Tanzania - River Basin Management and Smallholder Irrigation Improvement Project

Document of The World Bank Report No. 15122-TA STAFF APPRAISAL REPORT TANZANIA RIVER BASIN MANAGEMENT AND SHALLHOLDER IRRIGATION INPROVEMENT PROJECT June 5, 1996 Agriculture and Environment Operations Eastern Africa Department Africa Region  CURRENCY EQUIVALENTS (as of May 1996) Currency Unit - Tanzania Shilling (Tsh) US$ 1.00 =Tsh 550 SDR 1.00 = US$ 1.45 WEIGHTS AND MEASURES Mctric System ACRONYMS AND ABBREVIATIONS ASMP Agriculture Sector Management Project CCM Chama Cha Mapinduzi (Political Party) DANIDA Danish International Development Agency FAO Food and Agriculture Organization GDP Gross Domestic Product IBRD International Bank for Reconstruction and Development ICB International Competitive Bidding IDA International Development Association IRR Intcrnal Rate of Return MOA Ministry of Agriculture and Cooperatives mOW Ministry of Water MTNR Ministrv of Tourism and Natural Resources MUV Manufacturing Unit Value Index MW Megawatts NAEP II National Agricultural Extension Project - Second Phase NAFCO National Food Corporation NALERP National Agriculture and Livestock Extension Rehabilitation Project NALRP National Agriculture and Livestock Research Project NCB National Competitive Bidding NORAD Norwegian Agency for International Development O&M Operation and Maintenance ODA British Overseas Development Administration OED Operations Evaluation Department RALDO Regional Agriculture and Livestock Development Officer RDD Regional Development Director SDR Special Drawing Rights SOE Statement of Expenditure SUDECO Sugar Dcvelopment Corporation UNDP United Nations Development Program URT United Republic of Tanzania GOVERNMENT FISCAL YEAR July I - June 30  CONTENTS I.BACKGROUND .......................................................................... I A. THE MACROECONOMY AND ADJUSTMENT .......................................................................... I The Adjustment Program ........................................................................... 1 Poverty Reduction During Structural Adjustment . ................................................................ 1 Macroeconomic Performance .......................................................................... 2 B. THE AGRICULTURAL SECTOR .......................................................................... 2 Constraints to Agricultural Growth .......................................................................... 3 The Nature of Future Agricultural Growth .......................................................................... 3 Natural Resource Management and the Environment ............................................................ 4 C. WATER RESOURCE MANAGEMENT AND IRRIGATION DEVELOPMENT ............. .............................. 5 Water Resource Management .......................................................................... 5 Irrigation Development .......................................................................... 9 D. GOVERNMENT ACTIVITIEs IN AGRICULTURE AND WATER MANAGEMENT .................................. 11 The Ministry of Agriculture and Cooperatives .................................................................... 1I1 The Ministry of Water . .......................................................................... 12 E. BANK INVOLVEMENT IN THE AGRICULTURE AND WATER SECTORS ............................................ 12 F. LESSONS FROM PREVIOUS IDA INVoLvEMENT ......................................................13.................. 13 H. THE PROJECT ...................................................... 13 A. RATIONALE ...................................................... 13 Rationale for the Project and IDA Involvement ...................................................... 13 B. OBJECTIVES AND SUMMARY ...................................................... 15 Project Objectives ...................................................... 15 Project Summary ...................................................... 15 C. DETAILED FEATRES .16 River Basin Management Component .16 Smaliholder Irrigation Improvement .20 D. PROJECT COSTS AND FINANCING .............................. 23 E. PROCUREMENT .............................. 25 F. DISBURSEMENT . ............................... 27 G. AccOUNTING AND AUDITS .............................. 28 This report is based on an appraisal mission in June and July 1995. The mission was led by J. Coates, (Principal Economist), and included M. Bekhechi (Environment Lawyer), J Bojb (Economist), S. Chamley (Anthropologist-Consultant), A. Elahi (Sr. rrigation Engineer), R. Hiiji (Basin Management Specialist), R. Rowe (Agronornist-Consultant), I. Payne (Environment Specialist-ODA), and D. Sungusia (Agriculture Services Specialist). The mission worked with a Govemnment team which included the Basin Management group led by W. Mutayoba (A.C. Water Resources) with S.A.S. Faraji (Hydrologist), I.E. Mwakalinga (Hydrologist), O.H. Rumambo (Hydrologist), and J.M. Kobalyenda (Principal Water Officer) and the Irrigation group led by Mr. E.H. Masija (A.C. Irrigation Dpt.), with Messrs. B.K. Nkuba, (Irrigation Engineer), I.A. Masenza (Irrigation Engineer), J.R.L. Kayumbe (Data Analyst), P. Riddell (Irrigation Engineer), and R.L. Masaki (Rangelands Specialist). The IDA project team also included E. Adu (Lawyer), P. Vandenheede (Disbursement Officer), Donna Criddle (Operations Analyst), and KeeSook Kim (Staff Assistant) and staff from FAO who participated in the Pre-Appraisal mission, J. Bisson (Economist), R. Dubois (Environment Specialist), and R. Sikkens (Irrigation Specialist - Consultant). The Peer Reviewer is G. Tschannerl (Sr. Municipal Engineer) and the Lead Advisor is U. Ktliffner (Sr. Irrigation Specialist). The Division Chief is S. Ganguly, and the Director is J. Adams.  m. PROJECT IMPLEMENTATION ................................................... 28 A. MANAGEMENT STRUCTURE ................................................... 28 Implementation Plans and Annual Workplans ................................................... 30 Monitoring and Evaluation ................................................... 30 River Basin Management Component ................................................... 