Tanzania - Urban Water Supply Project

Document of FILE Copy The World Bank FOR OFFICIAL USE ONLY Report No. P-1928a-TA REPORT AND RECOMMENDATION OF THE PRESIDENT OF THE INTERNATIONAL BANK FOR RECONSTRUCTION AND DEVELOPMENT TO THE EXECUTIVE DIRECTORS ON A PROPOSED LOAN TO THE UNITED REPUBLIC OF TANZANIA FOR AN URBAN WATER SUPPLY PROJECT December 7, 1976 This document bas a restricted distribtion and may be usd by recipients only in the performance of their official duties. Its contents may not oterwise be islosed without World Bank authorization.  CURRENCY EQUIVALENTS Tanzania Sh = US$0.12 US$1.00 = TSh8.30 (As the Tanzania Shilling is officially vaLued at a fixed rate of 9.66 TSh to the SDR, the US Dollar/Tanzania Shilling exchange rate is subject to change. Con- versions in this report were made at US$1.00 to TSh8.30 which is cLose to the short-term average exchange rate.) Government - Government of Tanzania MWEM - Ministry of Water, Energy and Minerals WRI - Water Resources Institute  FOR OFFICIAL USE ONLY INTERNATIONAL BANK FOR RECONSTRUCTION AND DEVELOPMENT REPORT AND RECOMMENDATION OF THE PRESIDENT TO THE EXECUTIVE DIRECTORS ON A PROPOSED LOAN TO THE UNITED REPUBLIC OF TANZANIA FOR AN URBAN WATER SUPPLY PROJECT 1. I submit the folLowing report and recommendation on a proposed loan to the United Republic of Tanzania for the equivalent of US$15.0 million to help finance the construction of an urban water supply project in Morogoro, Tanzania. The proposed loan will be for 20 years including 4.5 years of grace with interest at 8.70 percent. PART I - THE ECONOMY Background 2. Although it has been some time since the last full economic and updating reports on Tanzania were finalized in May and December 1972, respectively, considerable staff economic work has focused on the country since that time. An agricultural sector report was issued in December 1974. The report of an industry and mining mission was dis- tributed in April 1975 and a study of the fiscal aspects of Tanzania's recent decentralization was completed in ApriL 1975. In March 1976, at the request of the Government, the Bank mounted a special mission which analyzed the fiscal implications of the Government's programs for Universal Primary Education and Universal Rural Water Supply. The work of this mission has provided substantiaL input to the work of the basic economic mission which visited Tanzania in July/August 1976. Throughout the last two years or so Bank staff have engaged in a continuous economic diaLogue with the Government on the serious balance of payments difficuLties which have been encoun- tered by the country and the Govern,ment's policies and programs designed to close the external gap. These policies and programs were reviewed in the Program Loan President's Report No. P-1517a-TA, dated November 25, 1974 and in a memorandum (M75-687) from the Secret- ary dated September 25, 1975. The Consultative Group for East Africa 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.  -2- met in April 1975 to discuss the progress and prospects of the Tanzania economy and the need for 3dditionaL resources to support the Government's development progr3m. 3. The TANU Party, under the leadership of President Nyerere, has been the guiding force in Tanzania's political evolution since the 1950's. Over the years following independence the political Leadership has deveLoped a phiLosoahy of egaLitarian sociaLism which has been articulated in many documents, most central of which is the Arusha Declaration of 1967. In restructuring the politicaL, economic and sociaL Life of the country the Leadership has introduced an impressive series of far-reaching institutionaL reforms. For the past decade Tanzania's sociaL and economic poLicy has been guided by three fundamental objectives: (a) the achievement of a partici- patory, decentralized socialist economic order; (b) the eradication of absoLute poverty and progress toward greater income equality; and (c) more rapid long-term economic growth with full participation of all regions and population groups in the development process. Reflecting this philosophy some of the Government's most significant decisions have been in the area of incomes poLicy. The tax structure is highly progressive and middle and upper incomes have been restrained while lower income wages have been increased. In the provision of social infrastructure services, poorer areas have been favored. As a result of these and related policies, differentials in living standards between upper and Lower income urban dweLLers have been narrowed and some impact has been made on urban/rural differentials and regionaL disparities of income. /e, Thp nverridino aoal is to develop a socialist economic system. An important consequence of this has been a series of nationalization measures and the accompanying proliferation of government parastatal enterprises. The state has increasingly intervened in the price mechanism; prices of domestic factors and products and imports are con- trolLed directLy or by such indirect methods as import licensing. These structural changes have resulted in implementatior and operation prob- lems. In addition, scarce manageriaL manpower has been spread thinLy, exacerbating aLready serious shortages of top and middle-level staff. Some of these confLicts shouLd be seen, however, as the inevitabLe consequences of a "frontaL" attack on poverty and underdeveLopment. 5. With a per capita income of $170 in 1975, Tanzania is cLas- sified as one of the 25 least developed countries as defined by the United Nations (couintry data are provided in Annex I). Between 1968 and 1973, Tanzania's GDP grew at 4.6 percent per annum. Domestic savings averaged 17 percent of GDP, while gross investment averaged 21 percent -- extremely high rates for a country at Tanzania's low level of per capita income. However, the growth rate of GDP was not commensurate with the  magnitude of the investment effort. The major reasons for this include the high proportion of investment which was directed into social service projects and infrastructure (such as the transportation links with Zambia),the sluggish growth in the agricultural sector and the stagnant or declining productivity in parastatal enterprises. During this period the balance of payments position of Tanzania was generally satisfactory despite the disappointingly slow growth of exports (overall export volume grew at only three percent per year from 1968 to 1972 and in 1973 export volume fell back to the 1968 level). With the sharp increase