 farmers, or 150,000 people. The program, however, suffered from delays in input distribution, a poorly organized and badly informed extension service and from lack of effective management. Consequently, the program did not achieve its aim, and the increased maize production in  - 8 - 1975 resulted mainly from the increase in the maize producer price and Govern- ment's measures to encourage an increase in the area under maize. The aim of the National Maize Project is to develop and expand the Maize Production Pro- gram, paying specific attention to increasing yields per area cropped. (Annex 6) 2.22 Seed multiplication is carried out on Government regional seed farms, (supported by USAID), and by a number of private growers and companies under contract to the Tanzania Seed Company. The latter is managed by the Tanzania Wattle Company, which itself is a major seed producer, and both receive financial and technical support from the Commonwealth Development Corporation. Major problems are inadequate coordination of seed producers' activities and the need to determine more accurately actual requirements of the various composites and hybridIs for each maize producing area. In this respect, additional expertise ancl finance are not required, but the Project would assist the Seed Coordinating Committee to operate effectively, and ensure an adequate supply of cert:ified seed by drawing up a quantified work program, and introducing the necessary inspection and reporting procedures. III. THE PROJECT AREA A. Location 3.01 The Project area would include virtually all high potential maize growing areas of Tanzania (Annex 3). These areas are located in the fol- lowing thirteen of the country's 21 regions: Mara, Arusha, Kilimanjaro and Tanga in the north; Morogoro, Dodoma and Tabora in the center; Iringa, Mbeya, Rukwa and Ruvuma in the southern highlands; and Lindi and Mtwara in the southeast. Within these regions suitable maize producing areas have been identified in 28 of the 41 districts. Project activities would be limited to 950 villages, or about 50% of the total number in these districts. Maps IBRD11872 and IBRD2803R show the regions and districts included under the Project, and rainfall distributioni throughout Tanzania, respectively. 3.02 The northern Project areas have a bimodal rainfall regime, with short rains in October - November, and main rains from February to May, which is the main maize planting season for these areas. The central and southern Project areas have a single rain season generally lasting from November to April. November is the main planting season but this extends into December in the more northern areas. 3.03 The land use potential map (Map IBRD2804R) gives details of soil types. Groups 5 and 6 which include the ferruginous and ferrisoils, res- pectively, have the highest potential; groups 4 and 7, which include the sandier soils, and the acid leached ferralitic soils of the higher rainfall areas, have generally low fertility and only moderate potential. 3.04 The Iringa, Morogoro, Mbeya, Ruvuma regions and the Sumbawanga district of Rukwa region have excellent soils and rainfall conditions for  - 9 - maize production, as also have the more fertile highland areas of Arusha, Kilimanjaro and Tanga regions. The Mpwapwa District of Dodoma region has soils of moderate potential while, except for certain localized areas, the soils of Tabora, Mara, Lindi and Mtwara regions are less productive. B. Land Use and Types of Farming 3.05 All Project participants would live in either ujamaa or develop- ment villages. In the former, farming is either on a communal basis or in blocks where individuals have their own holding, although normally each family also has a subsistence plot. Development villages are made up of scattered individual holdings but with village committees, and they will gradually be converted to ujamaa villages with blocks cultivated either individually or communally. 3.06 Cropping systems vary widely throughout the Project area. Maize is the predominant food crop, being grown in combination with various other food and cash crops. The latter varies from tobacco in Tabora region, cotton and sunflower in Morogoro, and pyrethrum, coffee and tea in regions of higher elevation, while in the coastal areas coconut and cashew are important. 3.07 Various agro-economic zones have been identified throughout the Project area by the Bureau of Resource Assessment and Land Use Planning, but further work is required to develop improved farming systems for these zones. (Annex 3). The expertise provided under the Project would begin to rectify this situation. C. Rural Infrastructure Villages 3.08 Village structure in the Project area has changed significantly under the impact of Government's villagization program and changes will con- tinue to occur for a number of years. Facilities in these villages range from less than the bare necessities, to schools, health centers, piped water and other social and economic infrastructure for which Government has been providing substantial assistance. Village size varies from about one hundred to more than a thousand families. This, and the fact that the economic base is extremely varied, makes village development planning very difficult, and emphasizes the need for a flexible approach to deal with local conditions. However, despite the generally low level of development, and the difficul- ties arising from the brisk pace of the villagization program, villages are the only practical means through which a large number of rural people can be reached under this Project with its large geographical coverage.  - 10 - Marketing and Input Distribution 3.09 The National Milling Corporation (NMC) purchases maize and sells maize meal in all regions included in the Project area. In most regions, NMC buys from cooperative unions; exceptions to this are Tanga, Mara and Ruvuma regions where primary cooperative societies sell direct to NMC. Coop- erative unions are the weakest link in the marketing chain, mainly because staff salaries are too low to attract management staff of the right calibre. It is expected that with the assistance being provided under the Nordic Cooperative Project these problems can be alleviated. (Annex 1) 3.10 With the exception of Arusha and Tabora regions cooperatives are not involved in the distribution of farm inputs. Where cash inputs are used, they are generally distributed either by the Tanganyika Farmers Association (TFA) which operates in Arusha, Kilimanjaro, Iringa and Mbeya regions, by the Regional Trading Corporations, or by cash crop authorities. The latter obtain financing for this purpose from the Tanzania Rural Development Bank (TRDB). Under the maize program, inputs have been distributed by regional and district authorities under the supervision of the Regional Development Director. These arrangements would continue in most areas until cooperatives are able to assume this function, or other means, possibly involving cash crop authorities, can be developed. 3.11 The National Milling Corporation is responsible for the transporta- tion of marketed maize in the Project area, assisted in some areas by coop- erative unions. Although NMC owns about sixty trucks these are insufficient, and a significant portion of the transport requirements is still being met through the hire of trucks from small private operators. Lack of transport caused by poor vehicle maintenance, inexperienced and indisciplined drivers, poor road maintenance, and a chronic shortage of vehicle spares is becoming a significant constraint to expanding food crop production, and to rural development as a whole. There is no clear solution to these problems which are national in scale and can only be cured in the long-term. The rural area transport problems could be alleviated in part by the railway, which should provide least-cost transport services in the areas served. The railway, how- ever, is plagued by operational difficulties and a continuous shortage of rolling stock. 3.12 Commercial grain storage facilities are largely owned by NMC and cooperative unions. NMC's godowns and silos have a total capacity of about 260,000 tons of which about 140,000 tons are located within the Project area, 75,000 tons in Dar-es-Salaam and 45,000 tons in rural areas outside the Pro- ject. The cooperative unions' utilizable grain storage is estimated at 75,000 tons, most of which is located within the Project area. The total storage capacity of district and regional godowns and silos in the Project regions available for maize and other grains is thus estimated at about 200,000 tons. In addition, cooperative societies own storage facilities which are, however, of generally poor quality and unsuitable for storage over periods longer than a few weeks. Losses due to inadequate storage have been high, particularly in years with large crops, earlier than normal rains and conditions particularly favourable to pest infestation.  - 11 - IV. THE PROJECT A. Brief Description 4.01 The Project, over a four-year development period, would assist Tanzania to increase efficiency in maize production and raise output in order to lessen future dependence on imports. It would provide 950 selected villages within the thirteen most important maize producing regions of the country with maize production inputs, improved extension services, and maize demonstration plots; improved transport and storage facilities; and establish a central Pro- ject Servicing Unit in the Ministry of Agriculture to assist Regional Authorities in Project implementation. Detailed investment proposals for the long-term development of food crop production and processing and a national grain reserve would be prepared, and assistance provided for the planning of a national agricultural development program. A Project implementation schedule is attached as Chart World Bank-15244. Specifically the Project would: a) Provide the maize seed, fertilizers, insecticides and, on a trial basis, herbicides required over the four-year Project period, and establish a Maize Production Inputs Account to finance the purchase of production inputs and their distribution; b) Strengthen extension services through staff training, a clear description and arrangement of extension staff activities and providing staff with adequate transport and extension aids, and with the production inputs required to arrange for field demonstration of improved maize husbandry; c) Improve farm input supply and maize marketing arrangements by: (i) Providing transport at regional level to ensure the timely delivery of farm inputs and maize marketing; and (ii) Constructing farm input and maize stores at Project villages and at regional locations; d) Expand ongoing food crop adaptive research programs; e) Provide each Project region with a mobile film unit, and expand the circulation of KILIMO's agricultural newspaper;  - 12 - f) Implement a pilot scheme designed to develop methods for expanding the use of oxen cultivation; g) Establish and staff within KILIMO a Project Ser- vicing Unit (PSU) responsible for input procurement and distribution, Project financial control, moni- toring Project progress, preparing Project annual work programs, and providing regional authorities with the extension aids, films and technical as- sistance required to implement the Project; h) Provide consultant and technical services to re- view Government's plans and prepare detailed in- vestment proposals for: (i) Establishing a national grain reserve; (ii) Expanding and modernizing the milling and storage facilities of the National Milling Corporation; and i) Assist in the planning and preparation of a national agricultural development program. B. Detailed Features Maize Production Technology and Input Packages 4.02 The Project would initially provide three standard maize production improvement packages for all Project areas. The packages would include fer- tilizer recommendations similar to those made recently by Government's Soil and Fertilizer Coordinating Committee. 1/ Maize production recommendations would be continually reviewed as itformation becomes available from Project field trials and demonstrations, and as changes occur in the price relation- ship between maize and purchased inputs. Revised recommendations would be incorporated in Project annual work programs (para 5.03). 4.03 The three maize production packages would be: Package 1: Intensive extension to improve time and density of planting using composite seed, and improved weeding and pest control. Package 2: When Package 1 has been adopted satisfactorily and village maize yields have ri:sen to about 1,500 kg/ha, fertilizer would be introduced equivalent to 50 kg/ha triple super- phosphate, and 100 kg/ha sulphate of ammonia. Some villages 1/ The packages also would be similar to those recommended for the IDA assisted Kigoma Rural Development Project.  - 13 - in high altitude areas would use hybrid rather than composite seed. Package 3: In areas most suited to maize production, some villages would move into Package 3 which would use the equivalent of an additional 50 kg/ha sulphate of ammonia, and in most cases hybrid rather than composite seed. Generally, Project villages would move from Package 1 to Package 3 in a sequential manner. This arrangement would ensure an improvement in husbandry standards before introducing fertilizer, and an increase in village incomes from which,.purchases of fertilizer could be met. However, as little is known about farmers' adoption behavior, the sequence of introduction of the Packages could be experimented with. Some villages are already at a stage similar to Packages 2 and 3. These would be assisted to improve further their present production levels. The use of more concentrated fertilizers would be encouraged, especially di-ammonium phosphate and calcium nitrate to replace sulphate of ammonia which increases soil acidity, while the long- term aim of the research program would be to develop crop rotations involving legumes to reduce the need for imported nitrogenous fertilizer. 4.04 Government would initially subsidize all maize inputs except herbi- cides because of the high cost of inputs, especially fertilizer, the diffi- culties and cost of organizing a national credit program and the limited scope for providing added incentives to participating farmers through higher maize producer prices (paras. 6.03-6.04). During 1975/76 the subsidy, which would be paid by Treasury from general revenue, would be equivalent to 75% of the farm gate cost of inputs. Project villages would pay in cash the difference between the Government subsidy and the actual farm gate cost. Government would review the subsidy level annually in the light of experience gained under the Project, and with regard to changes in input and maize pro- ducer prices, and develop a program to gradually reduce and eventually elimi- nate the subsidy burden on Treasury. Assurances were obtained from Govern- ment that IDA would be consulted during the annual review of subsidy levels. 4.05 Maize production inputs required for Project year 1975/76 would be purchased by Government. 1/ These total about 16,000 tons of fertilizer, 4,000 tons of seed, and 1,800 tons of pesticide. The returns from the sale of all maize inputs (at subsidized prices) would be channeled into a Maize Production Inputs Account (MPIA) which would be established for the purchase and delivery of maize production inputs. In addition to the sales proceeds, Government would deposit into this account sufficient funds to cover payments for the purchase and delivery of inputs. The minimum amount to be deposited in the account at the beginning of each calendar quarter would be equal to the expected payments during that quarter. Total input requirements for 1975/76 and annual incremental requirements for Project years 1976/77 - 1978/79 would be financed under the Project. The establishment of MPIA would be a condition of Credit effectiveness. 1/ Government could be reimbursed for the subsidy portion of total input cost for that year from a USAID loan of July, 1975.  - 14 - 4.06 Since during the 1973/74 and 1974/75 cropping seasons maize pro- duction inputs were distributed free, the degree to which villages would be able and willing to purchase maize production inputs is unknown. Consequently, should some of the inputs purchased by Government in the first year remain unsold at the end of the 1975/76 planting, these inputs would be retained for the next season. Incremental inputs which are provided under the Project would only be purchased once inputs had been sold and the Maize Production Inputs Account had been funded in the agreed manner. Extension Services, Training and Demonstrations (Annex 2) 4.07 Short courses in maize agronomy and extension methods would be given to all field assistants and assistant field officers at their local rural training centers (RTC). These courses, lasting for two weeks, would be conducted by senior agricultural staff from the regions concerned. In the first year all staff would take one course with a refresher course every year thereafter. All staff responsible for teaching at the RTCs would attend a two-week course at the Ilonga Research Station. More intensive courses in maize agronomy and extension methods would be arranged for assistant field officers and the most progressive field assistants. These would be given at the agricultural research and training centres. In addition, a few senior agricultural staff would be sent overseas for intensive courses in advanced maize agronomy. 