Tanzania - Second National Sites and Services Project

Report No. 1518a-TA FILE COPY Tanzania: The Second National Sites and Services Project June 9, 1977 Urban Projects Department FOR OFFICIAL USE ONLY U Document of the World Bank This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  TANZANIA SECOND NATIONAL SITES AND SERVICES PROJECT Currency Equivalents Currency Unit Shilling (T Sh) Tanzania Cent 100 = T Sh 1 T Sh 1 US$0.12 US$ = T Sh 8.30 Measurement Equivalents and Abbreviations 1 meter (m) = 39.4 inches = 3.28 feet 1 kilometer (km) = 0.62 mile 1 square kilometer (km2) = 0.386 square mile 1 cubic meter (m3) = 35.3 cubic feet (cu ft) 1 liter = 0.26 US gallon 1 cubic meter per day (m3/day) = 35.3 cubic feet per day I cd - liters per capita per day 1 hectare (ha) 2.47 acres Other Abbreviations and Acronyms Ardhi = Ministry of Lands, Housi-ng and Urban Development Treasury = Ministry of Finance and Planning ECC = Economic Committee of Cabinet THB = Tanzania Housing Bank NBC = National Bank of Commerce SIDO = Small Industries Development Organization RDD Regional Development Diriector ADDD = Assistant District DevelDpment Director RLDO = Regional Land Development Officer DLDO = District Land Development Officer TANESCO = Tanzania Electricity Supply Company MCH = Mother and Child Health CCM = Chama Cha Mapinduzi SSI = Small Scale Industry Government's Fiscal Year: July 1 - June 30  FOR OFFICIAL USE ONLY TANZANIA SECOND NATIONAL SITES AND SERVICES PROJECT TABLE OF CONTENTS Page No. SUMMARY AND CONCLUSIONS . ................. . . . . . . . . . . i-iii I. INTRODUCTION .. *.................................... 1 II. BACKGROUND ......................... ............ 2 A. The Urban Sector ..... ......................... 2 B. The National Sites and Services Program ....... 5 III. THE PROJECT .. ... ........ ...... 8 A. Objectives .................... ... ..0..................... 8 B. Main Components ..... ................... ....... 8 C. Detailed Features 9.......................... 9 D. Implementing Schedule ... .... .. ..... .... ....... 14 IV. COST ESTIMATES AND PROJECT FINANCING .............. ... 15 A. Cost Estimates .............. ......... ........ 15 B. Financing .................... 15 C. Land Rent and Service Charge .................. 19 D. Accounts and Auditing ...................... 20 E. Procurement and Disbursements ................. 20 V. ORGANIZATION AND MANAGEMENT ....... . . ...................... . 22 A. Ministry of Lands, Housing and Urban Development (Ardhi) .................. 22 B. Regional and District Administration .......... 23 C. Tanzania Housing Bank (THB) ................... 24 D. Small Industries Development Organization (SIDO) ........................ 25 VI. PROJECT JUSTIFICATION AND ECONOMIC BENEFITS ........ 27 VII. AGREEMENTS REACHED AND RECOMMENDATIONS ............. 30 This report was prepared by the mission, Ms. Nassau, Ms. Garrity and Ms. Mayne-Nicholls. This document hIs a etricted distribution aed may be usd by recipients only in the peirfamsme of their olcWl dutie. Its contents may not otherwie be disclosed without World Bnk uthorization.  ANNEXES ANNEX 1 Squatter Settlements and Project Towns ANNEX 2 Community Facilities ANNEX 3 House Improvement and Construction Loans Program ANNEX 4 Small-Scale Industry Program ANNEX 5 Implementing Agencies and Project Organization and Management ANNEX 6 Design Standards and Cost Estimates ANNEX 7 Cost Recovery of Urban Infrastructure ANNEX 8 Technical Assistance ANNEX 9 Economic Evaluation: Squatter Upgrading and Surveyed Plot Program ANNEX 10 Monitoring and Evaluation ANNEX 11 Estimated Schedule of Disbursement CHART 1 IBRD 16747 Implementation Schedule MAPS IBRD 12572 Tanzania - Project Towns IBRD 12573 Dar es Salaam - Site Locations IBRD 12574 Tanga - Site Locations IBRD 12575 Tabora - Site Locations IBRD 12576 Iringa - Site Locations IBRD 12577 Morogoro - Site Locations IBRD 12595 Typical Upgrading Layout - Hanna Nassif (DAR)  SUMMARY AND CONCLUSIONS i. The proposed project is the second IDA financed project in support of Tanzania's National Sites and Services Program. Like the currently on-going first project (Credit 495-TA), it is aimed at enhancing the living conditions of the urban poor through basic improvements in their environment and creation of economic opportunities. The most important objectives underlying IDA sup- port of Tanzania's urban sector are (a) to demonstrate a replicable approach which responds effectively to the shelter and employment problems of the urban poor and (b) to strengthen the institutional and financial capability of Government