31 Smallholder Irrigation Improvement ................................................... 31 IV. AGRICULTURAL DEVELOPMENT .............................................. 34 A. AGRICULTURAL DEVELOPMENT ................................... 34 The Smallholder Irrigated Rice Based Farming System ....................................................... 34 B. FINANCIAL VIABILITY OF THF FARMS ................................................. ..... 36 C. WATER CHARGES . ...................................................... 37 D. MARKETS AND PRICES ....................................................... 37 V. PROJECT BENEFITS AND RISKS ............................................................... 38 A. ECONOMIC VIABILITY OF IRRIGATION SCHEMES ................................................................ 38 B. THE RETURNS TO BASIN MANAGEMENT ............................................................... 39 The Rufiji Basin ............................................................... 39 The Pangani Basin ............................................................... 39 The Value of W ater in Competing Uses ............................................................... 40 Returns To Improvements in Irrigation Efficiency ............................................................ - 41 Rate of Return, Net Present Value and Sensitivity Analysis ....................... ................. ....... 41 C. FI SCAL ASSESSMENT ............................................................... 43 D. ENVIRONMENTAL EVALUATloN ................................................................................................. 44 E. WOMEN IN RURAL AREAS ................................... 45 F. RISKS ................................... 46 VI. AGREEMENTS AND RECOMMENDATION .......................................... 47 A. AGREEMENTS .......................................... 47 B. CONDMONS OF EFFECTIVENESS .......................................... 48 C. CONDmON OF DISBURSEMENT .......................................... 48 D. RECOMMENDATION .......................................... 48 ANNEXES A. STATEMENT OF WATER RESOURCES DEVELOPMENT POLICY B. STATEMENT OF IRRIGATION DEVELOPMENT POLICY C. IRRIGATION POTENTIAL BY RIVER OR LAKE BASIN D. ASSESSMENT OF COSTS AND BENEFITS E. SELECTED COST TABLES F. PASTORALISTS AND IRRIGATORS IN THE USANGU PLAINS G. ESTIMATED SCHEDULE OF DISBURSEMENTS H. TERMS OF REFERENCE 1. DOcuMENTS CONTAINED IN THE PROJECT FILE J. KEY PERFORMANCE INDICATORS K. MAPS  TANZANIA RIVER BASIN MANAGEMENT AND SMALLHOLDER IRRIGATION IMPROVEMENT PROJECT CREDIT AND PROJECT SUMMARY Borrower: United Republic of Tanzania Implementing Agency: Ministry of Water, and Ministry of Agriculture and Cooperatives Beneficiaries: Farmers on Selected Irrigation Schemes, Ministry of Water, Ministry of Agriculture and Cooperatives. Poverty: Not applicable. Amount: SDR 18.2 million (US$26.3 million equivalent) Terms: Standard IDA terms, 40 years maturity. Commitment Fee: Standard I Financing Plan: See para. 2.39 Economic Rate of Return: 34 percent Net Present Value: US$34.8 million at a discount rate of 12 percent. Staff Appraisal Report: Report No. 15122-TA Map: IBRD 27610, 27611, 27612, 27613 Project ID No. TZ-PA-38570 A variable rate between 0 and 0.5% of the undisbursed credit balance set annually by the Executive Director of IDA.   I. BACKGROUND 1.1 Tanzania's estimated GNP per capita of US$ 100 places it amongst the poorest countries of the 2 world . The social indicators present a dismal picture: gross primary school enrollment rates of 63 percent are lower than virtually all of sub-Saharan Africa and secondary school enrollment rates of 4 percent are the lowest in the world. An infant mortality rate of 115 per thousand exceeds sub Saharan Africa's average of 104 and adult mortality has more than tripled in the last decade. 1.2 Six years after Independence, in 1967, the Government announced the Arusha Declaration and called for establishment of a socialist society. The public sector was given a leading role and state control was extended throughout the economy. While the Government was able to forge a unified nation, and to make progress in improving social conditions, structural weaknesses within the economy began to emerge. By the mid-1980s economic decline and macroeconomic imbalances were severe with increasing budget deficits, high rates of monetary expansion and inflation, and a significantly overvalued shilling. Real per capita incomes were falling, exports collapsed, industrial output declined sharply, imports were reduced, and severe inefficiencies in production and distribution resulted in widespread shortages. Social progress achieved in the 1 970s began to be reversed and the lack of revenues severely hampered the maintenance of infrastructure. A. THE MACROECONOMY AND ADJUSTMENT The Adjustment Program 1.1 Beginning in 1986, the Government embarked on a program to reform and fundamentally change the existing approach to economic development by dismantling the system of pervasive economic controls and encouraging more active participation of the private sector in the economy. Structural reforms, particularly relating to traditional exports and the parastatal and financial sectors, were not fully completed, and macroeconomic stabilization remained elusive. Nevertheless, the economy responded well to the reforms that were implemented (notably, liberalization of food crop marketing and progressive improvements in foreign exchange management) and the accompanying increased availability of external resources. Official estimates indicate that both GDP and exports grew by more than 4 percent per year during 1986-94, with increased production of food and traditional export crops and strong rise in nontraditional exports. Poverty Reduction During Structural Adjustment 1.2 Recent household surveys have shown that the adjustment program has been successful in reducing the incidence of poverty. The devaluation of the shilling and removal of restrictions on the marketing of food crops boosted incomes of smallholder agricultural families. The increased availability of consumer goods as a result of liberalization directly benefited the rural poor (as well as the urban poor who also benefited from increased supplies of food from liberalized agricultural marketing), and liberalization opened new earnings opportunities for rural women and for off-farm employment. Sample surveys for the preparation of a Poverty Profile suggest that the percentage of poor declined from about 70 percent of the rural population in the early 1980s to about 50 percent in the early 1990s. The liberalization program 2 However, this official estimate of GNP may be seriously understated, as indicated by a recent survey of household expenditures.  