4.08 To assist in arranging suitable training courses in maize the International Maize and Wheat Improvement Center (CIMMYT) in Mexico would provide a senior training officer for a period of about four weeks. 4.09 Courses in organization and management would be arranged for all regional and district agricultural development officers, regional Project coordinators and district supervisors. These courses would last three weeks and be held at the Tanzanian Management Development Institute at Morogoro. In addition management seminars for regional and district staff would be arranged to ensure an exchange of views and experience between participating regions. 4.10 Two maize demonstration plots would be planted annually in each Project village. One plot would be planted and managed by the village school, while the other would be the responsibility of a progressive farmer selected by the village committee. In both cases, the necessary inputs and technical advice would be provided free by the Project. The school demon- stration plot would be about half a hectare in area with four equal-size plots. Three plots would demonstrate the three production packages provided under the Project and one the prevailing traditional system. The village demonstration plot would replicate the package being adopted by the village. These demonstration activities would also provide an opportunity for exten- sion staff to gain practical experi:ence.  - 15 - Input Supply and Maize Marketing (Annexes 4 and 6) 4.11 During the Project period, regional cooperative unions (or in certain regions, NMC) would become increasingly involved in the distribution of maize production inputs and the movement of maize grain. For the trans- port of the estimated Project input requirements and incremental maize pro- duction, an additional 44, 7-ton, and 44, 10-ton trucks would be required. These would be provided under the Project. 4.12 There is a shortage of storage facilities in most of the newly established villages. Under the Project about 530 villages would be pro- vided with stores of about 200 tons capacity. Stores of this size would be expensive if constructed by contractors. Consequently, the Project would provide construction materials and technical advice, while construction would be carried out by villagers on a self-help basis. The materials cost of village stores constructed in this manner would be about US$6,000 each. 4.13 Project villages would be selected by Regional Authorities during Project implementation, and since there is no previous experience of farmers' ability and willingness to purchase maize production inputs it is impossible to estimate precisely how many stores would be required and how big individ- ual stores would be. Consequently, a lump sum for village store construc- tion has been included, and the exact size and location of village stores would be determined during Project implementation by Regional Authorities with the assistance of the Project Servicing Unit. Inevitably, there would be a time lag between identifying the need for a store and completion of construction. To overcome this problem, and to ensure that adequate tem- porary storage was available, the Project would supply cooperative unions with tarpaulins which would be used by Project villages until permanent stores were constructed. 4.14 There is a need also for additional regional storage for maize and farm inputs. The Project would provide for the construction of storage facilities for about 34,000 tons of farm inputs and 80,000 tons of maize. Similar to village storage, the precise location of the new regional storage would only be determined during Project implementation. All Project storage and transport requirements would be incorporated in regional annual work programs. These work programs would be appraised by PSU and would be sub- ject to approval by the Project Steering Committee and IDA (para. 5.03). Crop Trials and Adaptive Research (Annex 3) 4.15 While there is sufficient experimental data to fully justify the general use of the varieties and the amounts and types of fertilizer recom- mended under the Project, further field research is required to determine more optimum recommendations for the various ecological conditions existing throughout the Project area and to find a suitable nitrogenous fertilizer to replace sulphate of ammonia (para. 2.12). About four trials would be established in each region every year and these would be organized by the Research Division of KILIMO. To enable the formulation of a national adaptive  - 16 - maize research program and to give a clearer idea of maize production con- ditions throughout Tanzania, a dei:ailed review would be made of all aspects of maize agronomy in the major maize growing areas of the country. The review which would include recommendations for the future agronomic develop- ments required to put the maize industry on a sound- long-term footing would be carried out by a consultant agronomist especially recruited for the pur- pose. Terms of reference for the review are detailed in Annex 7. Mobile Film Units and Agricultural Newspaper (Annex 2) 4.16 The Project would provide a mobile film unit for each of the 13 Project regions. Provision would also be made for the purchase and produc- tion of suitable films on maize production. 4.17 KILIMO publishes a monthly agricultural newspaper (Ukulima wa Kisasa) although few copies are distributed in the newly established villages. The Project would increase the number of copies printed and arrange for their distribution in all Project villages. Articles on maize production would be incorporated in the paper. Maize production leaflets would also be prepared and distributed. Oxen Pilot Scheme (Annex 6) 4.18 A pilot scheme to develop methods of improving oxen utilization and for proving suitable ox-drawn equipment would be carried out in up to twenty selected villages located in those Project regions where oxen use is most common. Selected villages would be required to own a minimum of ten pairs of oxen. Such villages would be provided with an experienced oxen trainer for two years, and up to ten sets of ox-drawn equipment, each including a cart, harrow, plough and cultivator. The equipment would be provided free with the proviso that it would be withdrawn if farmers failed to participate effectively in Project activities. In addition, the Project would provide oxen to four mechanized agriculture centers which are being established with UNDP/FAO assistance. The animals would be used for the development and testing of new ox-drawn implements, in particular a maize planter and interrow cultivator. Project Servicing Unit (PSU) 4.19 Regional authorities would implement the Project at regional and district level. To assist them a PSU would be established within the Crop Production Division of KILIMO. PSUr would be headed by a Project Manager and would include the following key staff: Extension Specialist, Farm Manage- ment Specialist, Financial Controller, and Procurement and Distribution Officer. It would be responsible for controlling Project finances, procuring and distributing maize production inputs and other Project-related materials and equipment, coordinating regional maize production plans, preparing annual Project work programs, providing regional authorities with extension aids, and monitoring Project progress. The Project would meet the full costs of estab- lishing and operating PSU, including the cost of purchasing and operating a  - 17 - light aircraft, which would provide the essential communication link between the Unit and the 13 Project regions. Terms of reference for the PSU are included in Annex 7. Pre-investment Studies 4.20 Government recognizes the need to establish a national grain re- serve, for which maize is the most important commodity, and for expanding and modernizing the milling and storage facilities of NMC. The Project would provide up to 60 man-months of consultancy services to carry out the required pre-investment work and prepare detailed investment proposals for a national grain reserve and for developing NMC's storage and milling facilities. Terms of reference for the pre-investment studies are included in Annex 7. Planning and Preparation of a National Agricultural Development Program 4.21 The Project would provide US$180,000 over a three-year period for agro-economic surveys, crop trials, land use planning and transport and allowances for KILIMO planning staff and consultants. A long-term national agricultural development program would be prepared which is compatible with Tanzania's regional approach to implementing its development effort and directed mainly towards those research, extension and investment activities that are better carried out at a national rather than a regional level. A preparation program is presently being prepared with RMEA assistance. C. Project Costs 4.22 Detailed Project cost estimates, based on July, 1975 prices which have been adjusted to take account of the October, 1975 devaluation of the Tanzanian currency, are presented in Annex 8 and summarized below:  - 18 - Foreign Local Foreign Total Local Foreign Total Exchange --(TShs. millions)--- ---(US$ millions)---- % Farm Inputs 30.8 84.6 115.4 3.822 10.508 14.330 73 Project Servicing Unit 4.6 5.3 9.9 0.566 0.667 1.233 54 Regional Staff and Support 18.0 16.2 34.2 2.237 2.008 4.245 47 Storage and Transport 24.1 43.8 67.9 2.989 5.440 8.429 65 Training 4.5 3.0 7.5 0.569 0.369 0.938 39 Air Transport 0.6 1.8 2.4 0.084 0.221 0.305 72 Pilot Ox Utiliza- tion 0.6 0.1 0.7 0.069 0.015 0.084 18 Project Prepara- tion Studies 1.3 4.2 5.5 0.152 0.527 0.679 78 Subtotal 84.5 159.0 243.5 10.488 19.755 30.243 65 Physical Contin- gencies 5.2 7.0 12.2 0.643 0.871 1.514 57 Price Contingencies 21.0 30.3 51.3 2.611 3.769 6.380 59 Total Project Cost 110.7 196.3 307.0 13.742 24.395 38.137 64 Physical contingencies have been included at a rate of 10% for all Project components except farm inputs, the pilot ox utilization scheme and project preparation studies, for which no physical contingencies have been allowed. Estimated input quantities are at maximum levels at which inputs could be usefully employed under the Project. Price contingencies have been included at the following annual rates: vehicles, equipment and general for 1975/76 - 1978/79 respectively: 11%, 9%, 8% and 8%; the foreign cost of buildings and civil works: 15%, 13%, 12% and 12%; and the local cost of buildings and civil works: 22.5%, 19%, 16.5% and 15% (Annex 8, Appendix 1). Local taxes, principallv including income taxes on earnings of local staff and some fuel taxes, are estimated to represent less than one percent of total Project cost.  - 19 - D. Financing 4.23 Financing of the Project would be as follows: Sources of Financing Total Project Costs US$ (Millions) % Government of Tanzania 1/ 11.1 29 Farmers 4.0 11 Arab Bank for Economic Develop- ment in Africa (ABEDIA) 5.0 13 IDA 18.0 47 Total 38.1 100 1/ Of which up to US$6.4 million is being made available to finance part of the cost of inputs in Year 1, from a US$12 million USAID loan to Tanzania's agriculture sector, thus reducing Government's share to 12% of total costs. 4.24 Under the proposed financing arrangements the cost of inputs in the first Project year would be financed by Government (US$6.4 million) and farmers (US$2.1 million). Of the remaining Project costs, Government would pay US$4.7 million, farmers US$1.9 million, ABEDIA US$5.0 million and IDA US$ 18.0 million. The ABEDIA loan, which would finance expenditures for all project vehicles, the light aircraft, village equipment and overseas training, would be at 2% annual interest over 20 years, with a 5-year grace period. The fulfillment of the conditions precedent to the effectiveness of the loan agreement between ABEDIA and Government would be a condition of effectiveness of the IDA credit. The ABEDIA/IDA contributions together would finance approximately $16 million of foreign exchange expenditures and US$7 million of local costs. 4.25 Funds for storage and transport would be channeled by Government to TRDB (Annex 9) at 4% annual interest, over 20 years, with a 5-year grace period. TRDB loans would be at 8-1/2% annual interest over 10 years for the construction of storage facilities and at 8-1/2% annual interest over 3 years for vehicles. These rates would be in line with TPDB's current standard terms for short-term loans and one percentage point above for medium and long-term loans. Should Government wish to make Project funds available to private transport owners for the purchase of lorries, such funds would be channeled through the National Bank of Commerce, through its wholly-owned subsidiary KARADHA Co. Ltd. The same terms as for TRDB would be applicable. Subsidiary  - 20 - loan agreements would be drawn up governing the terms under which Government provided funds to lending institutions. A loan agreement between Government and TRDB would be a condition of credit effectiveness. E. Procurement and Disbursement 4.26 PSU's procurement of Project incremental fertilizer, insecticides, herbicides, vehicles, motor cycles, platform scales, safes, aircraft, village store construction materials and the contracts for the construction of regional and NMC storage (US$21 million) would be by international competitive bidding, following IDA guidelines. Local manufacturers would be allowed a preference margin of 15 percent or the existing rate of duty, whichever is lower. Local civil works contractors would have a 7-1/2% preference margin. All maize seed, and office equipment, bicycles, tarpaulins and small miscellaneous items costing less than US$120,000 per contract would be procured locally in accordance with normal Government procedures which are satisfactory to IDA. The required maize seed, mainly local varieties, would be supplied by the domestic seed industry. Thirteen fully equipped film vans would be obtained at special discount from UNESCO through the kudio-visual Institute in Dar-es-Salaam. Assurances were obtained during credit negotiations that PSU would bulk orders for vehicles and other equiLpment wherever possible; that draft tender documents for all contracts expected to cost in excess of US$300,000 would be submitted to IDA for approval before invitations are issued: that bid analysis and recommendation for award woulel be submitted to IDA for comment before contracts are awarded; and that all other procurement would follow standard Government procedures which are satisfactory. 4.27 Against appropriate documentation, IDA would disburse: (i) For incremental maize production inputs required in Project Years 2 to 4, including inputs for research trials and demonstration plots: (a) 100% of foreign expenditures if directly imported; (b) 80% of total expenditures if locally procured; (ii) For equipment and materials: (a) 100% of foreign expenditures if directly imported; (b) 80% of total expenditures if locally procured; (iii) 80% of the cost of constructing regional and NMC storage as well as offices for Project Servicing Unit staff; (iv) 100% of the foreign expenaditures of consultancy services, including PSU staff;  - 21 - (v) 80% of the cost of local training; local PSU and regional staff salaries; and the cost of operation and maintenance of PSU and regional vehicles, excluding lorries. Documentation for expenditures under (v) above would not be submitted to IDA for review, but would be retained by the Borrower and be available for ins- pection by IDA during Project supervision. A schedule of quarterly disburse- ments is in Annex 12. Any surplus funds would be subject to cancellation. 4.28 For the crop season 1975/76 retroactive financing would be required for the items detailed in Annex 8, Table 12. These items have a total esti- mated cost of TShs. 4.2 million (US$0.5 million) of which TShs. 2.1 million (US$0.3 million) would be eligible for disbursement by IDA and TShs. 1.5 million (US$0.2 million) for disbursement by ABEDIA. F. Accounts and Audit 4.29 Government would maintain separate accounts for maize production inputs and for disbursements through the Project Servicing Unit which would be audited by the Auditor General. TRDB and NMC would also maintain separate accounts for subloans under the Project. The TRDB and NMC accounts would be audited by the Tanzania Audit Corporation or any other independent auditors acceptable to IDA, and the accounts and auditors' reports would be submitted to IDA within six months of the close of each financial year. Auditing of the regional and village cooperatives is the responsibility of the Prime Ministers Office; all audited cooperative accounts would be available for examination by IDA. G. Ecological and External Effects 4.30 The Project would have a favorable effect on the ecology, in as much as soil fertility would be improved and erosion controlled by better husbandry, crop rotation and fertilizers. However, as part of the input package, the Project would recommend the use of relatively small amounts of DDT 5% powder (10 kg.ha) against stalkborer and DDT 25% liquid against army worm for about 10% of the Project area. While DDT is the most effective and economic known control of the two pests, the Project would experiment with other control methods as they become available. Although the use of DDT has been prohibited in some industrialized countries, there is no evidence to suggest that the proposed quantities would have any harmful effects in the Project area.  - 22 - V. ORGANIZATION AND MANAGEMENT 5.01 Project organization which would be based entirely on Government's existing administrative structure would be as follows: National level 5.02 A Project Servicing Urnit (PSU) would be established in the Crop Production Division of KILIMO. Experienced staff would be appointed to PSU, including a Project Manager, a Financial Controller, and specialists in extention, farm management, and procurement and distribution. Only if suitably qualified Tanzanians could not be made available, these positions would be filled with internationally recruited staff. Appointments to the above positions would be made in consultation with IDA. It would be a condition of Credit effectiveness that staff in those positions, with the exception of the farm management specialist, have taken up their duties. Assurances were obtained during negotiations that, in case the farm manage- ment specialist could not be recruited by April, 1976 alternative arrangements to carry out the functions of that specialist satisfactory to IDA would be made until such time as the post could be filled. 5.03 PSU would have the following key functions: (a) Project Planning and Cioordination: The maize production targets for each region have been established with the aim of attaining self-sufficiency in regions where this is economically justified and increas:ing surplus maize from those regions which are particularly suited to production and reasonably close to consuming areas. Within these guidelines, a detailed work program would be prepared annually for each region and district. The PSU in conjunction with regional Project coordinators would consolidate the regional plans into an overall national plan. This national plan would be reviewed annually by the Project Steering Committee anc. submitted to IDA for approval (para. 5.10). (b) Financial Administration and Procurement of Inputs: PSU would be responsible for the administration of Project funds and for operating the Maize Production Inputs Account. Inputs would be purchased at their full market price and sold to farmers through the regional cooperative unions or regional authorities at subsidized prices. Inputs would be sold in accordance with Government policy which, at present, provides for uniform prices throughout the Project area. Government is aware that uniform producer prices would distort resource allocation in the long-run and is keeping its pricing policies under constant review. The Inputs Account would receive sufficient payments from Government to meet the difference between the full cost of farm inputs, including transport, and the subsidized sale price to farmers. PSU would keep separate accounts for maize production inputs transactions and its other operations.  - 23 - (c) Technical Support for the Regions: PSU would provide technical support to agricultural staff in the regions under the Project. This would include assistance with training programs for exten- sion staff, production of leaflets on maize production methods, organization of demonstration plots and film shows. (d) Transport and Storage: PSU would arrange the distribution of farm inputs to regional cooperative unions or other agencies as appropriate. PSU would not operate its own transport for this purpose but would use the railways and transport companies. PSU would also play a major role in helping to prepare plans for development of storage and transport in the regions and approving expenditure for this purpose. It would arrange for assistance to be provided to villages to aid in village storage construction. (e) Project Evaluation: An important aspect of Project implementa- tion would be the continuous improvement of maize production technology and the development of improved farming systems. The Project field research program, village demonstration activities and maize production statistics would provide valuable data for this purpose. PSU would have responsibility for collecting appropriate statistics, through crop sampling and other evalua- tion means; for continually monitoring Project progress; and for revising Project plans as appropriate. PSU would be assisted in this task by USAID supported staff. Regional Level 5.04 Within each region the Project would be administered through the regional government administration with its chain of command going from the regional headquarters through the districts, divisions and wards down to the villages. In each region the regional authorities would appoint a Project coordinator who would be assisted in each district by a Project supervisor (Annex 7). Appointment of staff in those positions would be a condition of Credit effectiveness. 5.05 A farm management officer would be appointed in each regional headquarters. He would liaise closely with the farm management specialist assisting PSU (para 5.02). He would be responsible for monitoring Project progress, estimating input use, maize yields and production in Project vil- lages and analyzing the results achieved on demonstration plots and trials. Also, he would advise on the most appropriate production techniques for the Project within the framework of local farming systems. 5.06 PSU would generally supply regional cooperative unions with maize production inputs, for onward distribution to village primary cooperatives. However, the establishment of the cooperative arrangements would take time, and until they are satisfactorily introduced inputs would continue to be  - 24 - supplied through existing parastatal organizations, including cash crop authorities and regional authorities. Project villages would buy inputs on a cash basis at subsidized prices. PSU would be reimbursed the cost of in- puts by the respective distributing agency in the regions as they are sold. 5.07 Arrangements for the purchase of maize from Project villages, and for financing such purchases would continue essentially as at present. In most regions the primary cooperative societies would purchase maize at the village level, acting as agents of the regional cooperative union, who in turn would act as agent for NMC. The cooperative unions would arrange for maize to be transported from villages to regional NMC depots. The National Bank of Commerce would continue to advance funds to cooperative unions, which in turn would provide financing to primary societies. Farmers would be paid cash on delivery at the official price. In regions where the cooperative system is unable to market maize efficiently, the NMC would purchase directly from Project villages. Village Level 5.08 The village would be the basic development unit. Within each village, either a newly established village cooperative society or an older multi-village primary cooperative would be responsible for distributing village requirements of seeds, fertilizers and insecticides and for the marketing of surplus maize. 5.09 The Project would provide one resident extension worker or village farm manager to each Project village. This would be difficult to achieve in some areas, and initially some villages would have to share staff. The vil- lage structure would make communication much easier for extension staff. Villagers would cultivate at least part of their maize in communal plots or in blocks of individual plots, thus facilitating contact with groups of farmers. Project Steering Committee (PSC) 5.10 To provide the inter-ministerial coordination required for policy decision making and for the approval of Project annual work programs, a Pro- ject Steering Committee would be established under the chairmanship of the Principal Secretary, KILIMO, and with members from the Prime Minister's Office and the Ministry of Finance and Planning, and the Commissioner for Rural Development. The Project Mxnager would be secretary of PSC. Establishment of PSC would be a condition of Credit effectiveness. Staffing 5.11 The Project Manager, a Tanzanian, and deputy Project Manager (Agri- cultural Development Service) were appointed in July 1975, and staff for the remaining key PSU staff positions are being recruited. Other Project staff, including regional coordinators, regional farm management officers, and  - 25 - district supervisors would be selected from the existing cadre. In order to achieve a staffing of one extension worker per Project village, an addi- tional 140 and 100 field extension staff, respectively, would be recruited in 1976/77 and 1977/78, mainly from Standard 12 school leavers of which sufficient numbers would be available. Thirteen mobile film unit teams, consisting of a driver/mechanic and projectionist/narrator, would be recruited by the Project, and their training undertaken by the Audio-Visual Institute in Dar-es-Salaam. VI. PRODUCTION, MARKETING, FARMERS' BENEFITS EFFECTS ON GOVERNMENT BUDGETS Yield and Production 6.01 Average maize yield in Tanzania are very low at about 750 kg/ha, although the range of yields is very large. The Project, which would include only the better maize producing areas, assumes a base yield of 1,100 kg/ha, increasing to 1,500 kg/ha for Package 1; 2,200 kg/ha for Package 2; and 2,700 kg/ha for Package 3. The higher yield levels are already being ob- tained by progressive farmers in high potential areas. The assumed yields for all Project villages would increase from an average of 1,370 kg/ha with- out the Project to 2,050 kg/ha at full development. The Project would in- crease maize production by about 20,000 tons in year 1 rising to about 195,000 tons at full development in 1982. Full development is not expected before 1982 because of an assumed three-year time lag between aDplication of a certain input Package and the full realization of its yield potential. Markets and Prices (Annex 10) 6.02 The Project would assist Tanzania to become more efficient in producing maize and substitute for imports which in 1974 and 1975 totalled about 291,000 tons (TShs. 360 million or US$50 million), and 247,000 tons (TShs. 301 million or US$42 million), respectively. During the Project period Government plans to establish a national maize reserve of about 90.000 tons and any Project production in excess of current requirements would be channelled into this reserve. Once the national reserve is estab- lished, Government policy would be to aim at maize production levels of about self-sufficiency. It would be assisted in this task by a strengthened Marketing Development Bureau which would be responsible for reviewing producer prices, and providing crop pricing recommendations. However, because of production fluctuations due to climate, it is inevitable that small maize surpluses would occur in some seasons which would be absorbed in border trade or added to the reserve. 6.!)3 According to projections prepared by the Economic Analysis and Projections Department of the World Bank, the international maize price (current terms: f.o.b. Gulf Ports) is expected to increase from US$128/ton (TShs. 1,030) in 1975 to US$154/ton (TShs. 1,240) bv 1980. The farm gate  - 26 - equivalent of these prices in Tanzania (import parity) is approximately TShs. 1,113/ton (US$138) for 1975 and TShs. 1,290/ton (US$160) for 1980. Actual farm gate prices for the 1974/75 season are TShs. 750/ton (US$93) and for 1975/76, TShs. 800/ton (US$99). Considering the sharp increases in producer maize prices in Tanzania in recent years (from TShs. 300/ton, or about 42% of import parity in 1973/74, to 73% in 1975/76), the unfavourable impact further increases would have with respect to the relative profitability of growing other crops, and the low estimated export parity price of Tanzanian maize of TShs. 550/ton in 1975, the domestic producer price for maize appears adequate at present. However, progressively higher producer prices relative to import parity would be justified from 1976/77 onwards if there is an indication that production targets are not going to be reached with present price levels. Input Subsidy, Government Budget and Foreign Exchange 6.04 The current subsidy level of 75% is intended to bring about a more economic use of inputs compared with the two preceding years when inputs were subsidized 100%, and still permit the target number of farmers to participate in the Project. The subsidy is justified in view of the high cost of inputs and because it is presently not feasible to provide sufficient incentives to farmers through such alternative means as credit or higher producer prices (Annex 5). Assuming that the 75% subsidy level would be maintained through- out the Project period, the cumulative cost of the subsidy is estimated at TShs. 289.7 million (US$36.0 million) by the end of 1978/79: if the subsidy level were then reduced to 60% the following four years and to 40% thereafter, the cumulative cost over the ten-year period 1975/76 - 1984/85 would be about TShs. 797 million (US$99 million). A reduction in the subsidy level should be possible as Project village incomes increase and with the anticipated fall in world fertilizer prices. It is Government policy to review future subsidy levels and options of recovery of the cost of the subsidy annually with a view to a progressive reduction in the budgetary burden. 6.05 With the proposed financing arrangements (para. 4.24) and the as- sumed future Government input subsidy policy (para. 6.04), the Project's overall negative impact on the Government budget would be a cumulative TShs. 281.6 million (US$35.0 million) by end of 1978/79, and TShs. 915 mil- lion (US$114 million) by 1984/85. This burden, which would be eased through contributions from a USAID loan of an estimated TShs. 52 million (US$6.4 mil- lion), would not be larger than the burden resulting from continued importa- tion of maize in the absence of the anticipated production from the Project. 6.06 The foreign exchange impact of the Project would be favorable, in spite of continuing fertilizer imports. The foreign exchange content of input costs would be TShs. 282 million (US$35.1 million) for the four-year period ending 1978/79, and TShs. 988 million (US$123 million) for the ten- year period ending 1984/85. Including the foreign exchange content of maize inputs and all other Project related costs, the cumulative outflow of foreign exchange over the ten-year period would be TShs. 1,194 million (US$148 million). Against these outflows are foreign exchange savings due to decreased maize  - 27 - imports. Over the ten-year period, gross foreign exchange savings from Project maize production would be about TShs. 1,931 million (US$240 million), and total gross savings from maize production plus inflows of external financing TShs. 2,116 million (US$263 million); the net savings would be TShs. 922 million (US$115 million). Farmer Benefits (Annex 11) 6.07 Farm budgets for the three improved husbandry packages (para. 4.03 and Annex 3) show that, at full development, farmer margins would be increased by 36%, 87% and 129%, respectively, over expected levels without the Project within three years of application of the packages. These margins would be sufficiently attractive to encourage farmer participation. 6.08 Assuming that the Project maize area per farm family would be 0.75 ha, the average increase in farmer income at full development, including the value of additional subsistence production, would be TShs. 342 (US$42) per family in 1975 prices. The estimated incremental cash income would be TShs. 228 (US$28). On a per capita basis these increases would be TShs. 68 (US$8) and TShs. 46 (US$6) respectively. This compares with the present estimated rural per capita GNP of TShs. 350 (US$43) (Annex 11, para. 16). Although the Project would not benefit the poorest segment of the rural population, the present incomes of Project beneficiaries would generally be below the average of rural incomes in Tanzania. VII. ECONOMIC BENEFITS AND JUSTIFICATION (Annex 13) 7.01 The primary direct benefits of the Project would be increased pro- duction of maize and a resultant reduction of grain imports. 