in capital inflows, the overall balance of payments was in surplus from 1971 to 1973, and there was a modest build-up in foreign exchange reserves. As a result of prudent financial management, net reserves were increased to approximately $145 million by the end of 1973, the equivalent of four months imports. The Economic Crisis of 1973-74 6. Events then occurred which resulted in a drastic change in the overall balance of payments of Tanzania. In 1973 and 1974 rains failed in many parts of the country necessitating substantial increases in imports of basic food items beginning in mid-1974. The drought also affected cash crops, leading to declines in the export volumes of coffee, cotton and sisal. The outcome of these forces was the rapid depletion of net reserves to $50 million at the end of 1974, equivalent to only three weeks of imports. Industrial production also leveled off, due in part to shortages of imported raw materials and spare parts stemming from the growing foreign exchange constraint. As both agriculturaL and industrial production stagnated, domestic demand increased rapidly due to expansionary fiscal, monetary and wage policies, causing severe pressure on the domestic price level. 7. During 1975, food product;ion recovered from the previous year's 30 percent shortfall, but grain imports were still required and the balance of payments situation continued to deteriorate due to falling export prices and production difficulties in the tradi- tional export crops. The resulting trade deficit rose from the 1974 level of $300 million to $325 million in 1975. Tanzania was able to cover the 1975 resources gap by curtailing imports to the bare minimum and by securing substantial program type assistance in addition to project related foreign aid, including a $30 million Program Loan (No. 1063-TA) from the Bank and funds from the IMF special oil facility. The outlook is for an improvement in 1976 due to good harvests of traditional export crops and it is expected  -4- that the trade deficit wilL drop to approximateLy $270 million. There has also been a gradual slackening in inflationary pressure. As measured by the wage earner's cost of living index, inflation was cut from 61 percent in 1974 to 32 percent in 1975. The rate of increase is expected to moderate further in 1976. 8. However, Tanzania' s present economic problems have not been exclusively caused by drought-related crop failures and import price escalation. The most disturbing aspect of Tanzania's recent economic performance has been the declining growth rate of output from the agricultural, industrial and mining sectors. The growth rate of agricultural output had been running slightly behind the rate of growth of population for six years prior to the harvest failure of 1974. Inadequate producer price incentives, inefficiencies in the transport and distribution systems for inputs and marketed production, and the effects of inadequately planned villagization contributed to this clisappointing performance. In the industrial sector many enterpr;ses have experienced declining labor productivity which has been cttributed to lack of incentives, poor discipline and ineffective management. Furthermore, while the ratio of domestic savings to GDP rose during the 1960'{,, jt began to decline in 1973 and feLL drastically to 8 percent- in 1974 as a direct outcome of the balance of payments crisis. ever, a slow recovery in the savings ratio to abou' 9 percent. 1ITw- occured in 1975 and further improvement is expected in 1976. The Government Response In late 197A, the Tanzanian r,overnment formulated a comore- hensive program of poLicy actions to bring the balance of payments problem under control and to restore the growth rate of output. The principal elements of the program included reallocation of invest- ment in favor of directly Droductive sectors, measures to raise agri- cultural output, steps to improve incentives in industry and mining and to increase the export orientation of these sectors, and constraints on public and private consumption demand. Imports were cut through direct controls and measures were introduced to slow down the rate of growth of private consumption. Indirect taxes were increased in 1974 and further increased in the 1975/76 budget. User charges for water and electricity were also raised and an extremely res- trictive wage and salary policy has been followed. In addition, cuts of about 7 percent in the civil service were made in March 1976. The Government has also raised agricultural producer prices to levels approaching world parity as an incentive to increase output. The overall planning capability of the Ministry of Agriculture was strengthened and a project coordination unit to improve project 1/ Current prices;not exactly comparable to data on paae 3 of Annex 1, which are in constant 1973 prices.  -5- implementation has been established in the Ministry. Furthermore, significant progress has been made in reallocating public invest- ment in favor of the directly productive sectors of agriculture, industry and mining. Finally in October 1975 Tanzania, in con- junction with Kenya and Uganda, devalued the shilling by approximately 14 percent as an incentive for increasing exports. 10. The Government's progress in implementing policies and programs designed to close the balance of payments gap in accordance with understandings relating to the program loan was the subject of a memorandum (M75-687) from the Secretary to the Executive Directors 'paragraph 2 above). The most encouraging aspect of the Government's response to the balance of payments crisis was the demonstration that Tanzania retains the ability to push through necessary but unpopular measures over a wide front. Tanzania's prospects for bringing the balance of payments deficit under control in the medium term clearly depend upon the effectiveness of the program of poLicy actions out- lined above. While it is too early to measure the full effect of all the policy measures introduced in 1974 and 1975, they have already had an impact on imports and domestic consumption, and the recovery of food production has clearly helped to alleviate the food deficit. In only one rmajcr problem area has the Government been unabLe to make major inroads: the rate of growth of public consumption expenditures. While the Government proposed no increase in nominal recurrent expend- iture in 1975/76 (implying a cut in real expenditure), actual recurrent expenditure appears to have increased by a substantial amount. Both budget control