2 increased the access of small-scale enterprises to production inputs which facilitated expansion of low- wage employment in the informal sector. Macroeconomic Performance 1.3 Progress has been made in reformiing the foreign exchange and trade systems over the last two years. Tanzania has moved to an interbank market and has abolished all export retention and import licensing. Excessive monetary expansion has been fueled by worsening fiscal management. The fiscal deficit (including grants) exceeded 8 percent of GDP in FY93 and 7 percent in FY94, after broadly balanced positions in the previous four years. This reflected widespread and increasing customs duty exemptions, an increasingly inefficient tax administration and the failure of the expenditure control system. Inflation has continued at over 25 percent and has recently accelerated. Real GDP growth averaged about 3-4 percent during FY92 to FY94. GroNvth in FY95 is likely to be seriously compromised by weak economic management and severe power shortages caused largely by less-than-average rainfall exacerbated by excess demand and delays in taking demand management measures by the authorities. These developments are limiting the scope for generating employment and, in particular, the high inflation rates resulting from macroeconomic mismanagement are eroding the real incomes of the poor. B. THE AGRICULTURAL SECTOR 1.4 Agriculture is the foundation of the Tanzanian economy, providing employment, food and exports. Some 84 percent of the employed population are in agriculture, producing 61 percent of both GDP and merchandise exports. Tanzania's 3.7 million farm families work small holdings with an average cultivated area of 0.9 hectares. Some 93 percent of all farmners cultivate less than 2.0 hectares each. 1.5 Food crop production dominates the agricultural economy totaling 55 percent of agricultural GDP, with livestock accounting for 30 percent, and the traditional cash crops (coffee, cotton, cashews, sugar, pyrethrum, tea, tobacco, sisal) for only 8 percent. Fishing and hunting contribute 6 percent and forestry I percent of agricultural GDP. 1.6 Average GDP growth for the whole economy since the mid-i 970s has been below the population growth rate. Between 1967 and 1978. the population grew at 3.3 percent per annum versus 3.6 percent per annum for GDP. While population growth slowed to 2.8 percent per annum between 1978 and 1988, GDP grew only at 1.7 percent per annum. As a result, GDP per capita fell during the early 1980s, and is only now beginning to rise to the levels enjoyed in the early 1970s. Presently real income per capita, evaluated using various deflators, is not much higher than it was in the mid-1950s. 1.7 Real growth in agricultural GDP3 has paralleled total GDP growth from 1966 through 1992. Agriculture averaged 2.8 and GDP averaged 2.7 percent per annum. The similarity is misleading; there have been sharp divergences between growth in agriculture and GDP. Growth in agriculture lagged significantly behind GDP during the initial period of increased Government intervention and control of the economy. Average annual growth rates differed by some 45 percent in 1966-75, and 65 percent in 1976- 80. In the early 1980s, the bottom of the economic decline, agriculture had actually started to recover, even while GDP growth was negative. Agriculture growth has led the economy through the periods of reform and economic recuperation since the early 1980s, responding to the earlier reforms in marketing of agricultural produce. 3 Unless mentioned specifically, all trend growth rates are average annual exponential rates, calculated using the least squares method.  3 Constraints to Agricultural Growth 1.8 Agricultural development in Tanzania faces tremendous challenges. To be successful farmers have to overcome the problems of distance from markets for products and inputs; diseases affecting both livestock and crops, and the normal variations in rainfall and climate. In a semi-subsistence agricultural economy such as Tanzania's, growth in food production will always continue at more or less the rate of growth in population. Provided new arable land is available and accessible, and there is political stability, the process of expansion whereby new rural families bring additional land into cultivation to provide themselves with their basic food needs, is continuous. For growth in agricultural production to surpass population growth, subsistence farmers need to be brought into a market economy where the availability of consumer goods provides incentives for cash generation and off-farm sales. The market also provides the ingredients for increasing the productivity of labor on the farm, as well as an outlet for marketable produce. Rates of agricultural growth will be greater than population growth to the extent that smallholder semi- subsistence farmners are brought successfully into the cash economy. 