7.02 The Project's interal economic rate of return is estimated to be 37% over a 20 year period. Project costs except the costs of the oxen utilization scheme and of NADP project preparation, and 50% of the cost of the storage and milling consultants were included in the calculation. No cost was attributed to family labour since the Project would assist only those farmers already growing maize, and only a small increment of presently under-employed family labour would be required under the Project. The costs of maize production inputs and value of Project incremental maize output are based on recent World Bank projections. Project incremental maize production is valued at import parity up to 1980, since self-sufficiency would not be reached before then. From 1981 onwards Tanzania would aim to maintain self- sufficiency and a national maize reserve of about 90,000 tons and since small annual surpluses are assumed to alternate with small annual deticits, Project incremental production is valued mid-way bet-ween imnort and export parity. Foreign exchange costs and benefits are va'ued at TShs. 1G = US$1.00 to re- flect the economic costs of foreign exchange to Tanzania: if foreign exchange were valued at the prevailing exchange -ate of TShs. 8.05 = US$1.00, the economic rate of return would he 30M.  - 28 - 7.03 The risks most likely to effect the Project's economic rate of return are the uncertainty about the extent to which farmers would be willing to purchase fertilizers and other inputs and to apply the recommended pro- duction methods. Only limited experience has been gained so far with the new village system and there is some doubt about how well this will work, especially where communal methods of production are emphasized. To the extent possible, these risks have been minimized by designing the Project flexibly so as to be able to utilize new experience gained during implementation and to eliminate those activities not yielding the expected results. Nevertheless, even if incremental output from the Project were 30 percent below expectation, the Project would still show a rate of return of 16 percent. VIII. AGREEMENTS REACHED AND RECOMMENDATION 8.01 During negotiations agreement was reached on the following principal points: (a) Inputs would be sold for cash at subsidized prices. The rate of subsidy would be reviewed from time to time in consultation with IDA (para. 4.04); (b) The Maize Production Inputs Account would be funded in the agreed manner (para. 4.05); (c) All remaining new appointments and subsequent appointments to replace senior staff in the Project Servicing Unit would be made in consultation with IDA (para. 5.02); 8.02 It would be a condition of effectiveness of the IDA Credit that: (a) The Maize Production Inputs Account had been established (para. 4.05); (b) The conditions precedent to the effectiveness of the loan agreement between ABEDIA and Government had been fulfilled (para. 4.24); (c) A subsidiary loan agreement between TRDB and Government, satisfactory to IDA, had been executed (para. 4.25); (d) All remaining key staff of the Project Servicing Unit, except the Farm Management Specialist, had been appointed and had taken up their duties (para. 5.02);  - 29 - (e) The regional Project Coordinators and district Project Supervisors had been appointed (para. 5.04); (f) The Project Steering Committee had been established (para. 5.10). 8.03 The Project is suitable for an IDA Credit of US$18 million on standard IDA terms.   ANNEX 1 Page 1 TANZANIA NATIONAL MAIZE PROJECT UJAMAA AND THE COOPERATIVE SECTOR A. Cooperative Structure Introduction 1. Cooperatives exist in large numbers throughout Tanzania. By the end of 1973, there were 2,200 registered cooperatives, including the Coopera- tive Union of Tanganyika, the Savings and Credit League, 25 cooperative unions, 1,518 cooperative marketing societies, 265 cooperative production and market- ing societies, 314 savings and credit societies, and 78 cooperative consumer societies. 1/ The existing cooperative structure is in the process of change. The present 25 cooperative unions are to be consolidated and reduced to 20, one for each region. 2. The history of cooperative movement in Tanzania goes back to the voluntary "growers associations" of the 1920s. In 1932 the first agricultural cooperatives were registered in the Kilimanjaro area to promote coffee as a cash crop for peasant farmers. The movement soon became popular and spread quickly to other areas where commercial peasant farming was relatively advanced. By 1952, there were 172 registered societies and by 1961 the number had increased to 857. Thus, cooperatives grew up spontaneously in the regions of Moshi, Arusha, Bukoba, Mwanza, Ruvuma, Rungwe&and Iringa. Shortly after independence, the Government decided that the cooperative movement should be made an integral part of the political structure and, consequently, 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. For example, Government now sets many of the financial margins within which the cooperatives have to operate; controls the hiring and firing of all union officials with a salary above TShs. 650 a month; controls all cooperative education;thas assumed virtually all responsibilities for inspection of cooperative societies; has transferred to the unions special tasks previously performed by the private sector; and has moved to link in- extricably the cooperative movement with ujamaa village development. 1/ Updated from annexes of the Kigoma (Cooperatives) and Geita (Ujamaa) Reports with assistance from the "Nordic Cooperative Assistance to Tanzania" Team.  ANNEX 1 Page 2 3. In addition, Government has appointed the executive committees of Coast, Dodoma and Mbeya regions, and the General Managers of Arusha and Coast Regional Cooperative Unions are appointed civil servants. Latest developments in the area of Government intervention include the liquidation of the Kigoma Regional Cooperative Union, and the dismissal of all senior officials of the Tanga Regional Cooperative Union, for reasons of inefficiency and gross mis- management. 4. 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. Increasingly, the cooperative structure is being used to further Government's political and socio-economic goals. 5. The principal responsibility of the cooperative unions, apart from various consumer, financial, transport and other services they might perform, is to move the marketable surplus of virtually all scheduled crops in all regions of the country from the primary societies to the appropriate marketing 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 pro- duction has sometimes suffered from such monopolistic practices, because the cooperativea have often suffered from 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 account- ability within the system, and the frequent allocation by Government of un- realistic costings and margins. Legislation, Organization, and Supervision 6. The first Cooperative Societies Ordinance, stipulating the legal basis for all cooperatives, was enacted in 1932. It was subsequently re- placed by the Cooperative Societies Ordinance of 1968. A further supplement is expected soon, with the aim of specifying the processes and criteria by which villages are to become multi-purpose primary cooperative societies. Two stages would be distinguished: i) a fully-registered multi-purpose cooperative society, with complete legal independence and status, and with limited liability; and ii) full control of the major means of production in the village by the cooperative society; the village would then be a true Ujamaa.  ANNEX 1 Page 3 7. 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 country's cooperatives. However, many of its functions have been transferred to the Government. For example, 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. 8. Following the Government reorganization of July 1971, the Rural Development Division, headed by the Commissioner for Rural Development, 1/ was established to deal with matters relating to cooperatives. Subsequently, this division was changed to the "Ujamaa and Cooperative Development Depart- ment" (see Chart 1). This central cadre is supported in each of the 20 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, who is responsible for the registration of all cooperatives. At the regional level, the Regional Develop- ment 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 dealing with the coopera- tives include the Unified Cooperative Service Commission, and the Audit and Supervision Fund. The former is charged with the recruitment, appointment (including promotions and transfers) and termination of all cooperative personnel earning TShs. 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. 9. The Audit and Supervision Fund currently has a network of 12 zonal offices distributed throughout the country, which it plans to expand into 20 regional offices. Total staff consists of 125 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 Coopera- tive College at Moshi or the Institute of Development Management at Morogoro. 1/ The Commissioner is head of the Ujamaa and Cooperative Development Department which is not within the Prime Minister's Office (PMO) but has the status of an extra-ministerial department.  ANNEX 1 Page 4 Only about 50 staff members are currently qualified to audit and supervise the financial accounts of primary societies. 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 disarray, requiring the accounts to be reconstructed before they can be audited. To simplify and standardise accounting practices at the society level, a simplified, uniform book-keeping system for ujamaa vi:Llages is in the final stages of preparation. In summary, problems of audit and supervision 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 supervision. 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 Traipnin 10. 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. The Cooperative College currently offers diploma, final 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 construction. In addition, rural training centers, coorespondence courses, and regional education programs contribute to education and training at the local level. 11. Past efforts have, however, only begun to address the pressing needs to train and educate staff of the primary cooperative societies, the coopera- tive unions, and the Ujamaa and Cooperative Development Department. The ujamaa and cooperative education plan (1973 - 1979), therefore, proposes to establish over the niext 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 1U93/74 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. 12. Substantial cooperative technical assistance is currently being provided by the five Nordic countries, under the "Cooperative Assistance to Tanzania" project. Of the 26 experts now assigned under an agreement origin- ally signed in 1968, six are working within the Ujamaa and Cooperative Devel- opment Department and the Cooperative Union of Tanganyika to establish sys- tems for wholesaling and transport, credit financing, management accounting,  ANNEX 1 Page 5 promotion and supervision and consumer and retail activities. Another 11 experts are working on the specific problems of 11 regional cooperative unions, and the remainder are working in the Cooperative College, of which five are stationed in Moshi and three in the regions. 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. B. The Ujamaa Village Program 13. Tanzania's socialist development policy is aimed at the simulta- neous achievement of economic growth and an egalitarian society through an increasing emphasis on agriculture and rural development. In recent years this rural development strategy has focussed on the establishment of co- operatives or ujamaa villages throughout the country. The immediate ob- jectives are social and political: to foster a sense of self-reliance through self-help, to achieve equality of income, and to raise rural incomes in order to close the rural-urban gap and reduce the flow of people to the major twowns. Government recognises, hcwever, that to be successful in these spheres, the strategy must be economicallv viable, and cooperative production -- only possible with the new social ;ittzitude -- is set as the final goal. The formation of ujamaa villages through resettlement, which is one of the more conspicuous features of the execution of the policy, is important in relatively sparsely populated rural areas of Tanzania. However, cooperative production can take place without any resettlement and, particularly in the most densely populated areas, will probably also be developed on that basis. 14. Recently, the definition of in ujamaa village has been changed and the term is now reserved for those villages that have reached full control of the major means of production. Newly-formeci villages would be immediately registered as primary societies (pare 6). 15. The Commissioner of 1'Jamaa & Cooperative Development is charged with the responsibility of developing the ujamaa ideals, and of evaluating the program. There is close coordtnation with TANU and the executive Ministries. The standardization of reporting and assessment of economic performance ;s still poor but these problems are receiving close attention. Technical support is given by various Ministries, especially the Ministry of Agriculture. 16. It is difficult to gauge progress with ujamaa in quantative terms, because this is marked by the acceptance of a cooperative attitude to working and living, as well as by economic success in production. The latter is only now being evaluated by the Tanzanians. A count of ujamaa villages and the population involved in them is some guide to progress, but village  ANNEX 1 Page 6 formation is reported in three stages and the first, unspecific stage, which dominates the statistics, tends to be over-estimated inspite of efforts to improve reporting. Figures of population in villages may also be high due to double counting. It is, however, now estimated that, by mid-1975, some 5,000 villages had been established, with a population of between 3 - 5 mil- lion. The final number of villages may be in the 7 - 10,000 range. 17. Until recently, three stages in ujamaa development were distinguished: (a) A formative stage during which resettled villagers require financial aid for activities such as bush clearing, access road construction, and purchase of initial equipment. In practice, they often also need food, as they cannot always produce sufficient initially. At this stage the Regional Development Funds are the major source of assistance. Villages that are members of primary marketing cooperatives also receive the same servicing facilities as individual members. (b) An intermediate stage in which the village members learn to live and work together, putting increasingly more of their work effort into commercial crops and development activities. At this stage the village adopts a workable constitution, and when it has become economically viable it registers as an Agricultural Association. Credit may then be obtained from the Tanzania Rural Development Bank (TRDB). (c) The final mature stage is reached when the ujamaa village registers with the Registrar of Cooperatives. It may become a full multi-purpose cooperative society, and would have adequate security to attract commercial credit. The assessment of when this stage is reached is carried out by the local Cooperative Officer, Agricultural Officer and TANU secretary. 18. Legislation now before Parliament proposes slightly different classi- fication, which would bring registration as a cooperative forward to the formative stage, and reserve 'Ujamaa' status for the mature stage. 19. Progress with ujamaa village formation varies considerably by region. The differences reflect both the variety of motives for the formation of villages, and the social and geographical conditions of the area. The great- est development has been close to the boundry with Mozambique where the dominant motivation has been defence. Elsewhere the largest number of villages are in areas which have limited involvement in the cash economy, are sparsely populated, and have a harsh environment. These include Dodoma and Kigoma where there has been a concentration of planning effort and technical  ANNEX 1 Page 7 assistance. It is in these areas where the concept of physical movement into concentrated villages (probably located on a reliable water supply), and the introduction of new crops (necessitating rather intensive technical guidance, modern inputs and credit) is a particularly appropriate approach. Because of the real need for services and little individual investment in their present situation these areas accept ujamaa most readily. 20. Progress in numbers has been noticeably slow in other areas, namely: (i) the densely populated highland areas where individually-owned land is under coffee, tea or bananas (such as in Kilimanjaro, Bukoba and Rungwe districts) and where ujamaa will be based on cooperative work on existing crops and is unlikely to involve any resettlement; and (ii) the generally less densely populated areas with established cash crops such as cotton, tobacco, and cashew. These are drier regions than the highlands, with more extensive cultivation and marked seasonal peak labor requirements (traditionally met by hiring). Technical and social problems seem partic-iiarly likely in the cotton area around Lake Victoria, including Geita, where an individualistic attitude, typical of first-generation settlers, exists. An unhurried approach to ujamaa seems especially appropriate in these areas. 