problems and actual spending overcommitment were con- tributing factors. This difficulty is receiving increased attention by the Government in cooperation with both the IMF and the Bank. 11. Since this basic program o' economic restructurinr will renuire more time before its full effects can be realized, the balance of pay- ments can be expected to remain weak in the short run. Still, Tanzania should have a stronger economy at the end of the restructuring process. The central focus of the assistance efforts should be, therefore, to ensure that the balance of payments gaps arising over the next few years do not prevent Tanzania from implementing a development prcgram designed to incorporate the required poLicy changes. In this connection, nego- tiations for a proposed second program credit were recently held in Dar es Salaam but have not been completed pending the successful outcome of discussions between the IMF and Tanzania regarding the terms of a proposed standby agreement, and the subsequent preparation of a satisfactory state- ment by the Tanzanian Government to the Bank regarding its economic program and policies. A Fund mission is presently in the field negotiating a standby agreement. In addition, and despite Tanzania's impressive domestic savings performance, a continued capital inflow in excess of the foreign exchange component of high priority projects will also be required. Financing of some local expenditures will, therefore, remain justified.  -6- 12. In terms of debt outstanding and disbursed, the Bank Group is Tanzania's second largest creditor after the PeopLe's Republic of China. Other major lenders are Sweden, Canada, Denmark, the NetherLands and the FederaL RepubLic of Germany. The current Low overaLL debt service ratio of about 8 percent (incLuding a notionaL 40 percent share of the debt of the East African Community Corporations) is expected to rise to about 12 percent by 1985. IncLuding a notionaL share of EAC debt, the IBRD is presentLy hoLding 15 percent of Tan- zania's outstanding externaL debt aid IDA 10.8 percent; the IBRD share is expected to faLL to about 10 percent by 1985 and the IDA share to rise to about 13.1 percent. Debt service payments to the Bank are about 30.9 percent of totaL debt service payments; the corresponding share for IDA is 2.2 percent. These two figures are projected to become 30.6 percent and 3.5 percent, respectiveLy, by 1985. The Bank's exposure is high because severaL major donors are now making their aid avaiLabLe either on grant basis or ',ery concessionaL terms and because, as a resuLt of prudent debt management, suppLiers' credits have been kept to a minimum. The average interest rate on Loans to Tanzania outstanding at December 31, 1975 amounted to onLy 3.5 percent and the average final maturity was 42 years. East African Community 13. The East African Community Corporations have experienced difficulties in recent years. While the 1967 Treaty for East African Cooperation is one of the most far-reaching and comprehensive economic cooperation agreements in existence among sovereign states in the developing world, in practice, the degree of economic integ- ration and cooperation among the Partner States is much less than what was envisaged in the Treaty. Polit-ical developments in the Partner States have created tension within the Community and impaired the growth of interstate trade. These difficulties have been compounded by the balance of payments constraints which alL three Partner States experienced. 14. During 1975, the Partner States recognized that because of their fundamental political and economic differences, a review of the 1967 Treaty was called for. For this purpose a Commission was established consisting of three Cabinet level officials from each country, with Mr. Demas, President of the Caribbean Development Bank, as Chairman. This Commission is expected to submit its report by the end of 1976. While there is hope that the Commission may recommend measures to deaL with the structuraL differences in the Long run, the Corporations are stiLL facing a difficuLt period.  -7- This was refLected in the significant deterioration in the relation- ships between Partner States in earLy 1976. Arrangements for debt service payments were again disrupted resulting in delays in making Bank service payments. As a result, disbursements to the East African Railways Corporation, the East African Harbours Corporation, the East African Posts and Telecommunications Corporation and the East African Development Bank were suspended by the Bank on ApriL 28, 1976 and, as the Executive Directors were informed on May 25, 1976 (SecM76-383), a mission visited East Africa to discuss with the Presidents and Finance Ministers the debt service probLems which were endangering both the Bank's operations in East Africa and the credit standing of the Com- munity and the Partner States (being jointly and severally responsible). Agreement was reached on May 28, 1976 to ensure remittances from the Regional Offices of the Corporations to complete all over-due payments and, as these were made by June 1, 1976, the suspension was lifted. Since then debt service payments have been made on time. Regarding future debt service payments, the agreement (i) specifies fixed contri- butions by each Partner State to meet external debt obligations, (ii) commits the Governments to ensure remittances from the Regional Offices of the Corporations, (iii) establishes a mechanism and procedure for timely payments, and (iv) provide:; automatic foreign exchange cover to debt service remittances. PART II - BAL4K-GRQUP-QPERAIIQNS-I_-ITNZ6NlA 15. Tanzania joined the Bank, IDA and IFC in 1962. Beginning with an IDA credit for education in 1962, 25 IDA credits and 11 Bank loans amounting to $403.8 million have so far been approved for Tanzania. In addition, Tanzania has been a beneficiary of 10 loans totalling $244.8 million which have been extended for the development of the common services and development bank operated regionally by Tanzania, Kenya and Uganda through their association in the East African Community. The only IFC investments in Tanzania to dat, totalling $4.7 million, were made to the Kilombero Sugar Company in 1960 and 1964. This Company encountered financiaL difficulties and in 1969 IFC and other investors sold their interest in the Company to the Government. Annex II contains summary statements of Bank loans and IDA credits to Tanzania and the East African Community organizations as of October 31, 1976 and notes on the execution of ongoing projects. 