1.9 Factors which have constrained agricultural growth over the past 20 years include areas where Government policy or institutional inadequacy have hindered the successful, voluntary, integration of the smallholder farmer into the cash economy. These factors can be characterized in three ways: (i) the consequences of centralized economic management from 1967 to 1982; (ii) a decline in the efficiency of public services; and (iii) the inadequate development of supporting markets. In particular, improved literacy and expansion of rural infrastructure, particularly market access roads, are preconditions to a successful development of the rural market economy. While the Government had made great strides in reducing illiteracy in rural areas over the past twenty years, these improvements have been hard to maintain. Rural roads have also been allowed to deteriorate. The Nature of Future Agricultural Growth4 1.10 To maintain high rates of growth in agricultural output Tanzania will have to focus in the short run on export crops, where world markets can absorb the increased production without affecting prices. The crops where growth increased in 1986-91 are those sold in the more elastic export markets (coffee, cotton, cashew, tea). For these commodities, growth rates have measured 2.4 percent per annum, up from -7.9 percent in 1981-85. The production of raw material for domestic light industry (cotton, sisal, leather, fruits and vegetables) will also be important. Those crops and livestock products with high income elasticities of demand can expect to see their domestic markets expand. Food production can also be expected to grow, although perhaps not as rapidly as for export crops, in response to population growth, increases in per capita incomes and as improved technology becomes accessible to small farmers. 1.11 Thus, the sources of growth in Tanzania's agriculture sector will come from two sets of factors: In the short term, the move to a market oriented economy has removed the distortions in the exchange rate and other prices, the inefficiencies of monopolistic parastatals and cooperative unions, and the need to circumvent restrictive regulations. This has resulted in adjustment and growth in all sectors. The returns to adjustment in the cereals subsector have already been realized, and growth will taper off. However, there should be a surge in exports, up to the ceilings determined by current capacity levels in the processing industries, as the breakup of monopolies increases efficiency in cotton ginning, coffee rmilling, cashew and pyrethrum processing, and as export marketing opens up to the private sector, providing new opportunities in traditional and non-traditional crops, . 4 From the World Bank, "Agriculture Sector Memorandum", Report 12294, July 1994.  4 * In the medium term, continued growth in agricultural output will depend mainly on (i) the traditional expansion of food production for the domestic market as rural population growth causes an increase in the area under cultivation and livestock holdings, and (ii) increases in food production for the domestic market as technological change (and investments in on-farm improvements, including irrigation works) causes labor productivity to increase and real prices to fall; and (iii) increases in production of food or plantation crops for intemational export, as investment in the processing industries for crops such as coffee, cotton, cashews, and non- traditional high value items expands production capacity and reduces unit costs. The following factors will also contribute to agricultural growth: (iv) increases in production of food, especially animal products and food crops other than starches and cereals, as higher per capita incomes cause effective demand to rise; (v) increases in production of food for regional export markets, as new markets are found for maize, pulses, and oilseeds for example; and (vi) increases in the production of export crops as technological advances reduce production costs at the farm level, for coffee, cashews and cotton in particular. Natural Resource Management and the Environment 1.12 The management of the country's natural resources is weak. While very large, these resources are threatened by factors linked to unmanaged development. The main threat to the country's natural forests is uncontrolled clearing for charcoal, logging, tobacco curing, and cultivation of all crops. Select hardwood species are also being extracted without regard to fees or permits. Lake Victoria is threatened by pollution from the urban centers on its shores, from the runoff of fertilizers and agrochemicals, and from uncontrolled proliferation of water hyacinth. Coastal waters are threatened by the practice of dynamiting of coral reefs in search of fish. Wildlife management is threatened by the encroachment of reserved areas by cultivation. In certain regions erosion and shortened fallows are affecting the sustainability of cultivation. While the country's natural resources are not near depletion, they will be if the current trends are allowed to continue. The need to improve management of water is discussed in the next section of the report. 1.13 Effective steps are needed to strengthen the management of forests lands, the conservation of reserves, and the collection of fees and royalties for wood harvested. The ability of a weak and sparsely distributed forest service has recently been strengthened under a recent donor and IDA-funded Forest Resources Management Project (Cr. 2335-TA). 1.14 The lead role in defining national policy on environmental issues is now in the hands of the Vice President's Office, and the autonomous semi-Governmental National Environment Management Council. The recently completed National Environment Action Plan focused on the need for priority action in the key areas of land degradation, water supply, environmental pollution, marine and freshwater resource management, habitat conservation and the loss of biodiversity, and deforestation. The action program includes continuing the NEAP consultation process and completion and adoption of the draft National Environment Policy. In addition, a revision of the legislative framework is needed to more fully enable local participation in environmental management and in order to introduce the incentives and regulatory mechanisms to encourage sustainable natural resource use, including implementation of new land, forest and wildlife protection legislation; developing the means for assessing the environmental impact of future development initiatives; strengthening environmental education and public awareness programs; and the preparation of a priority long-term investment program. The Government intends to design incentives such as pricing policies, concession arrangements, revenue collection systems, and systems of resource tenure and property rights which are consistent with the objective of minimizing the further depletion of natural resources. The project will support these initiatives in the water sector.  