21. The economic justification for communal farming rests on th'z greater production arising from economies of scale and division of labor, aae the more economical (and therefore increased per capita) provision of Gover"ment services such as extension advice, water supplies, schools and dispensaries. The case for economies of scale is based on (i) sharing bulk oreers of inputs and group marketing; (ii) making more efficient use of labor and hand or powered implements, enabling a greater area to be cultivated; (iii) allowing specialization of function, such as herding and tool repair in the village; and (iv) enabling wider access to technical advice. The channelling of agricultural credit is assumed to be facilitated by cohesive ujamaa units, as can the economic distribution of packages of inputs supported by extension services. Potentially, ujamaa villages would extend the coverage of coopera- tive marketing organizations, but until they reach a sufficient size to pro- vide marketing services economically, Ujamaa villages would be members of existing cooperative societies. A further argument in favor of ujamaa villages is that a group of farmers will accept ar. Jnnovation to which individuals are indifferent: the validity of this argument has yet to be proved. Also, there is a lack of empirical evidence on where economies of scale are achieveable in commercial production, and the optimum scale of different farming operations in not known. However, it is clear that radical changes are needed in farm management as village production becomes increasingly communal and is carried out on a larger scale. A well-trailed extension service, familiar with or trained in decision making on larg farms, then becomes a necessity.  ANNEX 1 Page 8 22. These management changes will be particularly required for crops like tobacco with a high management requirement, and crops such as cotton which have a high pay-off from timely operations, and are typically included in a rotation with food crops. The application of ujamaa principles to cotton growing, particularly with mechanization, calls for major changes in farm management techniques. These would include new rotations, introduction of minimum tillage techniques, intensification of cultivation by use of sprays and fertilizers, and possibly changes in picking and sorting. Mechanization needs to follow intensification and a reorganization of land-holdings. It cannot economically be imposed on the existing low-yield system. 23. Difficulties within villages-- which are likely to be a continuing concern -- are the lack of leadership and management skills among village members, and problems of measuring the individual's share of communal pro- duction. The former can be helped by planning, training, and assistance from extension services. The most important factor is to ensure that ujamaa villages develop from the desire of the people and at a pace at which they can absorb the new ideas and techniques essential for viable communal produc- tion. TANU 24. The mobilizing force in ujamaa development has been the political party TANU (Tanganyika African National Union). Organized down to cells of ten families, TANU reaches the grassroots of the rural areas, and in its pyramid structure rising through village development committees and ward organizations provides a hierarchy for chanelling both policy decisions and local opinions. At the village Level it provides a forum within which a group of farmers can debate possible cooperation through which technical and planning advice (e.g. on site selection, crops, planting areas) can be channelled, and through which farmers' problems and wishes can be passed to the party and Government. 25. The unpaid leader of the 10-family cell, known as the 'Balozi', is a TANU member elected by party members of the cell. It is his job to keep records of the cell membership, collect dues from party members, and pass on political and policy informatz~on to the cell. He is also to keep the party leadership and -- through 1ANU -- the Government informed on local problems. The Balozi tries to settle local disputes which otherwise would go to the courts. The position is primarily political, but with Party and development matters so closely linked, the Balozi obviously fills an economic role as well. A Balozi may become the chairman of an Ujamaa village formed in his area only l.f the villagers consider his technical knowledge adequate for the post. Although the 10-cell system has not been working uniformly well, partly because the duties of the Balozi have become overburdensome, it does provide an effective two-way channel for information available.  ANNEX 1 Page 9 C. Training Program for Cooperatives 26. As pointed out in Section A, the responsibility for mobilizing the people, once settled in villages, (which are also primary societies) falls on the cooperative movement which unfortunately has many problems. Some of these problems have been created recently, as a result of the Government's social development policy. Personnel within the cooperative movement are not prepared, by experience or training, for their tasks, and this explains the attention Government is now giving to technical assistance and training. 27. The training requirements have been identified at all levels of the cooperative movement, and concern mostly management and accounting, al- though such areas as marketing, storage and transport management also lack properly trained staff. The Nordic Group is assisting Government with the formulation of an overall cooperative development plan, the identification of specific training requirements, and with technical assistance at regional level. The project has been in operation for seven years and at present comprises 26 Nordic personnel. The Project Coordinator is attached to the Prime Minister's Office and the specialist staff and management technicians of the project assist with the work of the Ujamaa and Cooperatives Development Department at headquarters and regional level. In addition eight staff are attached to the Cooperative College at Moshi which provides the training for the staff of the College. 28. At the regional level, the Nordic Group has developed accounting systems for transport and storage, and they have provided for technical assistance to introduce these systems. In addition, they have provided for limited numbers of staff as advisers to management. The progress of this assistance program is slow, and cooperative unions still have a long way to go before they would be able to perform their functions efficiently. 29. The National Maize Project, would, in the execution of its develop- ment activities, make full use of the regional cooperative unions for storage of inputs and their distribution to villages, and for collection and storage of maize. In order to perform these functions effectively, and at low cost, the cooperative unions would need technical assistance in marketing, account- ing and transport management. 30. Detailed requirements have been drawn up, in cooperation with the Nordic Group; they are presented in Appendix 1 to this Annex. The proposal estimates that 8 expatriates would be needed, in addition to vehicles and accounting machines; 4-year costs would be approximately US$2.0 million. 31. Although this assistance would be of great significance to the successful execution of the Project, the component has not been included for the following reasons. Firstly, training of, and technIcal assistance  ANNEX 1 Page 10 to the cooperative movement would place an intolerable burden on Project Management, who would be occupied full-time with the agricultural components. Secondly, the high expatriate staff component could be better provided on a bilateral (grant) basis. However, there is a real danger of complicating the integrated improvement and development approach of the Nordic Team by a proliferation of agencies and/or projects operating in the same field. 32. At the village level, the Project would rely on the ability of village secretaries. Their tasks would include ordering inputs and their storage, storing maize and arranging for timely evacuation, applying for storage construction credit, and in general proper accounting of all NMP related activities. Their task would also include distribution of inputs and receipt of produce to be marketed together with all necessary records in villages, where farming is done on an individual rather than a communal basis. 33. It is recognized that, in general, village secretaries are not able to perform these functions effectively, and that additional training is re- quired. Such training is all the more urgent, now that the entire rural population would be villagised by 1976, and now that practically all agri- cultural development would be based and centered on these villages. 34. The 5th Education Project, proposed for World Bank financing, would address itself to the re-training of cooperative staff at village level. The education project would train Village Management Technicians (VMT) who each would be assigned 5 villages, whilst living in one of them. These VMT's would be trained primarily in cooperative organization and marketing, management-oriented bookkeeping, economics, and mechanical and construction topics; training periods would be for 5-14 weeks. The VMT is expected to introduce accounting systems, developed with assistance from the Nordic Group, at village level, and to give individual village secretaries on-the-job train- ing. The education project would start its VMT training program early 1976, and the first graduates would be available for posting in mid-1976. It has been agreed that VMT's would be posted on a priority basis to wards with NMP villages, so that the National Maize Project might be the first to benefit from this training program. D. Cooperative Marketing 35. Marketing for virtually all food and cash crops in Tanzania is the monopoly of parastatals and cooperatives. Many small-scale producers are members of cooperative societies, which were originally developed as market- ing units for coffee, tobacco, cotton and cashew nuts. In addition, they took on the marketing of surplus food production on behalf of their members.  ANNEX 1 Page 11 36. In most regions the primary societies market crops through a regional cooperative union which acts as agent for the parastatal organiza- tion responsible for the particular crop. Usually, the cooperative union arranges the transport of produce from the villages to the regional centres, where it is delivered to the parastatal concerned. Frequently, the coopera- tive union operates its own transport fleet for this puprose. In some regions, however, where the regional cooperative union is not too effective, the primary societies may deliver crops directly to the parastatal without going through the union. 37. Official policy does not encourage private traders and/or transporters to engage in marketing and transport of farm produce, be it food or cash crops. The consequent lack of competition and the forced pace of development of cooperatives (paras 20-22) and parastatals have created serious inefficiency problems for these organizations. There has been a tendency for the Government to re-organize and re-name parastatals at frequent intervals in an attempt to overcome some of these difficulties. The poor state of organization of several regional cooperative unions and very many small primary societies is a serious constraint to agricultural development. However, the Government is engaged in a major program for cooperative development and training with assistance from the Nordic countries (para 27). 38. In the more developed agricultural areas of Tanzania farmers have had relatively little difficulty in obtaining farm inputs. Larger-scale farmers have been able to obtain supplies themselves from organizations like the Tanganyika Farmers Association (TFA) in the main towns. Likewise, in areas where a prosperous small-scale agriculture had been developed on the basis of high value export crops like coffee, tobacco and cotton, farmers have been well-served by cooperative societies and crop authorities. How- ever, the majority of small-scale farmers in Tanzania are not so well served. 39. Most cooperative unions and primary societies have c.nfined them- selves to marketing farmers' surplus food crop production anc have not made a serious effort to supply inputs to farmers. Part y for thi3 reason, regional authorities in the main maize growing areas have undertaken a maize production program for the last two years under which they supp'lied fertilizers, seeds, insectivides and, sometimes, machinery services to selected villages. Given that all villages will be registered as cooperative societies, that they will all be affiliated to regional cooperative unions, and that they will all be responsible for marketing farmers produce, it seems logical that farm inputs should also be supplied to farmers through the cooperative system. Thus preference will always be given to loaning lorries provided under the Project to regional cooperative unions, as well as for them to control the additional regional storage. As all "villages" will be multipurpose coopera- tives, the village stores will naturallv belong to the village society. October 13, 1975  ANNEX 1 Appendix 1 Page 1 TANZANIA NATIONAL MAIZE PROJECT UJAMAA AND THE COOPERATIVE SECTOR PROPOSAL FOR ASSISTANCE TO REGIONAL COOPERATIVE UNIONS 1. The Nordic Project for ('ooperative Assistance (N-PCAT) to Tanzania has suggested a program of assist.nce to regional cooperatice unions, 1/ which would enable these unions to effectively discharge their functions and responsibilities. The follow:Lng is a summary of these proposals. 2. All the 14 unions in the 13 regions would be included in the assistance proposals made below, even though the National Hilling Corpora- tion (NMC) is currently undertak:ing the maize services, normally carried out by the regional cooperative unions, in 5 of the regions covered by the Scheme. The unions' functions most directly affected by the NHP would be agricultural marketing, input .nd crop transport and accounting, all of which need assistance. Marketing 3. Four agricultural marketing experts would be needed to assist the unions in their marketing operations. One expert each would be stationed in Mtwara, Mbeya, Dodoma and Mosh:i, and would cover the other unions in the NMP as indicated by the four groups of regions in para. 1, footnote 1. The qualifications of the experts would include training in economics and accounting and experience in marketing. The main tasks of the experts would be: i) To assist management of the unions in the planning and operation of all marketing activities; ii) To supervise the keeping of marketing accounts and records; 1/ The 13 regions affected by the NMP have been grouped, and the proposed centre of technical assistance for each group is underlined, as follows: 1. Mtwara, Lindi, Ruvuma; 2. Mbeya, Rukwa, Iringa; 3. Central, Tabora, Morogoro; 4. Kilimanjaro, Tanga, Arusha, Mara.  ANNEX 1 Appendix 1 Page 2 iii) To cooperate with officials of the Ministry of Agriculture in making local and regional crop estimates prior to and during the crop season; iv) To assist union officials in coordination of activities with the National Milling Corporation on tha regional level; and v) To assist with the implementation of cooperative marketing systems, as developed by the Agricultural Marketing Section in the Prime Minister's Office and the Cooperative Union of Tanganyika. Transport 4. Four transport experts would be needed to assist the unions in their transport operations. Since such experts are difficult to obtain, flexibilitv should be exercised in deciding on the type and extent of qualifications re- Suired. The main tasks of the experts would be: i) To assist in transport management, including allocation and service of vehicles, manpower organization and administrative routines; ii) To supervise the keeping o- transport accounts and records, as developed by the Nordic-PCAT: iii) To introduce economic analyses with reference to freight rate structure, costs and long-term budgeting; iv) To assist in conducting training zourses for union transport officers; and v) To assist the Transport Section -i the Prime Minister's Office and the Cooperative Union of Tanganyika in the implementat.an of cooperative transport systems. Accounting 5. A uniform accountancy system, presently beirg -inalized with Nordic-PCAT assistance, would be well suited for mechanization. A pilot scheme with mechanized bookkeeping equipment, involving some 70 simple and robust machines, would be undertaken in one union. If successful, the mechanized accounting system would be introduced in all 14 cooperative unions.  