16. In keeping with Tanzania's overall development strategy Bank Lending operations are increasingLy focusing on the rural sector and  -8- directly productive projects. Up to the end of FY72, 10 out of 15 Loans and credits made individually to Tanzania had been for infrastructure. Of the 21 Tanzania operations approved since then all but six were for directly productive projects. The directly productive projects are supporting both the agricultural and indus- trial sectors, including the Kigoma Integrated Rural Development Project (Credit No. 508-TA), the NationaL Maize Program (Credit No. 606-TA), the Fisheries Development Project (Credit No. 658-TA), the Tanzania Investment Bank Project (Loan No. 1172-TA), and the Mwanza Textile Project (Loan No. 1128-TA). The first Bank Group assisted project in the forestry sector, the Sao HiLCl Forestry Project (Loan No. 1307-TA), was recently approved and will provide the raw material for Tanzania's first paper and pulp plant, also under preparation for consideration by the Bark Crcup for financing. We plan to continue assistance for productive projects in the indus- trial and agricultural sectors: a project to support selected industries at Morogoro is expected to be ready for consideration by the Executive Directors in the near future, a proposed second rural deveLopment proiect in Tabora Region has been aopraised and a third regional rural development project is under preparation for Mwanza and Shinyanga Regions. In addition, the agricultural stra- tegy recommended in the sector report (paragraph 2) is providina the basis for the first project in the National Agricultural Develop- ment Program (NADPi presently under preparation. Other projects under preparation included a second urban prcject and a trucking project. CART III - THE WATER SUPPLY SECTOR IN TANZANIA UarkgrQund 17. Tanzania is gnerally an arid country, poor in surface and ground water. The rivers belong to three drainage basins: the eastern basin drains into the Indian Ocean, the western basin drains into Lake Tanganyika and the Congo River, and the northwest basin drains into Lake Victoria and the Nile. The driest regions are in the center of the country, where there are no rivers and ground water is scarce. The presence of turbidity in surface water, salinity in many ground water resources and high fLuoride content in some aquifers also creates problems in water quality.  -9- 18. Until recently, hydrological and hydrogeological informa- tion concerning available water resources in Tanzania was almost nonexistent. However, since the Government decided in 1969 to prepare an inventory of the water resources in each region, consider- able work in this area has been initiated. At present, the work has been completed in two regions and is underway in the remaining important regions. Sectoral Organization and Development 19. The organization of the water supply sector has gone through many changes over the past decade. Until several years ago, the activities in the sector were administratively divided between the Ministry of Works (responsible for urban water supplies), the Ministry of Local Government (for sewerage) and the Ministry of Agriculture (for water resources and rural water supplies). In 1969, the Ministry of Water Development and Power was established to consolidate within one ministry all activities in the water and power sectors. In 1975 the Ministry's responsibilities were expanded to include mineral resources and its name changed to Ministry of Water, Energy and Minerals CMWEM). 20. Under the decentralization policy adopted by the Government in 1972, the role of what was then the Ministry of Water DeveLopment and Power was changed from one of managing and controlling the entire water program to one of providing policy direction, technical assist- ance to the regions, procurement, manpower training and disposition, and initiating and executing "national water projects" (including those in major urban areas). Since that time each region's water department has been headed by a Regional Water Engineer, who is responsible for operation and maintenance and for the design and implementation of extensions and improvements to the water supply systems within his region. The Regional Water Engineer is responsible to the Regional Development Director for administrative and financial matters and to the central ministry for technical matters. 21. Urban Water Supply: About 85 percent of Tanzania's urban population has access to water, but the level of service varies widely in both quality and quantity. About 20 percent of the consumers are served through private connections and account for approximately 50 percent of total consumption. The remaining domestic consumers are served through public standpipes. Water is usually treated and chlorinated. At present water supplied through private connections is metered and charged at a uniform rate of TSh2.20/m3 Water supplied through public standpipes is free of charge. Industry's share averages approximately 10 to 15 percent of the total urban water consumption.  -10- 22. During recent years, the expansion of urban water supplies has fallen noticeably behind the growth of domestic and industrial needs due to the shortage of budgetary funds and the emphasis on ex- panding rural water supplies. Many urban systems are operating over or at their maximum capacities with no production reserves. Water shortages are common during periods of peak consumption and water quality has deteriorated. Extensions of distribution systems to new housing estates have been halted for lack of water and the non- availability of a reliable water supply has become a bottleneck to the implementation and/or operation of a number of industrial and agro-industriaL projects. 