5 C. WATER RESOURCE MANAGEMENT AND IRRIGATION DEVELOPMENT Water Resource Management 1.15 Surface Waters. Tanzania has defined nine river basins for water resources administration. The Rufiji is the largest and covers 177,000 kin2, 18 percent of Mainland Tanzania. It drains the Southem Highlands into the Indian Ocean. The smaller Ruvu/Wami basin flows into the Indian Ocean north of Dar es Salaam, and supplies water for domestic and industrial uses. The highly developed Pangani basin carries waters from Mt. Kilimanjaro and the Northern Highlands East into the Indian Ocean. The waters in the Northwest drain North into Lake Victoria, the Nile Basin. The Malagarasi river runs West into Lake Tanganyika. Extensive wetlands line rivers across the country and provide essential habitat for fisheries. The deltas of coastal rivers are especially irnportant as nursery areas for aquatic life. Mangrove trees line the brackish edge of these important wetlands. About one-third of Tanzania is arid or semniarid, with rainfall below 800 mm. Another third lies in highland areas with precipitation in excess of 1,000mm. The long dry season (June to October) results in low river flows, causing seasonal scarcity. Recently, scarcity and conflict amongst users has been exacerbated by a few years of less than average rainfall, coupled with the expansion of irrigation and the increase in demand for hydropower and water for urban uses. 1.16 Groundwater. Very old "Precambrian " rocks underlie more than 75 percent of the country. They generally yield water only in joints, or fractures and are therefore fairly low in yield, although suitable for village water supply. There are groundwater aquifers of quite variable yields -- from productive volcanic alluvium near Mt. Kilimanjaro and high-yielding coastal limestone deposits to very low yielding clay deposits of old lake beds. Less than one-third of the country can be characterized by groundwater in productive areas of sand, sandstone, and alluvial aquifers. Knowledge of groundwater resources is sketchy. Groundwater use is greatest in the Pangani, and some abstractions take place in the Ruvu/Wami and the Inland Drainage Basin (around Lake Natron). 1.17 Groundwater is a key source of water for both rural and urban areas, including the cities of Arusha, Dodoma, and Mtwara. Before 1978, emphasis was on construction of deep wells to tap protected, lower aquifers. By 1980, as much as 50 percent of these boreholes and water systems were inoperative because of maintenance problems with the diesel pumps. Since that time, shallow aquifers have been emphasized for water supply despite their vulnerability to pollution and uncertain yield. 1.18 Irrigation. Irrigated agriculture is the largest "consumptive" user of surface and groundwater. Demand is growing rapidly with the incentives provided under a liberalized agricultural market. Developments in irrigation are described in paras 1.36 through 1.39. 1.19 Hydropower Demand. Eighty percent of the electricity in Tanzania comes from hydropower. Practically all the hydropower generating capacity in the countly is in the Rufiji or the Pangani Basin, the two basins targeted for project interventions. The Mtera and Kidatu plants with a total of 284 MW installed capacity in the Rufiji River basin account for most of the hydropower installed capacity. The installed capacity on the Pangani Basin is now 91 MW5 following the commissioning of the Pangani Falls project in early 1995. The Kihansi project in the Rufiji Basin will provide an additional 180 MW by 1999. In both basins the power stations are downstream of most of the irrigation schemes and urban water abstraction points. This makes water use for hydropower highly subtractable from other uses. Conflicts have already developed between the needs for irrigation and water supply for storage, and hydropower flow requirements, including evaporation losses from the regulating reservoirs. 5 Pangani Falls with 66 MW, Hale with 17 MW and Nyumba ya Mungu with 8 MW.  6 1.20 Domestic Water Use. Some 68 percent of urban dwellers have a piped water supply, much of which is unreliable. In Dar es Salaam, domestic consumption accounts for approximately 70 percent of consumption from the water distribution system. Industry and urban agriculture account for the remaining 30 percent. Demand for water exceeds supply in most urban centers. Tariffs are too low. Technical, financial and managerial problems result in an under utilization of existing capacity. Of the total of 62 urban areas with piped water systems, over 90 percent reported serious deficiencies in the systems in 1990. About 46 percent of rural dwellers have an adequate water supply. Water pollution and uncontrolled abstractions interfere with water supply for humans and cattle, particularly in the dry season. 1.21 Water Quality. Serious water quality degradation was found in urban areas of Tanzania due to poorly treated industrial waste water discharges and sewage. An industrial pollution discharge inventory indicates that the priority areas of concentration include Dar es Salaam, Arusha-Moshi, Tanga, Lake Victoria, and the Morogoro area which serves as the head-waters for the Ruvu River and the water supply for Dar es Salaam. Adequate treatment of piped water by chemical additives, filtration and disinfecting is often hampered by limited funds. Sewer collection systems are inadequate and may leak into groundwater. Lack of proper sanitation and industrial pollution control have contributed to outbreaks of water bome diseases such as cholera, typhoid and dysentery. 