ANNEX 1 Appendix 1 Page 3 Costs 6. Estimated costs for such a 4-year assistance program would be around US$2 million, broken down as follows: Total Unit Cost Number Years Cost (US$) A. Staff Marketing Experts US$35,000/year 4 4 560,000 Transport Experts US$40,000/year 4 4 640,000 B. Material 4-Wheel Drive Vehicles US$ 7,500/each 4 1 30,000 Accounting Machines US$ 5,000/each- 70 1 350,000 C. Miscellaneous Vehicle Operations US$ 0.20/km for 30,000 km 4 4 96,000 Training, Teaching Materials - 1 50,000 1,726,000 Contingencies 274,000 Total 2,000.000  NATiONt1L nrZE PROJECT UJAIQA AND COOPERATIVE DEVELOIUT DEPARTMENT ICONISSIONER oBjECTrVES Provide central government policy and technical leadership to ujamaa and cooperative development. WUJOR TASKS: - Overall administration of the Department. - Act as Pl'Os senior expert on ujsmaa and Cooperative issus. - Perform duties as Registrar. - Administer and review law relatiny to ujamee and Cooperative societies. - Work with the Comisatemer for Regional Planning and Control. - Work with the Director of the DDC Project Team. - Ensure that concerned parties are regularly briefed in ujeasa and cooperative issues. - Maintain working relationships with other institutions. POLICY DEVELOPMENT & RESEARCH SECTION AUDIT AND INSPECTION SECTION MANAGEMENT ASSISTANCE SECTION CARDE DEVELOPMNT SECTION MAJOR TASKS - MAJOR TASKS MAJOR TASKS MAJOR TASKS Translate policy direction provided - Define procedures on legal - Assist and strengthen specialised - Advise Manpower Dev. Division and regions on staff by the President, Prime Minister aspects of registration of activities (e.g., smallacale in- required a.d TAN)l into operational proced- ujamac villages and co-opera- dustries, savings and credit, - Recomend schemes of service and raise related policy ures for implementation by the tive societies, and provide building and construction, retail issues regions and other ministries technical support for Regis- and wholesale, agricultural and - Develop & recommnend policies on secondment of cadre Monitor the implementation of trar function livestock) by officers to movement policy in the regions through the - Maintain register Developing staniard accounting - Forecast cadre manpower req ,ir-ne"ts X propose plans planning process and by maintaining - Advise Registrar on dissolu- and management systems appro- to meet them current knowledge tion of cooperative societies priate for each type of - Propose annual manpower development plan for cadre Analyse ujamaa and cooperaLive - Act as central point of contact activity - Plan career dev. of key cadre officers section of annual performance for Chief Accountant, Audit and Developing manuals that describe - Submit request for allocation of school leavers and report and cou=enting through Supervision Fund, and work how to start up and organize graduates to Manpower Development Division the Principal Secretary to the closely with tonal auditors to each activity - Assist Unified Coop. Services Commission in staffing Planning and Control Department ensure societies and unions are - Provide a management advisory team cooperatives for action audited on time and audit re- to assist individual villages and Seconding staff as necessary Developing annual planning cosmendations followed up societies Assisting in recruitment of people from overseas guidelines for the regions and - Spot-check regional unios' - Act as liaison point at centre for Advising on submissions to high level manpower allo- ensuring they are transmitted, financial and management perfor- individual villages and societies cation coamittee working through the Principal mance in consultation with ROD - Inform Policy Development and - Prepare training programses for regions' use Secretary and advise Registrar on action Research Section of major problema - Manage training of cadre staff in PID Providing assistance in plan required with policy implications - Nominate cadre officers for training at higher education preparation, as required - Process CUT's and SCULT's - Prepare educational publicity institutions locally & colleges overseas Monitoring progress & problems estimate for approval by Regi- material - Act as parent Minist. for NICE through quarterly reports, strar and analyse their final - Assist UCSC determine manag nt - Work vith CUT and Unif. Coops. Service Com. to ensure reports of Zonal Offiders and accounts structure at union level, includ- Cadre and movement training programs are compatible reports from Audit & Inspection - Maintain copies of approved esti- ing staff calibre requirements Funds necessary for ujaisa training are available & Management Asst. Sections sates of unions - Work vith NICE on introduction of lMovement training programs are effective Develop ujamaa & cooperative posi- - Inform Policy Development and new mmgemnt systems in teaching - Propose transfers of cadre officers tion in rural development decisions Research Section of major prob- - Carry out annual promotion exercise Identify emerging policy issues & lems with policy implications - Maintain full and current records of cadre officers recommend the steps required to - Advise Management Assistance resolve them Section on need for system de- Fotmslate & propose longer term sign or changes Dev. strategies for consideration by the Commissioner, PS, PM & TANU Research specific problems & issues as agreed by the Coimissioner Maintain statistics on ujamaa and cooperative dev. & analyze trends annually Source: "The Nordic Cooperative Assistance to Tanzania" Team Septemher 3, 1975.  ANNEX 2 Page 1 TAN 2ANIA NATIONAL M0.IZE PROJECT EXTENSION & TRAINING I. EXTENSION A. Background 1. The extension service is centered on the newly-formed villages, and the ultimate aim is to have one extension worker, a Field Assistant (FA), for each village who would operate as Farm Manager of that village. It will take some years to reach this national objective; but for Project villages provision is made to recruit the requisite additional staff. At ward level an AFO is responsible for supervision of agricultural activities, and he in turn is supervised by the District Agricultural Development Officer (DADO). The DADO normally has between 3 to 4 wards in his district, and his total staff varies between 15 and 60 people (Table 1). 2. The Regional Agricultural. Development Officer (RADO) fulfills more of a service role, since all extension staff are operating in the districts, and are as such responsible to the DADO. However, the RADO influences the agricultural development activities of the DADO's through regular meetings, visits, and discussion and approval, of work programs. He himself is respon- sible to the Regional Development Officer (RDO), who comes under the Prime Minister's Office, and the RADO only receives technical support and advice from Ministry Headquarters. 3. The Ministry of Agriculture (KILIMO) in Dar-es-Salaam is mainly a policy making body with technical staff who assist but do not direct regions in the preparation of annual work programs and their execution (see Chart World Bank 8522 (R)). In addition, KILIMO is responsible for the national research program, agricultural training and identification and preparation of national agricultural projects, which are, however, executed by regional authorities and/or parastatal organizations. 4. Staffing in Project regicns is shown in Table 1; the figures are the latest available to KILIMO but only reflect the 1974 position. At present staff are inadequate in both number and training, and are poorly supervised, which is partly due to the chronic shortage of transport at all levels.  ANNEX 2 Page 2 B. The Maize Project 5. Improving the quality of the agricultural extension services would be critically important for successful implementation of the National Maize Project. The new village system would facilitate this, since the villagers would live together and contact with groups would be easier. Upgrading of extension staff and improving their mobility would be an important component of the NMP, and is discussed in paras. 20-34. 6. Ideally, each village would have a resident extension worker (FA), although during the initial Project years some villages would have to share an extension worker. These staff would ensure that well-run maize demonstra- tion plots were established in every village (one on farmers' fields and one at the school), and that farmers attended regular field days at these plots. Inputs for the demonstration plots would be supplied free under the Project, and yield data would be used in monitoring Project progress. These demonstra- tions would be half a hectare in size, and include the village's recommended package, and the next one to be reached. 7. The Project would also provide funds and transport for the Research Department to establish maize variety/agronomy trials at about four villages in each region, thereby expanding their ongoing program. Information from these trials would be collected by the Regional Farm Management Officers to be appointed under the Project. This would be used to continually improve the production recommendations made under the Project, a task assigned to the Farm Management Section in PSU. 8. Leaflets on improved maize production would be distributed in every village, while mobile film units would show films on maize production at regular intervals in Project village, (para. 16). 9. In each region a senior member of the agricultural staff would be appointed as full-time Regional Maize Project Coordinator (RMPC). He would be responsible for drawing up Project work programs and coordinating the Project within the region and with the Project Servicing Unit. In each dis- trict a field officer would be appointed as the full time Project supervisor; his title would be District Maize Project Supervisor (DMPS). 10. Each Regional MPC and District MPS would be supplied with a vehicle. Assistant Field Officers responsible for supervision of village level staff would receive financial assistance for purchase and upkeep of motorcycles; a similar program would be extended for bicycles for field assistants at the village level. 11. A Farm Management Officer to the Project would be appointed in each region; his title would be Regional Farm Management Officer (RFMO). He  ANNEX 2 Page 3 would ensure that the farming methods recommended under the Project were appropriate within the context of the local farming systems. He would also help to monitor Project progress, and be in close liaison with the Farm Management Section of PSU in Dar-es-Salaam. C. Communications 12. Various communication media have been tried to keep the farmer in- formed of new agricultural techniques and inputs, and to refresh him on hus- bandry details of established crops and/or livestock but existing coverage is insufficient. The Project would strenthen the existing visual and audio- visual programs, by expanding the printing and distribution of KILIMO's monthly agricultural newspaper, and by providing Mobile Film Units. 13. KILIMO publishes a monthly agricultural newsletter in Swahili under the title Ukulima wa Kisasa (Modern Farming), first published in 1955. The paper covers the following major subjects: i) Agricultural and veterinary practices. ii) Nutrition and home economics. iii) Agricultural activities in villages. iv) Entertainment and sports. The average number of copies is 27,000 per month, which are distributed via Regional Agricultural Development Officers, parastatals and cooperatives. A readership survey of several years ago indicated that each copy is read by about 5 people. Even so, it is clear that the majority of farmers never see a copy of Ukulima wa Kisasa. It also appears that very few copies reach the newly established villages. 14. The Project would, therefore, provide financing for an increased production of the paper, which is at present partly subsidized by Government. Most of these additional copies would be distributed for sale to Project villages, through the RMPC's and DMPS's. The cost per copy, a proposed TShs.0/50 against the present TShs.0/30, finances around one-third of printing costs. 1/ 15. PSU would contribute to Ukulima wa Kisasa articles on maize hus- bandry, fertilizer application, etc., as well as on farm management in general. Leaflets and/or pamphlets on these topics can also be printed at the paper's offset printing machines, thus reducing present under-utilization of equipment. 1/ NMP would finance the subsidized portion of printing costs of copies printed in addition to the current 27,000/month, which by Year 4 would total an extra 26,000 copies.  ANNEX 2 Page 4 16. Experience has shown that films are an excellent media for mobilizing the interest of rural people in improving agriculture. A few mobile film units exist, but are insufficient to permit regular visits to the large number of existing villages. Furthermore, an improved mix of films is required. Con- sequently, the Project would provide a new mobile film unit for each of the thirteen Project regions. Each unit would be staffed with a driver, a pro- jectionist and a commentator, who wotuld be trained in Dar-es-Salaam at the Audio Visual Institute (AVI). Finance would also be provided for the hire and/or purchase, of suitable films. Special requirements, i.e. agricultural films in the Tanzania context, would be produced by AVI, and the Project would finance the production of one such film per year. AVI has a modest collection of films which can be hired at no charge. However, with the limited selection available and the intention to increase viewer interest by screening a variety of films, it would be necessary to buy such films for which funds would be provided. II. TRAINING 17. Training under NMP would be restricted to agriculture and manage- ment; although there is a need for training in the cooperative sector, these requirements are being met through the activities of the Nordic Group and the 5th Education Project, which is proposed for IDA financing (Annex 1). Provision has also been made for seminars of senior staff at regional and district level. A. Agriculture 18. The training envisaged for agricultural staff consists of: (i) Agronomy and extension course: for APO and FA (ii) Interphase course: for AFO and FA (iii) Maize production course: for diploma and degree holders (iv) Degree courses: for degree holders. Agronomy and Extension Course 19. The object of the course is to upgrade all field extension staff -- Assistant Field Officer (AFO) and Field Assistant (FA) -- who would be responsible for day-to-day contact with farmers in Project villages. The course would consist of an agronomic section, which would emphasize husbandry and correct fertilizer application; a demonstration section, where layout and importance of the demonstration plots would be taught; and an extension  ANNEX 2 Page 5 section, which would cover methods of communicating with farmers. The curri- culum would be put together by KILIMO staff in the Training Division, by the seconded CIMMYT Adviser and by staff at the Ilonga Research Station. For the communication aspect assistance would be sought from the Audio Visual Institute (Dar-es-Salaam) and the University of Dar-es-Salaam. 20. The trainers of the Agronomy & Extension course (and subsequent refresher courses) would be experienced agricultural officers from the 13 regions of the Project. They would receive a 14-day course at the Ilonga Research Station. 21. The Project would attempt to train all field staff in villages affected by the Project in Year 1, before the start of the 1975/76 season. Some 540 villages would be reached in Year 1, increasing to 950 by Year 4 of the Project. It is estimated that 50% of the villages have one extension worker each, whereas in the remainder one extension agent covers 2 villages. This would give a total of 405 FA and AFO to be trained in Year 1. All AFO at ward level, who supervise the village staff, would also need to attend this course: on average there are 10-12 villages to one ward level AFO. Many villages will not be included in the Project and, therefore, each AFO may only be responsible for five to six Project villages, and, hence, around 95 would need to be trained. Thus, in Year 1 a total of 500 FA and AFO would participate in the Agronomy & Extension Course (Table 2). 22. Using a ratio of 15-20 students per course, some 26 courses would have to be organized in the 13 regions. Trainers for these courses -- senior agricultural officers in the regions -- would be trained at the Ilonga Research Station. There would be one trainer per region, with the exception of Iringa and Mbeya, who would have 2 trainers each due to the large number of students to be trained. In each of the subsequent years (Project years 1976/77, 1977/78 and 1978/79) a total of 110 AFO and FA would be trained, to reach staff in new villages covered by the Project; these numbers include ward-level AFO's. 23. The aim of the Project is to reach an establishment of one extension worker per village. Therefore, new staff would be recruited, mostly Standard- 12 school-leavers, who would be given a 3-month training course. Provision has been made for 140 trainees in Year 2, and another 100 in Year 3; courses of this type are already being organized by KILIMO, but the Project would provide for financing of this training program. 