23. Rural Water Supply: Capital expenditures for the rural water supply sector have been mainly funded by Sweden since the Swedish aid program commenced in 1962. These funds have reached TSh55 - 65 million (US$6.6 - 7.8 million) yearly to which the Tanzanian Government used to add 20 - 30 percent. However, during the past two years the Swedish contribution has been the only development allocation. As of June 1976 it was estimated that approximately 22 percent of the total rural population of 13.7 million were served with water systems developed by MWEM. It is also estimated that an additional 10 percent of the rural population is served through non-MWEM systems or acceptabLe traditional sources (stream, lake, etc.). The level of service provided by MWEM systems ranges from a single protected spring or well with a hand pump to a distribution system serving consumers through public standpipes. Water is free of charge; house connections and water billing are not used. 24. Selection of villages within the rural water supply cons- truction program is fully decentralized to the regional administration. The program is implemented by the Regional Water Engineer with funds supplied through the Central Government. Water systems are run and maintained by the regional organization but are funded centrally. Budgetary allocations for recurrent cost requirements of the existing systems are inadequate -- in some cases lower than required to supply fuel needed to keep them running, let alone maintain them properly. Other important issues in the rural water sector include identifying how the limited available funds could be most effectively utilized (in particular how village priority could be related to least cost technology) and the role a system of charging for rural water could play in mobilizing additional resources for water deveLopment and in ensuring the efficient use of available supplies. The Bank has been involved in these policy issues through its discussions with the Government on the implications of the Universal Rural Water Supply Program (para. 2). Both the needs and problems of the rural water  -11- systems are being focused on in the Bank regional project under imple- mentation in Kigoma and those being prepared in Mwanza/Shinyanga, Tabora and Mara. QrgaonitiQnal_EzegbJlma 25. The standards of administration and technical and financial control in this sector are weak. These difficulties reflect the shortage of trained personnel at all levels and problems in organi- zational structure. The training and technical assistance components of the proposed Project were specifically designed to address the manpower problems (paragraphs 35 and 36). Extensive Government/Bank discussions have focused on the organizational issue and these resulted in the Government hiring a consultant to review existing institutional arrangements. While the recommendations which resulted from this review were not fully accepted by MWEM, the Ministry followed up the study with its own proposals, the most important of which relate to financial operations. 26. At present all capital and recurrent expenditures in the water sector are financed through the Government's budget and all revenues are paid directly into the national treasury. With the recent national resource constraint this has resulted in severe re- ductions in oDerations and maintenance funds and subsequent deterioration of individual systems. In addition, it has prevented the collection of any surplus in the sector and made it impossible for MWEM to enforce discipline in the cotlection of water bills (as it is not directly res- ponsible). In the MWEM's reform proposal, which was recently approved by the Cabinet, Dar es Salaam would be serviced by a separate corpo- ration and an urban water supply fund would be established for Morogoro and other towns. This fund would receive water revenues from these cities and disburse against operating costs, debt service and expend- itures on routine system expansions. As a long term objective the fund should be self-sufficient but in the interim Government subsidies would meet any shortfalL. This fund wouLd be administered by MWEM and dis- bursements would be made in accordance with budgets for each urban water system prepared by the local water officials and approved by MWEM. The financial advisor financed under the Project (paragraph 35) would be attached to the fund and would play an important part in setting up its procedures. Commercial accounting has been accepted in principle by the Government and it would be implemented as soon as feasible in the sector. Although these changes do not go as far toward establishing an independent and commercially oriented sector as might be desired, they specifically address a number of the most pressing problems with the present arrange- ments and represent the Government's initial positive reaction to issues raised by Bank work in the sector.  -12- Sewerage and Drainage 27. There are only six towns with partial water-born sewerage services: Arusha, Dar es SaLaam, Moshi, Mwanza, Tabora and Tanga. The population served totals about 200,000. In addition, various governmental institutions near towns are served by separate sewerage systems which usually have oxidation ponds for treatment. Expansion of water-born sewerage systems has low priority; public sewerage is not essential where density is low and disposal of wastes by other means is possible. The sewerage systems are always separated from storm water drainage, which is generally carried out in conjunction with road construction. Both sewerage and storm water drainage systems are operated and maintained by local authorities, but to very low standards. Long Term Sectoral Development Programs 28. The long term program for urban water supply is considered separately for each town, stressing the need for securing adequate services to the growing urban population and industry by timely extensions of existing water systems. However, due to the country's economic problems in the last few years, the expansion program has been implemented in only the three towns where external finances were available (Dar es SaLaam, Tabora, and Tanga). Implementation of the remaining part of the program depends on availability of further external funds and Government decisions on the priority of the systems involved. The most acute current shortages which could be resolved with additional outside financial assistance are in the seven urban water systems which were reviewed in the Bank Group financed feasibility study (Shinyanga, Lindi, Morogoro, Mbeya, Mtwara, Mwanza and Iringa). The African Development Bank has agreed to assist in financing the Shinyanga and Lindi water systems and the European Development Fund is considering supporting the expansion of the water systems in Mbeya, Mtwara and Mwanza. Morogoro water supply would be financed through this Project but up to now no funds have been found to finance the Iringa water supply system. 