1.22 Transboundary Water Pollution. Tanzania shares many waterbodies with neighboring countries. Of these, only Lake Victoria has been identified as having a serious pollution problem. Untreated discharges of sewage from the country's second largest city (Mwanza) reach the lake. In addition, large textile and tannery mills discharge toxic substances and oxygen demanding pollutants. Coffee processing, cotton ginneries, vegetable oil, cosmetics, soap and 8-10 fish processing plants discharge oxygen demanding wastes to the lake watershed. Other sources of contamination include mercury waste, coming from mining areas in the Lake Basin, and runoff from the overuse of agrochemicals and fertilizer in cotton production. A joint intemational effort to reduce pollution loading in the lake, and better manage fishing and other activities is being supported by IDA and the Global Environment Fund. 1.23 Legislation, Policy, Water Allocation and Regulation. The regulatory and institutional framework for water resources management is provided under the Water Utilization (Control and Regulation) Act of 1974, amended in 1981. The Act, as amended, declares that all water is vested in the United Republic of Tanzania, sets conditions on the use of water and authorizes the Principal Water Officer, with Statutory Authority, to be responsible for setting policy and allocation of water rights at a "national" level. For designated River Basins with their own water officer, these responsibilities are delegated to the Basin Water Officer. A Central Water Board advises the Principal Water Officer. A Basin Water Board advises the Basin Water Officer. Membership of both boards is at present limited to officials from sectoral ministries and regional government, and would be broadened under the project to include private water users and other stakeholders. 1.24 Despite these statutory powers, management of water resources by the Government has for a long time been fragmented. The effectiveness of water resource management, and the application of the legal framework, depend on the quality of a water rights administration. Under the legislation, and through the above offices, the Ministry of Water (MOW) grants rights for each withdrawal of water for use in domestic, industrial, hydropower, livestock, irrigation, environment and habitat preservation, or mining activities. A maximum of 50 percent of the expected low flow of a river at a point can be granted as a right to a person or corporation. This is done through a fairly lengthy procedure of application via the office of the Regional Water Engineer located in each of the 20 administrative regions of the country. The administrative system, centralizing information for the river basin, should: (i) be in a position to control withdrawals of surface and groundwater by issuing and revoking water rights; and (ii) know at all times the  7 quantity of water available in the basin, and its use, by monitoring both the sum of water rights granted, and physical availability. 1.25 Such administrative systems have recently been created in the Pangani and Rufiji Basins, and are still to be installed in the remaining basins. Even so, management of the waters in the Pangani and the Rufiji is still rudimentary. Water resource management plans for the basins are only now being put together, hampered by the lack of accurate and timely information. Data on precipitation, hydrometric data, and actual abstractions for irrigation is inaccurate and sketchy. Planning is more advanced in the Pangani Basin, where NORAD has provided assistance in the context of putting the new hydropower plant in place at Pangani Falls. Much further work is needed to strengthen information gathering and analysis, and to model water flow, in order to improve water management, and increase societal retums to the use of this scarce resource. The project would respond to these requirements. 1.26 Planning in the water sector has also been disjointed, and only recently have efforts been made to bring together a national picture of water availability and use. Virtually all work related to water resources management and development has been carried out at the regional level. This has included the formulation of Water Master Plans for 17 different regions all of which, with the exception of one, have been financed and prepared by donor agencies in different formats and containing different types of information. The main river basins were divided up into fragmented areas making water resource management difficult. The authority of regional water engineers is also fragmented, as competition between regions for water become an issue and essential data from regions do not get transferred to a central office. 1.27 National Water Policy. In an effort to improve the management of water resources, and with the aim of achieving the ambitious goal of providing water within easy reach (400 meters or less) to all by the year 2002, the Party adopted a National Water Policy for Tanzania in 1991. This policy, however, deals mostly with water supply issues, and a number of important points relating to water management, allocation and pollution control are not adequately addressed. A framework for preventing and resolving conflicts among competing users and for regulating demand for water is needed. The conflicts surrounding the inflow and use of water in the Mtera Reservoir crystallize the issue. This is not dealt with in the policy paper. In addition the Policy proposes no change in law or introduction of incentives to improve the effectiveness of water pollution controls. Nor does it fully address the fragmentation caused by the continued existence of 20 regional water organizations and the resultant problems in resolving cross-cutting regional water resources management issues. With these shortcomings, and an emphasis on drainage of wetlands so land can be used productively and on other water development and flood control structures, the 1991 National Water Policy may result in actions which further degrade environmental quality in Tanzania. The Bank and the Government would collaborate on the refinement of the National Water Policy under the project. 