24. The Project would also organize annual refresher courses for all field staff; during these 12-day courses any new developments in maize agronomy could be covered, the experience of the previous season reviewed, and emphasis would be given on extens:Lon methods. The senior agricultural staff in charge of the Agronomy and Extension course would also teach the refresher courses.  ANNEX 2 Page 6 Interphase Course 25. This course would last one entire growing season, and be given at the maize research stations. Extension agents (AFO and FA) would come for 2-3 weeks at the start of the season, and would be trained in phase 1 (land preparation and planting) of the maize calendar. They would then go back to their villages to implement this knowledge, after which they would return to the research station for training in phase 2, which would include weeding, and the application of top dressing and pesticides. Finally, the third period would comprise yield estimation, harvesting and marketing and some training on the economics of maize production. 26. The "interphase" approach, which has been successfully tried in other countries, has several advantages: (i) The agent is not completely taken off-duty; (ii) There is an immediate follow-up in practice of the teaching and demonstration at the research station; and (iii) There is an immediate interaction between extension and research. 27. This type of course could take a maximum of 15 students per research station; included in the program would be the Ilonga Research Station (Kilosa), the Ukiriguru Research Station (Ukiriguru) and the Agricultural and Research Training Institute (ARTI) at Mbeya. Thus 45 students could be trained per season, for a 4-year Project total of 180 extension staff. CIMMYT has agreed to make a Senior Training Officer available to organize this type of training in Tanzania. 28. Trainers for the 1st year's interphase course would be recruited from research staff at the above mentioned research stations. In subsequent years certificate and diploma holders who had successfully graduated from the CIMMYT or IITA courses (para. 29) would be able to assume full teaching responsibility. Maize Production Course 29. A few well-qualified degree holders would be sent to Mexico (CIMMYT) or Nigeria (IITA) for a season (5 months) course of intensive maize production. These students would then become the trainers of extension agents for the interphase course. 30. In addition, graduates would be employed on maize research stations, particularly Ilonga, Ukiriguru and Mbeya. KILIMO has produced a staffing plan, aimed at providing the technical manpower to make Tanzania self-sufficient in maize production research. This national staff development plan emphasizes production-oriented research, and the IITA/CIMMYT courses would be the first  ANNEX 2 Page 7 stage of this program. Consequently, the NMP would provide training funds for 21 intensive maize production courses, to be taken at IITA (Ibadan, Nigeria) or at CIMMYT (Mexico). Since the IITA is concentrating on maize and legumes, and also on the adopticin of improved maize varieties to African conditions, most students would go to IITA. Degree Training 31. The manpower analysis referred to above (para. 30) also points to the need for B.Sc. and M.Sc. degree holders. The Project would assist in providing some of these staff, by financing training, as follows: 1975/76 1976/77 1977/78 1978/79 Total B.Sc. (2 years) 3 3 2 1 9 M.Sc. (2 years) 1 2 3 3 9 32. In the past, most degree training has been undertaken in American or European universities, which renders more difficult the transition to Tanzania, and the application of the training to tropical agriculture. There- fore, the NMP would provide financing only for degree courses taken at African universities. B. MIanagement 33. NMP would be implemented at the regional level by regional and dis- trict agricultural staff (RADO and DADO) and by Project-appointed Regional Coordinators and District Supervisors. These staff, though usually well-trained in agriculture, have had little cr no management training and, thus, they are not always able to effectively organize and supervise their staff. 34. Consequently, the Project envisages the organization and financing of a management training course (some 3 weeks) for RADO's (13), DADO's (28), Regional Coordinators (13) and District Supervisors (28). These 82 staff could not be trained all at once; some 45 would be taken in in the first half of 1976, and another 45 in the latter half of 1976, with 10-day refresher courses to begin early 1977. The training of 90 staff takes into account transfers. 35. The management course, which would consist of case-studies and demonstrations, in addition to lectures, would be organized and given by the Tanzanian Management Development Institute at Morogoro.  ANNEX 2 Page 8 36. In addition, the Project would provide financing for regular seminars, to be convened by the Project manager. The purpose of these meetings would be to discuss NMP work programs and to develop solutions to problems of implementation. October 13, 1975  ANINU 2 TANZANIA Table 1 3ATIONAL MAHXZ PROJECT ErMnD u uosSTAFF - Plow AND DISIRICT 173e 11COk M Region/DlxtritC Agrictul- riald Ai t nt Filil totel RegignfDintrict Arictl- pield Aeeietant FilId Total FIed Field P.1 Toa Neie tttt- ua f la lad ..i zf,_ offi. L. St 0 OffL ~~~~~~~~~~~~~~~~~~Aaaitetae Staff Officer Officer Offer AsetAtt Staff Officer OfficerA. R.gisoal Staff 3 2 7 - 12 Regionol Stff 1 4 16 14 35 District Staff - Ar-earo0 - 4 19 8 31 DLacrit Staff - Kiltro - 1 5 8 14 HlD ag* - 3 12 10 25 KiloecO - 3 20 21 44 Mboloc - 2 10 13 27 Mhi*fg. - I 1i 10 22 MHduli - 2 5 3 10 HorogorIo 1 3 16 1 38 Toa.l 3 13 53 36 105 Total 2 12 68 71 153 Joco.--- Mcvar qrgio-al S-ff 2 a 6 9 25 Roional Staff 1 3 13 9 26 Xit-riot Staff - D7ds - 7 48 11 66 Ditrikt Staff - rSaaai0 I 1 14 18 34 itodoa - 2 30 19 51 Hr -te - 4 10 18 32 Mgoap.,a - 7 9 9 45 Naal. - 3 11 14 28 Total 2 24 113 48 187 Total 2 11 48 59 120 tRalocal Staff 1 3 7 1 12 Regioal Staff - 3 2 - 5 Datrlict Staff Inioga0 - 4 21 28 53 Diatrict Staff - Hp nda - 3 13 11 27 Hufi.dio - 4 15 16 35 Sonba gac 1 2 16 12 31 Ojonibc - 6 21 35 62 Total I a 31 23 63 To-aI 1 17 64 80 162 Ruv f-egioal staff 1 4 12 5 22 Raglo-al Staff I a 8 2 19 Dietrict Staff - Mbirpg - 1 22 9 32 Merri- t Staff -Mfhai 1 4 32 71 108 S."...a 1 1 27 12 41 fore - 6 29 34 69 Tondoro - 2 24 9 35 Ro.b0 - 1 S 16 25 Total 2 8 85 35 130 Total 2 19 77 123 221 Tabora Li-di Regioral Stff I 1 13 S 23 itotiocol Staff 4 4 -loDi.trict Staff - NM9g - 6 20 14 40 Macriot Staff - Kilaw - 2 9 9 20 Tbor - 2 23 12 37 Llndil - 3 14 19 36 Urboc (rely ford diatriot; ea.ff i-ccidad order NahinRg-.N 1 2 15 12 30 Tabor- District) Total 1 42 40 94 Total 1 9 56 34 100 'fDa- Tape RegLoral Stff 2 4 4 1 11 Regional Stoff 1 4 19 10 14 co t-i Staff - Chuny - 2 14 9 25 Di-trict Staff - Handei - 1 16 24 41 fyrla 1 a 3 12 forego. _ I 11 14 26 Sbryat - 3 25 15 43 Lu.hoto* _ 7 54 it 72 irlotic - 2 1! 17 30 *zheaa 2 23 24 49 R ngre - 3 13 20 36 PanRgni - 1 5 8 14 rOr-1 2 15 75 65 157 Kl Total 1 16 128 91 236 I/ Ar-rr ir iodi-ta NM Dierricat. otporo-r.o- ov3llebl5 for thit Ragios. sopc;r'cr 3, 2971.  TANZANIA NATIONAL MAIZE PROJECT TRAINING OF EXTENSION FIELD STAFF No. of Field Staff Courses No. of 50% of Villages, 50% of Villages, Ward Level Total to (15-20 Villages 1 Field Staff 1 Field Staff AFO, One per. be Trained Students Region Year 1 Each for 2 Villa&es Total 5-6 Villages in Year 1 per Course) Arusha 68 34 17 51 11 62 3 Dodoma 19 10 5 15 3 18. 1 Iringa 88 44 22 66. 19- 85 4 Kilimanjaro 36 18. 9 27 9 36 2 Lindi 31 16, 8 24 5 29 2 Mara 34 17 8. 25. 6. 31 2 Mbeya 60 30., 15 45 10 55 3 Morogoro 46 23, 12 354 7 42 2 Mtwara 42 21 10, 31. 6.- 37 2 Rukwa 24 12 6, 18- 4 22, 1 Ruvuma 24 12 6 18^. 4 22 1 Tabora 26 13 6 19 4 23 1 Tanga 42 20 11 31 7 38. 2 TOTAL 540 270 135 405 95 500 26 September 3, 1975.  ANNEX 3 Page 1 TANZANIA NATIONAI, MAIZE PROJECT TECHNOLOGY AND POTENTIAL FOR MAIZE PRODUCTION Introduction 1. Maize is the most important cereal crop in Tanzania and the staple food of much of the rural population. Varying estimates place the area under maize at 0.8 to 1.2 million ha with average yields of about 700 kg/ha. The crop is grown in all regions of Tanzania, but in some areas conditions are marginal. The bulk of production comes from Arusha, Tanga, Dodoma, Iringa, Mbeya and Tabora Regions; other regions of increasing importance include Kilimanjaro, Morogoro, Ruvuma, Rukwa, Lindi, Mtwara and Mara. Table 1 of Annex 5 indicates the relative importance of maize as a food crop in Tanzania, while Table 2 of Annex 5 shows marketed production from 1966 to 1973. 2. With only an estimated 25% of total production marketed, maize is primarily a subsistence crop. Cultural practices are often poor, resulting in the low national average yielcl estimate of 750 kg/ha. Yields, however, vary widely throughout the country and from season to season with a range of 600 kg to 1,800 kg/ha. Under good husbandry in suitable areas yields of up to 5,000 kg/ha are obtained, and the potential for yield improvement is, therefore, very good. Furthermore, given the wide adaptability of maize, the crop is better suited to Tanzania's variable conditions than any other cereal. However, because Tanzania has such an extremely varied environment, and although on-going basic research work has provided sufficient data for formulating maize packages for irmediate use, their application to particular areas would be refined as more production data becomes available. Maize Areas 3. In Tanzania, generally, the yield of maize has been found to increase with higher altitude and more rainfall, with a longer growing season and with moderate temperatures. Top yields of maize can be expected from areas with over 1,000 mm of rainfall well distributed throughout the growing season on a fertile soil. Good crops can normally be produced with over 750 mm of rain- fall, but below this level of precipitation, and especially if distribution is poor within the season, yields will be reduced and crop failures can be expected. The altitude is also a key determinant in the length of the growing season, which increases with elevation, and maize breeders have evolved varieties suited to four broad altitudinal zones, as well as for varying rain- fall characteristics. The zones, and general guidelines on the number of days to reach maturity, are shown below:  ANNEX 3 Page 2 Altitude Zones and Length of Growing Season Altitude Days from Planting Days from Tasselling in meters to Tasselling to Maturity Below 300 44 - 55 50 - 60 300 - 900 55 - 75 60 - 70 900 - 1,800 85 - 115 70 - 80 Over 1,800 115 - 125 80 - 90 4. Considerable research on maize breeding has been done within Tanzania at the Ukiriguru (Mwanza Region) and Ilonga (Morogoro Region) Research Stations, and by staff of the East African Agricultural and Forestry Research Organization. Varieties have also been introduced from Kenya, Zambia and elsewhere. Ukiriguru and Ilonga have both produced composites now in general supply for farmers, while Kenya hybrids are favored for the higher altitudes where they produce better results than composites. In more marginal areas, the short-season Kenyan Katumani Composite has been introduced. Table 1 summarizes the 1970-72 findings of research variety trials on the effects of altitude. Further varieties are now being evolved and tested by the research stations for possible introduction in two or three years' time. Present indi- cations are that these new varieties will give substantial yield improvement, and they include a short-season variety, for use in areas where returns from maize are at present marginal. 5. In selecting areas suitable for the Project, emphasis has been placed on those regions and districts that are environmentally most suited to maize production. However, areas which might be regarded as having intermediate potential but where maize is still a major crop have also been included to ensure that these areas reach at least self-sufficiency in maize production, and that field research is carried out in areas faced with climatic problems in certain seasons. At the same time, other areas which are only marginally suited to maize growing, given present technology, have been excluded from the Project. 6. Although published physical resource data in Tanzania is limited, the Bureau of Resource Assessment and Land Use Planning (BRALUP) of Dar-es- Salaam University has published a paper on the Agro-Economic Zones of Tanzania, which is supported by a more detailed series of publications for different geographic regions of the country (see Bibliography). A further BRALUP publi- cation summarizes mainly in map-form rainfall and evaporation data. From these sources, it has been possible to derive Table 2, which shows into which rainfall and altitude zones the majority of the selected Project districts fall. The highest potential maize areas are those of the Southern Highlands  ANNEX 3 Page 3 in Iringa, Mbeya and Ruvuma Regions, the higher elevations of Arusha and Kilimanjaro Regions, and the Usambara Highlands in Tanga Region. The remainder of the Project area has a medium potential for maize, including the inter- mediate elevation and rainfall areas of Central, (Morogoro, Dodoma, Tabora), Northern (Mara, Arusha) and North Eastern (Kilimanjaro, Tanga) Tanzania, and the high rainfall coastal and hinterland areas of the South East (Mtwara, Lindi). Table 2 also shows that only five districts in Arusha, Tabora, Dodoma and Rukwa Regions have areas where annual rainfall is below 750 mm, and villages falling within these areas of the districts will be excluded from the Project. However, since these districts are all important maize growing areas, it is regarded as vital to study the problems of maize growing in areas of lower production potential, often only slightly above 750 mm average rain- fall, and make recommendations for yield improvement or changes in cropping patterns. Research 7. Considerable research on maize production has been carried out in Tanzania, based on a comprehensive network of research stations and substations. There are now nine agricultural research institutes and twenty three agri- cultural experimental stations in the eighteen identified ecological zones of Tanzania (Map IBRD 10799). Recently, KILIMO's research activities have been internally reviewed, and a new system of Crop Research Coordinating Conmmittees introduced to streamline and more effectively coordinate national research activities. 8. The present maize research program is evolved through the Maize Research Coordinating Committee, and has the following major elements: i) Maize variety trials to evaluate present and new varieties at 24 locations; ii) Introduction and progency testing of new varieties; iii) Village research trials in all maize growing regions combining varieties, fertilizer use and the cultural practices; iv) Cultural practice trials on seedbed preparation, time of planting, spacing, weed control and use of herbicides; and v) Fertilizer trials. Cultural Practices 9. A good knowledge exists of recommended cultural practices in most ecological areas and more attention to these alone could produce considerably higher yields, even without using fertilizer. At present  ANNEX 3 Page 4 prices, it is generally not economic to use fertilizer without improved practices. Thus, the first package recommended under the Project (para. 21) concentrates on improving cultural practices, combined with improved seed, and pesticides to control stalk borer, which are relatively cheap, easy to use, and technically proven. Improved Seed 10. For the higher altitudes, hybrids developed in Kenya are generally recommended, and these are being produced locally for the Tanzania Seed Company. For medium and lower altitudes, the Ukiriguru and Ilonga composites have been developed locally, and are being multiplied on Government seed farms and also by individuals or companies under contract to the Tanzania Seed Company. For low-rainfall areas, the Kenya Katumani Composite is recommended, and is being multiplied on a Government seed farm at Arusha. The Government is receiving USAID assistance to its seed multiplication program, and the Tanzania Wattle Company (TANWAT) is the major company involved with seed production outside Government seed farms. TANWAT is partially financed and receives technical assistance, including some resident staff, from the Commonwealth Development Corporation, and TANWAT is also the managing agent of the Tanzania Seed Company. On the research side, the Government receives assistance from two international research organizations, CIMMYT in Mexico and IITA in Nigeria, and with this technical assistance with the breeding program promising new varieties are at present under trial. 