29. The Long term program of rural water supply identified by the Government in 1971 proposed that the entire rural population be provided with reasonably accessible water by 1991. With the subsequent speed-up of the villagization effort and the resulting increase in demand for various social services (including water), TANU and the Government decided in 1974 that by 1980 every village should have at least a source of water supply. More recently, however, the Government requested Bank assistance in reviewing comprehensively the  -13- financiaL implications of the acceleration of this target (para 2). The resulting joint Government/Bank study projected that in order to achieve the official target of Universal Rural Water Supply by 1980, approximately 20 percent of total Government development funds wouLd be required for the rural sector program alone. As Government budgetary decisions have allocated a far lower proportion of develop- ment funds than this to the rural water sector, it is apparent that the target itself will not be achieved. However, by accelerating the target, the Government did highlight the importance it gives to providing social services to the rural areas and put its achievement within a time frame which is not so distant that it could be ignored. PART IV - THE PROJECT 30. With the growth of industry and urban pcpulation and the resulting rapid increase in deficits of individual urban water supply systems, the Government requested Bank assistance in identi- fying the highest priorities in thE urban water sector and develop- ing appropriate investment proposals. As a result, a Bank Group financed feasibility study of seven urban areas facing acute water shortages was initiated in 1975 (financed under Credit No. 495-TA, the National Sites and Services Project). With the completion of that study, a Bank appraisal team in December 1975 reviewed and confirmed the technical and economic viability of the urban water supply systems proposed for these seven towns. However, because the total Project as appraised was larger than available Bank financing, the Government requested that the Bank support the development of the Morogoro sys- tem (the largest investment and an essential prerequisite for the proposed Bank investment in the Morogoro Industrial Estate) and both the Government and the Bank actively sought additional financing for the remaining cities. 31. In November 1976 negotiations for this Loan took place in Washington and Mr. F. K. LwegaruliLa, Principal Secretary of MWEM led the Tanzanian delegation. An appraisal report entitled "Urban Water Supply Project" (Report No. 1199-TA) is being distributed separately to the Executive Directors. A Loan and Project Summary is provided as Annex III to this report.  -14- Obiectives 32. This Project would make a substantial contribution to the Government's investment program in the urban water sector by financing the most costly system of the seven towns selected as having the most urgent need for water in the country. Morogoro is a rapidly expanding city located at an important transportation hub of Tanzania and pre- sently faces a growing water shortage. Furthermore, the implementation of Tanzania's largest industrial estate at Morogoro will depend directly on the availability of water supplied by the proposed Project. The physical need and priority of the Project are therefore clear. 33. In addition, the Project has provided a focus for studying and attempting to deal with major institutional and training problems in the water supply sector. One of the major institutional problems, that of financial independence and discipline, is being specifically addressed by the urban water supply fund concept which developed from discussions between the Bank and Government on this Project (para 26). The Project will also finance a number of advisors who will assist MWEM with its management and technical responsibilities and in addressing its training needs (para 35). Finally, the Project will provide funds to assist the Government in expanding its facilities for the training of manpower in high priority skill categories (para 36). Description 34. Under this Project, the Ngerengere River wouLd be impounded by the constjuction of the Mindu Dam. This earth dam would retain 10 million m of water at full capacity and would safely yield more than 52,000 m3/day of water. This reservoir would meet (with a small surplus) projected urban water demand and downstream estate demand in Morogoro until 1986 when the water treatment capacity developed under the Project will be fully utilized. The small surplus of reservoir capacity in 1986 reflects the fact that the dam has a minimum economic height. The dam would include a 100 m long spillway sited on its rock abutment and draw-off arrangements. Raw water would fLcw by gravity to a 24,700 m /day capacity treatment plant designed to meet projected demand up to 1986. There, treated water would be pumped through a long pipeline to a 5,000 m3 reservoir. From the reservoir, one gravity main would supply the new industrial area. A second outlet would be connected to a booster station which would pump the water through a pipeline to an elevated tank connected to the existing distribution system. Replacement and extension in the reticulation system would involve the laying of additional pipes.  -15- 35. In addition to financing this construction (including the related consultant services) the Project would include financing of management advisory services and a training program for MWEM. The management advisory services to be financed under the proposed Pro- ject would provide three senior advisors to be attached to MWEM headquarters: a technical expert, a financial expert and a personnet and training expert. In addition to filling expertise gaps w-ithin MWEM and assisting in Project implementation, these experts would have a significant training role with respect to upgrading existing staff. Agreerient has been reached that senior advisors would be employed by MWEM by June 30, 1977 (Section 3.02b of draft Loan Agree- ment). 