1.28 Pricing Policies. The legislation on water use does include a set of "economic water users fees," set forth in Subsidiary Legislation passed September 23, 1994. In addition the entities supplying water to the final consumer charge for this service, with dues payable in cash in urban areas, and often in terms of labor contribution to system maintenance in rural irrigation schemes. The price of water does not at present reflect its true scarcity value. Neither the basin level "economic users fees" nor the charges for delivery of water cover the true cost of the resource. This causes two types of problems. In both the water supply sector, as well as in irrigation, insufficient revenues are generated to cover operation and maintenance costs. The quality of the service, and of the water received, is undermined. A second problem is that the low tariffs encourage inefficient use of water and waste by industry, consumers and irrigators. As a result much of the urban population in Tanzania is being undersupplied. Wastage levels are very high in irrigation schemes, especially small scale traditional schemes, where irrigation efficiency is very low.  8 1.29 Plans are in effect to progressively increase water tariffs throughout Tanzania. With regard to the "economic water users fees" to be collected by basin water offices, it is proposed under this project that these rates be raised to a level which would provide sufficient funds to support the administration of basin water offices, including the collection of information on water quality and availability, the enforcement of pollution standards, and the administration and monitoring of water rights. Charges for delivery of water in irrigation schemes to be collected by water user associations, would be adjusted to cover operation and maintenance costs and any basin water charges. These principles would be used by smallholder irrigation schemes, for example, especially those which received assistance under this project, in determining the charges to be paid by irrigators for water delivered. 1.30 Relationship With Other Riparians. The project seeks to improve river basin management and improve smallholder irrigation efficiency in the Pangani and the Rufiji river basins. The Rufiji Basin is contained wholly within the territory of Tanzania. The Pangani basin (in Northern Tanzania) does extend marginally into Kenya. Lake Jipe stretches across the border between the two countries, just east of Mt. Kilimanjaro and receives some of its water from Kenya. This lake supplies water to a tributary to the Ruvu river, at the northern end of the Pangani Basin. The Ruvu itself, with its main catchment in Tanzania, supplies water to some 7,000 ha of traditional irrigation schemes, which have been in operation for decades. While the selection of the exact schemes to be rehabilitated over the six year implementation period have still to be defined, the project is likely to rehabilitate a few of the schemes along the Ruvu river. Project objectives are to rehabilitate and upgrade infrastructure on existing schemes. No additional water will be drawn from the river, and no additional area will be brought under irrigation. Indeed, an objective of the project is to improve irrigation efficiency, and reduce water abstractions from the river for a given area under irrigation. It may be the case that traditional schemes along the Ruvu would be rehabilitated under the project. 1.31 Under the provisions of OP 7 .50,6 the proposed project comes under the exception to notification requirements granted for "rehabilitation" of ongoing irrigation schemes. It should be noted that in this case Tanzania is the lowest downstream riparian, and the waters from Kenya flow into a system which is otherwise completely contained in Tanzania. On the issue of whether the viability of the Bank funded investments in improved irrigation infrastructure "will not be adversely affected by the other riparians' possible water use," it is the Bank's judgment that the possibility of Kenya materially affecting flows out of Lake Jipe into the Pangani basin and the Ruvu river is minimal. For these reasons it is the Bank's policy that Tanzania is not required to notify Kenya on its intention to undertake this type of project, although it may choose to do so in the spirit of intemational cooperation on riparian issues. 1.32 Water Use by Pastoralists in the Usangu Plain. Immigration of both pastoralists and cultivators into the Usangu Plain (in the upper catchment of the Ruaha river) has been sustained for many years. Gradually, the cultivable areas and the grazing lands have been occupied, and recently conflicts over water and land management have erupted between the pastoral groups and the cultivators. 1.33 Conflict between herders and farmers arises from two main sources: 1) livestock trespass on farms without permission and damaging crops and fields, and 2) farmers cultivate in areas that have been explicitly designated as grazing areas by village councils. Forage is scarce in the dry season. Crop stubble remaining on harvested fields is a desirable resource. Herders having large livestock herds - especially the Sukuma -- graze their animals on rice and maize stubble without permission from farmers, often destroying remaining crops, irrigation canals and bunds, and damaging fields in the process. This phenomenon has led to violent conflict between the Sukuma and farmers and remains unresolved. Herders are desperate for dry 6 World Bank Operational Policy Guidelines, No. 7.50, "Projects on International Waterways".  9 season resources. The more those resources dwindle, the greater the conflict. Conversely, farmers in sorne villages experience land shortages for irrigation during the wet season and establish fields on lands that have been specifically set aside for grazing by village councils. Herders then suffer wet season pasture shortages, and have taken farmers to court. 1.34 Formal village demarcation and titling activities, and land use planning to formally designate cultivation and grazing zones with land titles verifying the boundaries of these areas would help to reduce herder-farmer conflict, and make individuals more accountable. Pastoralists would benefit from secure land tenure rights, which would in turn facilitate pastoral resource management and rangeland conservation. These activities would be undertaken under the auspices of the World Bank funded Forest Resources Management Project (-r. 2335-TA), with support from this project. A special study of pastoral land tenure and resource use on the Usangu Plains would be financed under this project to assist in resolving these problems. The situation is summarized in Annex F. 1.35 Water Quality Management. The Water Utilization (Control and Regulation) Act of 1974 was amended in 1981 mainly to provide for the creation of basin management, to strengthen water pollution control and to establish standards for effluent discharges. In 