11. Despite this external support for breeding and seed multiplication, the existing National Seed Production Committee has not proved fully effective in coordinating seed production to meet national requirements. It is essen- tial for an executive subcommittee of the main committee to define precisely national requirements, and then to ensure that seed production from all sources is geared to meet the demand. Thus for maize, as also for other crops, a specific seed production program must be evolved and closely monitored with an effective inspectorate for assisting private growers. Government and regional seed farms and assuring the maintenance of correct standards. Prices paid to growers also need review to ensure the additional effort required to produce seed is sufficiently rewarded. An essential element of the proposed Project would thus be the effective coordination of all parties involved in seed supply. 12. Although composite seed can be retained for use over two or three seasons, it has been decided to replace composite seed annually under the Project, for reasons of a slight genetic yield loss, an even greater germina- tion loss due to poor storage, and a yield increase of up to 10% from con- tinuous improvemenit achieved each year in the composite selection carried out at the Ilonga Research Station. It would take one or two years before composite seed production has increased to Project requirements. In the early years of the Project, when some composite seed would have to be re- tained, a seed dressing, sinuthion, would be nade available under the Project for use by villagers, to protect the seed in store from insect pests.  ANNEX 3 Page 5 Fertilizer Use 13. In many areas of Tanzania more than one essential element is lacking in the soil, and fertilizer must be applied to obtain optimum yields. Generally speaking, nitrogen and phosphorus are required together (and their interaction is often important), while in some areas potassium is also essential if yields are to be raised above subsistence levels. Fertilizers, however, are relatively expensive and in view of the poor husbandry standards and low average yields throughout Tanzania, and the inadequacy of information on which to base localized fertilizer recommendations, the Project would initially broadly follow the recommendations of the Government Soil and Fertilizer Use Coordinating Committee. These recommendations for Project districts are summarized below: Recommended Fertilizer Rates for Project Districts N P K N P K N P K N P K 30-30-0 30-20-10 20-20-0 10-10-0 150 Kg/ha SA and 150 Kg/ha SA 100 Kg/ha SA 50 Kg/ha SA 65 Kg/ha TSP 50 Kg/ha TSP, 44 Kg/ha TSP 25 Kg/ha TSP 24 Kg/ha M. Potash Tarime Njombe Iringa Arumeru (part) Serengeti Mufindi Morogoro (part) Tabora Songea Kilosa Urambo Mahenge Mbeya Mpwapwa Morogoro (part) Sumbawanga (part) Sumbawanga (part) Nachingwea Mbozi (part) Newala Lindi Mtwara Rombo Masasi Arumeru (part) Hanang Nbulu Korogwe Lushoto Source: Draft Handbook for Fertilizer Use in Maize in Tanzania. Ministry of Agriculture 1974.  ANNEX 3 Page 6 14. Much higher optimum rates are listed for the better areas, but the draft handbook stresses the high cost of such packages. For instance, at present fertilizer prices (revised from 1974 publication) it would cost TShs. 1,900 to apply a recommended optimum application of N9OP60K20 per hectare and require at least an additional 2,400 kg/ha yield increase to break even on the extra cost. Such a yield increase is unrealistic for farmers just being introduced to improved practices, and hence the lower rates shown in the above list have been recommended. These rates are broadly in line with the modest fertilizer recommendations in the Project packages discussed in para. 20, which comprise 100 kg/ha SA and 50 kg/ha TSP in Package 2, and 150 kg/ha SA and 50 kg/ha TSP in Package 3. 15. The benefits of the modest rates recommended by Government and utilized in the proposed Project are supported by results of a series of trials on farmers' fields in ten districts of Northern Tanzania that are documented in the EAAFRO Journal for 1969. Compound fertilizers used in the trials were Diammonium Phosphate (18N-47P), a 25N-10P Compound, and a 20N-20P-10K Compound. In every trial the application rate was only 112 kg/ha (100 lbs per acre). The response levels indicate yield increases in the range of 300 - 500 kgs for the more marginal areas, which would not be included in the Project, and from 600 - 1,300 kgs for higher potential areas, in some instances reaching 1,800 -1,900 kgs per hectare (Table 3). 16. Using yield data from Project village demonstration plots, yield surveys of Project farmers' field and further research trials, fertilizer recommendations for the differing ecological conditions existing within the Project area would be developed during the course of the Project. Attention to this aspect would be a key responsibility of Project Servicing Unit staff and the regional farm management officers, who would also draw on the assistance of all those engaged in maize agronomy in the country. Dependent on economic considerations, it seems possible that at elevations above 1,500 meters, where yield potential is greatest, a higher rate of fertilizer application could be recommended for the more successful villages. Evidence supporting this is available from six maize experiments conducted at six off-station locations in Mbeya Region by Dr. Mattson and Mr. H. Zuberi in the 1974-75 season. The data presented in Table 4 show a significant increase in returns, when the level of N is increased from 50 to 100 kg/ha, using unsubsidized prices. Further experimentation during the 1975/76 season will assist in determining the need for any changes in the recommendations. In addition to further technical assistance from CIMMYT, the Project also pro- vides for the service of an experienced maize agronomist to review existing maize agronomy in different areas of Tanzania, and make recommendations for its long-term improvement (Terms of Reference are given in Annex 7). 17. Present fertilizer recommendations are based on using Triple Superphosphate (TSP) as the phosphorous source, and initially Sulphate of Ammonia (SA) for nitrogen. Both are produced by the Tanzania Fertilizer Company (TFC) at Tanga. In view of SA's soil acidifying effect, and the  ANNEX 3 Page 7 high cost of transportation per unit of nitrogen, a more suitable form of nitrogen, such as CAN or Urea combined with the use of Diammonium Phosphate (DAP), would be investigated under the Project. The TFC factory could be modified to produce DAP but, until this is done, DAP requirements would be imported. Pesticides and Herbicides 18. The major pest affecting maize in Tanzania is stalk borer, and the use of 5% DDT (powder) as a preventative measure is included in the recom- mendations for all areas in the Project. Research is ongoing in neighbouring countries to replace DDT by an equally suitable and cheap insecticide; once a recommended alternative is available it would be adopted. Localized out- breaks of army worm are another problem, and provision is made under the Project for their control by 25% DDT (liquid) on up to 10% of the Project area. 19. Since each farmer under the Project is only assumed to grow an average 0.75 to 1 hectare of maize, he would have sufficient labour for adequate weeding. However, as maize is only part of his farm program, and as some more successful maize growing villages might wish to increase the acreage under maize, the Project would arrange for two percent of the area to receive a herbicide - Gesaprim is at present recommended - on a pilot basis. Villagers using Gesaprim would be expected to bear the full cost to ensure that its use releases labour at a true economic cost. Weeding is undoubtedly a constraint on the size of holding for many farmers, and the effectiveness of the herbicide has been well demonstrated on research trials and by some villages, where Gesaprim has already been introduced. The Project Technical Packages 20. The national average yield for maize is estimated at 750 kg per ha. However, since the Project would concentrate its efforts only in the better maize growing areas, it is assumed that average Project area yields are about 1,100 kg/ha; this level has been used as the yield without the Project. In view of the present low husbandry standards, the initial extension effort would be concentrated on improving husbandry methods including early planting, correct spacing and timely weeding. The initial stage (Package 1) couples this with the use of improved seed and herbicides. Once this stage has been mastered, farmers would move on to Packages 2 and 3 which incorporate modest increases in fertilizer use. 21. The Project packages are as follows: (i) Package 1: Farmers using this package would receive improved composite seed and insecticides, mainly 5% DDT for stalk borer control as a blanket recommendation, and also some 25% DDT to combat incidences of army worm outbreaks, provided on the basis of requirements for 10%  ANNEX 3 Page 8 of the cropped area. These inputs coupled with the use of improved practices advocated by the strengthened extension services, would be expected to raise average yields to 1,500 kg/ha (36 percent) over a three- to four-year period. (ii) Package 2: Inputs for this package would include those in Package 1 together with modest use of fertilizer (50 kg/ha of TSP and 100 kg/ha of SA). The SA would be applied half at planting time, and half as a top dressing. The expected yield from this stage would be 2,200 kg/ha, which would further increase yields by 700 kg/ha. At higher altitudes hybrid rather than composite seed would be used. (iii) Package 3: In the more favored growing areas with rainfall over 1,000 mm, a third stage would be introduced adding one additional 50 kg bag of Sulphate of Ammonia, or its nutrient equivalent. The nitrogen would be applied half as a basal and half as a top dressing. The eventual yield increase for this stage is placed at 2,700 kg/ha, in part attributable to the increased fertilizer use and in part to restricting this stage to only the best growing areas. 22. Although Package 1 realizes a 400 kg/ha yield increase, it is envisaged that all villages would move to Package 2 over a two- to five-year period. The yield increase could not be maintained in Package 1 on a con- tinuous basis, and the advance to Package 2 is necessary to maintain fertility, as well as to further raise yields. Other Technical Improvements 23. In addition to the various forms of assistance discussed under extension and training in Annex 2, the Project would support technical innovation in the following ways: i) By making additional funds available to the Research Section of the Ministry of Agriculture to expand their network of village trials from the existing 34 up to 52 trials per season. These village naize trials include a variety trial to compare local and recommended varieties with new ones; a fertilizer trial; a management trial incorporating various cultural practices including herbicide use; and also a research management block to demonstrate existing recommendations; ii) By making additional funds available to the Research Section for conducting rotational maize trials with legumes as well as intercropping trials with the object of reducing the dependence upon imported nitrogen. In this sphere the findings  ANNEX 3 Page 9 of the complementary FAO/UNDP Assistance to Crop Production Project would provide an important input (Annex 7); and iii) By providing consultancy funds for a maize agronomist to review existing agronomic recommendations for maize for the various areas under the Project and make recommendations for the long-term improvement of maize agronomy. It is anticipated that the consultant would be in the field for four months in Year 1, and a further two in the second year after more data is available on which to review recommendations.  ANNEX 3 Page 10 BIBLIOGRAPHY References Date 1. Agro-Economic Zones of Tanzania, BRALUP Res. Paper 25 1973 2. Agro-Economic Zones of: i) N.E. Tanzania BRALUP Res. Report 13 1971 ii) Southern Highlands BRALUP Res. Report 22 1971 iii) Kigoma and Tabora Regions BRALUP Res. Report 46 1971 iv) Dodoma and Singida Regions BRALUP Res. Report 47 1971 v) Coast and Morogoro Regions BRALUP Res. Report 48 1971 vi) Southern Tanzania BRALUP Res. Report 49 1971 3. Rainfall and Evaporation in Tanzania, BRALUP Res. Paper 24 1973 4. National Agricultural Research Program 1974-75, Ministry of Agriculture 1974 5. Response of Maize to Application of Compound Fertilizers on Farmers Fields in Ten Districts of Tanzania, Anderson, EAAFRO Journal, January 1969 6. A Handbook for Fertilizer Use in Maize in Tanzania, Draft Edition, Cusani-Visconti and Samki, Ministry of Agriculture 1974 December 1, 1975  ANNEX 3 Table 1 TANZANIA NATIONAL MAIZE PROJECT RESEARCH MAIZE YIELDS IN 2A1ZANIA. BY ALTITWI AID VARIETt. 19-070-1971/72 1/ 1/ Altitude Nae of Composite Adjusted Adjusted LocatiormTested Range Vait rHybrid Yed ils o (meters) Variety orAll Sites 50% (Bags/ 1970 1971 1972 (bags/acre) acre) Ilonga Coup Coup 12.2 15.8 4 4 2 Ilonga Coup A Coup 13.9 17.8 4 7 2 Below 300 H 512 Hybrid 12.4 15.1 - 3 2 H 511 Hybrid 11.0 14.3 - 3 - SR 52 Hybrid 13.8 17.6 - 3 - H 632 Hybrid 12.0 15.4 4 7 2 Ilonga Comp Coup 15.4 19.4 7 1 2 Ilonga Coup A Coup 15.1 19.6 7 8 1 300-900 H 512 Hybrid 14.2 18.2 - 4 2 H 511 Hybrid 13.0 16.1 - 5 - SR 52 Hbrid 16.7 17.0 - 4 - H 632 Hybrid 14.5 20.3 7 9 2 Ukiriguru Comp Comp 17.7 26.5 12 13 5 H 611 Hybrid 18.9 25.2 12 5 5 900-1800 H 632 Hybrid 17.9 23.3 12 13 5 H 512 Hybrid 20.5 25.0 - 6 5 H 511 Hybrid 25.7 27.4 - 6 - SR 52 Hybrid 21.6 27.8 - 6 - Ilonga Comp A Coup 9.9 - 3 4 3 2/ Iloaga Coup CoUp 9.6 - 3 3 4 900-1800 Katuabili Coup 10.0 - - - 3 H 512 Hybrid 9.3 - - 2 3 Low SR 52 Hybrid 8.4 - - 2 - rainfall M 6 Coup 7.7 - - - 3 Katummni Comp B Coup 6.6 - - - 3 H 613 Hybrid 28.3 34.4 - 3 2 H 611 Hybrid 25.9 32.8 - 3 2 Above 1800 Ukiriguru Comp Comp 23.4 26.5 - 5 2 H 512 Hybrid 23.3 28.1 - 3 2 H 511 Hybrid 24.1 31.1 - 3 - H 632 Hybrid 22.5 24.8 - 5 2 1/ Figures are adjusted for seasonal variations according to yield by standard varieties. Yields are expressed as 90 kg bags of shelled grain. Differ- ences between yields are not necessarily statistically significant. 2/ Sites selected from several altitudes, all with low rainfall. September 3, 1975.  TANZANIA AMNEX 3 NATIONAL MAIZE PROJECT Table 2 MAJOR ALTITUDE AND ANNUAL RAINFALL ZONES OF PRDJECT DISTRICTS Altitude Over 1,800 metres 900-1800 netres 300-900 metres Under 300 metres PROJECT PROJECT Annual 1,000mm 750- >750mo 1,000mm 750- 750 (1, 000m 750 >7501 (1,0OOOmM 750 > 750mm REGION DISTRICT Rainfall 1, 'OODC1,00000 1,0001ma l,000 _ __ ARUSNIA Arumeru X X X X Hanang X - I _X Mbulu __ _ _ _ _ _ _ _ _ _ X I_ __ KILIMANJARO Moshi X X X I _ Rombo _X X X _ TANGA __hez __ _ .. X X X Korogwe X X X_ Lusboto X _X X MOROGORO Kilosa X X _ Morogoro _ __X DODOMA _____ _____ ____ _____ __Mabenge X =X TABORA Tabora _ _ X X UramboX IRINGA _r_I _ X XX x _I _--_---_ Mufindi X X X _ MBEYA Mbeya_ _ _ __ X__ _ _ __XX Mbozi X, _ LINDI Nachingwea _ X_X_X MTWARA M__asi X X _ Mtwara x , =-? _ _X X_ RUVUMA Song_ea X X _ MARA Sumbawanga X _ _X X _ Serengeti I X Recommended Seed Dy Zone Hybrid Hybrild - ybrid UCA IX IC IC Katumant IC IC _ 632 632 632 Katumani UCA UCA IC [C N. B. At least 10% of the District area ust fall within the zone cited for inclusion of that zone to be Marked on the table. Villages in those parts of Districts vith annmul rainfall below 750im will be excluded frem the project. Source: Agro-Economic Zones of Tanzania and Rainfall Data - BHALUP, 1973. September 3, 1975.  TANZANIA NATIONAL MAIZE PROJECT MEAN Y1TLDS AND RESPONSES OF MAIZE GRAIN FOR LOCALITIES IN EIGHT DISTRICTS OF TANZANIA, 1966 District Locality No. of Yield (Kg/Hectare) Mean Responses Trials Unfertilized Fertilized Mean Kg/Ha Per Cent DODOMA Matambulu 6 590 1,025 307 435 73.8 Nkungu 5 309 665 488 355 114.9 Mndemu 3 610 855 733 245 40-3 Makulu 6 399 790 595 391 98.0 Mean for District 20 478 834 656 356 74.6 MBULU Hareabi I 12 3,421 3,865 3,643 444 13.0 Hareabi II 8 1,579 1,926 1,753 346 21.9 Kainam 6 2,611 3,402 3,007 791 30-3 Endabash 6 2,883 4,228 3,556 i,345 46.7 Mbulu town 4 3,708 4,349 4,029 641 17.3 Mean for District 36 2,841 3,555 3,198 714 25.6 SOUTH Uru 7 2,014 2,794 2,405 779 38.7 KILIMANJARO Machame 9 3,216 4,143 3,680 927 28.8 Kibongoto 7 2,847 4,236 3,542 1,389 48.8 Masama 4 3,i64 4,561 4,i6_ 79 21.2 Old Moshi 8 1,590 1,887 1,739 297 18.7 Central Hai 8 2,094 2,949 2,522 855 40.8 Mwika 6 2,762 3,726 3,244 964 25.9 West Vunjo 6 2,316 3,092 2,704 776 53.5 Mean for District 55 2,576 3,424 3,000 847 32.9 ARUSHA District 6 1,905 2,677 2,291 772 40.6 EAST KILIMANJARO Keni and Mkuu 5 657 1,528 1,093 871 132.3 MUHEZA- Muheza 11 1,961 2,799 2,380 839 42.8 KOROGWE Korogwe 4 1,397 1,999 1,698 602 43.0 Mombo 7 1,037 1,451 1,244 414 39.9 Mean for District 22 1,464 2,083 1,773 619 42.3 LUSHOTO Mtae 7 1,767 2,235 2,001 469 26.5 Mgwashi 7 2,652 4,608 3,631 1,956 73.8 Ubiri 7 919 1,470 1,195 550 59.9 Bumbuli 7 2,390 3,129 2,760 739 30.9 Soni 5 3,946 5,748 4,847 1,803 45.7 Balangani 6 2,054 3,124 2,590 1,071 52.1 Manga 4 2,088 2,430 2,760 1,342 64.3 Mean for District 43 2,260 3,392 2,826 1,133 50.1