36. The trainina comnonent reflects an attempt by the rank to address important training needs which are not being adequately met by other available sources. Its focus on subprofessionals reflects a judgment that the professional level training which already has firm financing cannot be productively expanded at this time. Its specific focus on the Water Resources Institute (WRI),operated by MWEM for the pre-service trainina of water technicians, is a direct ouit- growth of a Bank examination of the needs of all categories of subprofessionals and how these needs can be most effectively met. Under the Project, the WRI would be expanded and diversified. While the WRI is already being expanded physically to raise the production of technicians from about 20 annually in 1976 to about 120 annually in 1979, no funds have been secured for construction of workshop, laboratory and library facilities. The Project would therefore pro- vide $1.5 million equivalent for construction of these priority facilities and would provide finarcing to help diversify the training program by upgrading courses for rechanics and electricians in the present curriculum. In addition, to provide flexibility as training opportunities arise or priorities change, $0.5 million is included for unallocated training expenditures. Finally, to assist MWEM in implementing its training programs, a second full time training officer would be appointed in the Ministry (Section 3.08 of the draft Loan Agreement). ErQir tZjaoasliolg 37. The estimated total cost of the Project is about TSh 160 million (US$19.2 million) including estimated taxes of US$0.7 million but. excluding interest and commitment charges during construction.  -16- Foreign exchange costs would be equivaLent to about US$9.4 million (49 percent). These estimates are expressed in December 1976 prices and are based on data provided by the feasibility studies. Physical contingencies of 15 percent have been added, except for construction costs of Morogoro dam where contingencies of 20 percent were added because rock excavation costs could be higher than in standard base estimates. Price contingencies toralling about 19 percent of total Project costs were included, based on the appraisal mission's best judgment of the future price changes in Tanzania. These cost esti- mates are detailed in Annex III. 38. The Project is expected to be financed as follows: US$ Million IBRD Loan 15.0 Government contribution 3.5 Total (net of taxes) 18.5 The Dank's contribution cf 31 percent of net Project costs is in line with other Bank projects in Tanzania. 6dminis Lali n_an _Eanauem ea 39.. MWEM would have responsibility for Project implementation. For that purpose MWEM would be assisted by the three senior advisors financed under the Project in the management advisory services com- ponent. To carry out the detailed design and supervise the construction of new works in Morogoro, MWEM appointed Sir Alexander Gibb and Part- ners as engineering consultants. This firm, which updated the feasi- bility studies for this Project, is acceptable to the Bank. To carry out desiqn and supervise construction of new training facilities for WRI (parcgrarh 76), MWEM has appointed NEDECO from the Netherlands. It is already working, financed by local funds. Engineering con- sultants acceptable to the Bank wouLd be retained during Project implementation for both of these components (Section 3.02a of the draft Loan ^greement). - 10. To actually run the water supply system (once constructed) an urban water supply unit would be created and maintained in Morogoro, responsible to the Regional Water Engineer (Section 3.05 of the draft  -17- Lcen 'greement). 'Headed by a manager and including assistant managers for operations and accounts and administration, this unit would handle operation, maintenance, billing and accounting for the Morogoro system. MlIEM would eventually set up similar units in other towns covered by the urban water supply fund (paragraph 26). The design of the fund itself and the timing of its implementation have been reviewed in detail and found acceptable to the Bank (Section 3.06 of the draft Loan Agree- ment). Because some adjustments in the sector!s orginizational frame- work may prove necessary with experience, these arrangements would be reviewed from time to time with the Bank to identify and correct any difficulties (Section 3.07 of the draft Loan Agreement). Annual financial accounts for the Morogoro urban water supply unit and the urban water supply fund wouLd be audited by the Auditor General or other independent auditors acceptable to the Bank and the audited accounts would be sub- mitted to the Bank not later than six months after the close of the year (Section 4.02b of the draft Loan Agreement). Procurement and Disbursement 41. All contracts over $100,000 (which would comprise most of the work financed under tha Project) would be awarded on the basis of in- ternational competitive bidding in accordance with the Bank's guidelines on procurement. A preference of 15 percent, or the applicable customs duty, whichever is lower, would be applied to the bids of local manu- facturers for equipment supply contracts. A preference of 7.5 percent would be applied to bids from local civil works contractors. Contracts under $100,000 would be awarded in accordance with Local procedures which are satisfactory to the Bank. Total expenditures under such contracts would not exceed $500,000. 42. The Bank loan would be disbursed for 80 percent of the total cost of civil works; 100 percent of foreign expenditures on equipment which is directly imported, 100 percent of ex-factory costs of locally manufactured equipment and 80 percent of total costs of locally pro- cured imported equipment; and 100 percent of foreign expenditures on consulting services, the senior advisors financed under the Project and training expenditures not related to equipment or civil works. The loan would be disbursed over 3-1/2 years as detailed in Annex III. To prevent delays in Project start-up, the Government has retained the consultants who prepared the feasibility study to complete detailed engineering for the Project. The expenditures incurred for these services since July 1, 1976 and expenditures related to the site investigation at Mindu up to an aggregate amount of US$500,000 are recommended to be financed retroactively.  -18- UrbAka.2rj_Ia riffa 43. Stated Government policy with regard to water tariffs is one of non-subsidization for the urban water suppLy sector as a whole. A second general policy is that of a unified price for water throughout the entire country. Finally, as a result of previous unsatisfactory experience with charging through kiosks and the Government's firm and consistent commitment to equity, no charge is collected directly from consumers for standpipe water. 