1988, these changes were finally implemented with the creation of a Central Water Board with executive powers for specifying pollution control measures and enforcing them. However, government has not been able to enforce these new standards. Penalties have been very low, and there is little incentive for industries, and polluters to install control measures. Government efforts are also hampered by resource allocations inadequate to support aggressive policing and data collection efforts. Irrigation Development 1.36 There are three reasons why the development of irrigation systems is an important aspect of an agricultural development strategy. First, the variability inherent in Tanzania's rainfed production systems creates problems of shortages of the main foodcrops in years of inadequate or poorly timed rainfall. For this reason, policy makers concerned about food security have always sought to increase food production in irrigated areas, and to reduce this variation. Second, irrigation schemes if properly managed, provide sustainable increases in small farmer productivity and income, addressing rural poverty alleviation and environmental management objectives. Finally, irrigated agriculture is the only way in which high value crops (vegetables, flowers) can be produced under the controlled conditions needed to meet market schedules in Europe and other demanding international markets, thus supporting the drive for diversified high value agricultural exports. The statement on irrigation policy in the 1983 National Agricultural Policy document reaffirmed Government's commitment to developing smallholder irrigation schemes, while addressing problems which had affected such projects in the past. The 1983 policy document noted that irrigation projects had floundered due to: * lack of an irrigation development policy * reliance on sophisticated irrigation techniques, which demanded heavy investment, highly trained manpower, and a lot of foreign exchange; * lack of expertise in the design and construction of large irrigation schemes; * poor planning, especially of small-farmer schemes.  10 1.37 Irrigated agriculture is one of the two largest "consumptive" users of surface and groundwater.7 Demand is growing rapidly with the incentives provided under a liberalized agricultural market. It is estimated that some 80 percent of all irrigation water is used for the 150,000 ha thought to be under "traditional" irrigation by smallholder farmers, using "run of the river" techniques. Rice and vegetables are the main crops produced in such schemes. On-farm water management is through traditional furrows or small basins. These are poorly drained and prone to water logging and salinization. Smallholder schemes are unable to control their abstractions of water from rivers and streams. There is no charge for water use, leading to waste. Fluctuations in river flows make continuous access to irrigation water unpredictable, diminishing crop yields. Poorly designed canals lose water. The irrigation efficiency in such schemes is low, estimated at 15-20 percent. Access to water is uneven within schemes. Allocations to farmers at the tail-end of the canals are often uncertain. Water rights for irrigation are either recognized under traditional riparian arrangements, or are obtainable at a fee. While larger users possess a "water right" issued by the Ministry of Water (MOW), the fines for excess water use are minimal and rarely applied. Some 25,000 ha are irrigated under large, modem, expensive, centrally managed schemes, with precision-engineered infrastructure. Most of these are State Farms run by National Food Corporation (NAFCO) or Sugar Development Corporation (SUDECO), producing rice and sugar. In addition, there is a Government- controlled smallholder rice scheme on the upper Pangani, and a few private farms in the Moshi-Arusha area producing irrigated flowers and vegetables for the European market. The area which is potentially irrigable in Tanzania is large, possibly as much as 850,000 hectares. The results of an FAO study in the late 1980s which assessed irrigable area (taking both soil and water constraints into consideration) are noted in Annex C.8 1.38 The Ministry of Agriculture and Cooperatives (MOA), with assistance from FAO and UNDP has been reviewing the Government's experience with irrigation projects, and have concluded that: - The emphasis should be on rehabilitation and improvement of existing smallholder schemes, and "future development should be based on staged improvement and expansion of existing local technology, which allows the farmers to adapt at their own pace. Equal emphasis should be given to operational and extension support to farmers at existing schemes." * In some cases, projects undertaken to date have been too sophisticated, requiring expensive structures and massive capital injections. The funds for completing these projects have often not materialized, causing a massive waste of resources invested in unused, half completed schemes; * If a project is expensive to begin with (certain recent donor funded projects in Tanzania have cost $15,000 to $20,000 per ha to cover the costs of precision land levelling and sophisticated water distribution systems), incremental yields and prices will have to be very good, for the economic rate of return to be satisfactory. The possibilities for marketing such high value crops, or for getting cropping intensities above 100 percent, is limited to small, well-managed private farms at present; * For smallholder irrigation schemes to succeed, management of water and land use should be in the hands of water user associations representing the interests of participant small farmners. 7 The other large consumptive use is due to the increase in evaporation from the reservoirs behind the Mtera and Nymba ya Mungu dams, created to support electric power generation. U IRT/FAO, "Field Document 14, Smallholder Irrigation Development Priorities", September 1990.  I1 1.39 Following this review, MOA has published the National Irrigation Development Plan (October 1994). This plan reiterates the priorities: * rehabilitation or upgrading of existing traditional irrigation schemes in all regions where this type of irrigation is significant;