44. To promote efficient resource allocation, tariffs should at least reflect the long-run marginal cost of water (which is taken here to equal the average incremental cost of water to be supplied from expansion schemes). Calculations of the incremental cost of water in the seven towns which were originally appraised indicate that, as- suming an opportunity cost of capital of 12 percent for Tanzania, the overall average incremental cost is roughly in line with the present tariff of TSh 2.20/m3 (which reflects the substantial 67 percent tariff increase of July 1975). However, since the Morogoro system is one of the most expensive of the seven appraised, the average incremental costs of its water are higher than the present tariff. While on efficiency grounds this would justify differential tariffs to reflect the dif- ferent costs of water supplied, the Government has decided to maintain its present policy of charging uniform tariffs for all urban water supply systems. Agreement was reached that water tariffs would be maintained at the present level in real terms through periodic review and adjust- ment in accordance with criteria to be agreed with the Bank (Section 4.03 of the draft Loan Aareement). Such adjustments would be based on an appropriate Tanzanian price index. 45. Since only a low proportion of water usage in Morogoro will involve standpipes (about 14%), this is not a major issue on pruject grounds. However, because the percentage of urban water con- sumed nationally through standpipes is much higher, the present Government policy of not charging for standpipe water would make it impossible for the urban water supply fund to be financially viable if there is no payment from any source for standpipe water. To resolve this problem it was agreed that the Government would compensate the fund for water consumed through standpipes (Section 4.04 of the draft Loan Agreement).  -19- Project Monitoring System 46. Important project monitoring concerns will be the staffing and training program, technical operations, financial performance and physical construction of the water supply system. The Government has agreed to establish a project monitoring system (Section 3.04c of the draft Loan Agreement) and proposed guidelines have been developed. Project Justification 47. The existing water supply situation in Morogoro is inadequate. Average water demand is far above the safe output of the river source and there are severe water shortages and low pressure problems in the dry season. With the projected population and industriaL growth in Morogoro the deficit in water production facilities would grow rapidly, presenting major health problems and a binding constraint to industrial development. This Project has, therefore, been developed to address the existing water problem and meet the growing water needs of Morogoro up to 1986. It would make up the deficits faced by Morogoro's present population (39,000) and provide adequate water to 27,000 additional people up to 1986. It would also increase sevenfold the amount of water available for industrial and commercial use. Industrial use is particularly important as Morogoro's central location has been fundamental in the Government's decision to place its first major industrial estate in Morogoro. Without this water supply project the estate could not be implemented as a number of key industries are large water consumers. The directly productive implications of this Project are reflected in the fact that by 1986 it is estimated that industrial and commercial water consumption will represent about 59 percent of totaL consumption. 48. This Project will also have significant institutional impact on the water supply sector. As a result of Government/Bank discussions it appears that the first step has been achieved in the direction of establishing commercial operations in the sector. With the establish- ment of the urban water supply fund, water revenues and expenses will now be separate from general revenues and expenditures, therefore providing a basis for ensurinr adequate financial resources for operations, maintenance and future minor expansions. Finally, the Project includes significant investment in manpower development, a major constraint to improvement in sector performance. This invest- ment would focus on the training of technicians and on improving the management and technical capabilities in MWEM.  -20- 49. The return on investment is estimated at 7.9 percent. This return does not measure the full benefits of the Project since the social and health benefits cannot be quantified. It is expected that the risk of diseases attributed to the lack of safe and adequate water supply would be substantially reduced. Risks 50. Detailed technical examination by the consultants and the appraisal team have indicated that there are no undue risks with regard to the physical construction of the water supply system in Morogoro. Nevertheless, because of the staff shortages in the sector, delays in the execution of the Project are possible. Provisions for the Project engineering consultant and management advisory services are expected to resolve this potential weakness. Furthermore, close Project supervision should ensure that any problems in this area are quickly addressed. 51. An area of major difficulty once the water system is estab- lished -- the provision of adequate funds for operation and maintenance -- is also a focus of concern. However, the institutional arrangements which were recently adopted by the Government significantly reduce the risks in this area. The urban water supply fund concept is expected to lead to far greater fiscal accountability in the sector. Further- more, it is expected that a continuation of the dialogue with the Government on the water sector (in regional rural development projects, as well as in the supervision of the proposed Project and in possible other efforts) wiLl lead to a further strengthening of these insti- tutional arrangements and lessen significantly the probability of future problems. PART V - LEGAL INSTRUMENTS AND AUTHORITY 52. The draft Loan Agreement between the United Republic of Tanzania and the Bank, the recommendation of the Committee referred to in Article V, Section l(d) of the Articles of Agreement of the Rank and the draft resolution innrnvinrl the nronosed Lnan ir hpinn distributed to the Executive Directors separately. The draft Loan Agreement conforms generally to the pattern of the Bank's agreements for water supply projects.  -21- 53. Features of the draft Loan Agreement of special interest are referred to in paragraphs 35, 36, 39, 40, 44, 45 and 46. 54. I am satisfied that the proposed Loan would compLy with the Articles of Agreement of the Bank. PART VI - RECOMMENDATION 55. I recommend that the Executive Directors approve the proposed Loan. Robert S. McNamara President Attachments December 7, 1976   ANNEX 1 Page 1 of Ij pages TAOLE 3A TANZANIA- SOC IAL INDICATORS DATA SHET LAND AREA (THOU KN2) --