to implement such an approach on a continuing basis. ii. Tanzania's cities are characterized by rapid population growth, a deteriorating employment situation, and proliferation of inadequately serviced squatter settlements. The urban population, currently growing at an estimated 9% per annum, increased from 5% to about 9% of the national total during 1967-76, with the bulk of the growth contributed by migrants from the countryside. The economic situation of urban residents has been worsening. Average household cash incomes in urban areas fell by about 50% between 1969 and 1976 as unemployment more than doubled to its present level of about 17% of the urban labor force. The rapid growth of the cities has not been planned or accompanied by a corresponding expansion of urban services. Consequently, between 40 to 70% of the residents of Tanzania's main towns are now living in uncontrolled, inadequately serviced squatter settlements. iii. Prior to the adoption in 1972 of the National Site and Services Program as official policy and the subsequent implementation of the program with IDA assistance, little was done to improve the living conditions of the urban poor. Government generally ignored the squatter population or when political pressure mounted, razed specific squatter settlements and relocated the residents. Low-cost housing programs sponsored by Government were inadequate and, because of high standards, priced the poorer groups out of public housing. iv. Under the first project, about 9,000 sites and services plots and improvements affecting 9,000 houses in existing squatter settlements are being provided to benefit approximately 160,000 low-income residents in Dar es Salaam, Mwanza and Mbeya. The execution of the physical aspects of the first project has been proceeding satisfactorily. There are, however, deficiencies in the social aspects, financial management and cost recovery arrangements in the project. The second project will address these defici- encies in the context of broader, sector-wide policy measures. v. The project components to be financed under the proposed IDA credit are: (a) Squatter Upgrading: Provision of secure tenure and basic urban infrastructure including water supply, drainage and roads to about 16,000 houses in existing squatter neighborhoods;  - ii - (b) Basic Surveyed Plots: Preparation of about 19,000 sur- veyed plots in planned residential layouts to be allocated to low-income applicants in the five towns; (c) Community Facilities: Primary schools, health and nutrition centers, and markets for each of the neighborhoods to be upgraded; (d) House Improvements and Construction Loans: Continuation of the lending program established under the first project for building materials loans for housing improvements in upgrading areas and construction of new housing or surveyed plots; (e) Small-Scale Industry: An experimental program of assistance to small-scale industries in Tabora and Tanga. The program would provide infrastructure for industrial clusters, credit for equipment, and management and technical assistance to selected small enterprises; (f) Equipment and Vehicles: Provision of the equipment and vehicles needed for project implementation, the surveyed plot program, maintenance and garbage collection; and (g) Technical Assistance: Financing for consultants' services for detailed engineering and construction supervision; training of Tanzanian personnel in executing agencies; management support; monitoring and evaluation of the project; a study of pricing and taxation of land and shelter services in urban areas; and preparation of a follow-up project. vi. The project has been designed with minimal standards that will make it affordable by very low-income households. The estimated cost of infrastructural improvements in the squatter upgrading areas is T Sh 3,500 (US$422) per plot. For surveyed plots, the cost per plot (including pre- paration, surveying and provision of minimal access to water) is approximately T Sh 210 (US$25). With household size averaging nine persons, the per capita costs are approximately T Sh 390 (US$47) and T Sh 23 (US$3) respectively. vii, Based on preliminary engineering of the upgrading sites, total project costs are estimated at T Sh 243 million (US$29.3 million) includ- ing T Sh 6 million for land acquisition. The direct and indirect foreign exchange content of the project cost will be approximately 36%. It is proposed that IDA finance 70% of the cost of infrastructure and community facilities, 75% of the technical assistance costs and 100% of the costs for equipment, vehicles and imported cement for THB's Materials Loans Program. The balance would be financed by Government and THB. Accordingly, an IDA credit amounting to T Sh 100 million (US$12 million) is recommended.  - iii - viii. The Ministry of Lands, Housing and Urban Development (Ardhi) will continue to be the principal executing agency, responsible for the infra- structure works and day-to-day management of the national program. Ardhi will be assisted in this task by the respective regional administrations and other concerned ministries and parastatals. Ardhi's planning and execution capa- bilities will be enhanced through actions agreed by Government as well as technical assistance provided under the project. The Tanzania Housing Bank (THB) and the Small Industries Development Organization (SIDO) which will be responsible for the housing loans and small industries components will also receive technical assistance support. ix. The project will be executed over four and a half years, 1977-1981. Contracts for civil works totaling T Sh 59.7 million (US$7.2 million before contingencies) will be awarded after international competitive bidding in accordance with Bank/IDA Guidelines for Procurement. Local contractors will be allowed 7.5% preference over foreign bidders. Lesser works comprising mainly community facilities worth about T Sh 7.8 million (US$0.95 million) will be awarded on the basis of local competitive bidding in accordance with Government procedures acceptable to the Bank. Equipment, vehicles and cement worth T Sh 16.6 million (US$2.0 million) will be procured according to Govern- ment' s normal procedures acceptable to the Bank. x. The proposed project will benefit more than 315,000 low-income urban dwellers (26% of Tanzania's urban population in 1976) providing improvements to over 40% of the existing squatter houses and satisfying 75% of the need for new plots in five major towns. With the benefits being provided under thd first project, the population affected would be some 475,000 people--almost 40% of the country's current urban population. Ex- pected benefits of the project include (a) increased productivity resulting from improvements in living conditions and health standards; (b) increased employment; and (c) increased housing production. Technical assistance provided under the project is expected to result in significant improvement in public policies of urban taxation for land, shelter and infrastructure; employment (through assistance to the urban informal sector); and health (preventive health care and nutrition). The economic rate of return is estimated at 23% for the squatter upgrading and surveyed plots programs. xi. The project is suitable for an IDA credit of US$12 million on standard IDA terms subject to conditions set forth in para 7.01.   TANZANIA SECOND NATIONAL SITES AND SERVICES PROJECT I. INTRODUCTION 1.01 The project for which the Government of Tanzania is seeking an IDA credit is the second phase of the National Sites and Services Program. The first Tnase, involving new serviced plots and improvements in squatter communities in Dar es Salaam and two regional centers, is being implemented with the assistance of an IDA credit approved in 1974 (Credit 495-TA). The second project will extend the program to four other regional centers as well as to new sites in Dar es Salaam. The primary emphasis of the project will be on upgrading; support will also be provided for surveying plots which will be serviced as settlement occurs. Implementation capabilities developed in the first project will be expanded and innovations introduced in project design based on lessons of the first project. 1.02 The Ministry of Lands, Housing and Urban Development (Ardhi) began planning the project in 1975 using project preparation funds included in the first credit. Missions visited Tanzania in March and July 1976 to assist and review project preparation and in October 1976 to appraise the project. This report is based on the findings of the appraisal mission consisting of Messrs. P. Patel (Chief of Mission), C. Goldfinger, B. Rush, D. Coyaud, and Ms. N. Barry (Bank), and Messrs. G. Hughes and F. Temple (Consultants).  II. BACKGROUND A. The Urban Sector 1/ Urbanization Trends 2.01 Although one of the world's poorest, least urbanized countries, Tanzania has been experiencing a rapid rate of urbanization. The urban population is currently estimated to be growing at 9% per annum and the proportion of the population living in urban centers increased from 5% in 1967 to about 9% in 1976. Both census and survey data indicate that the rapid growth of Tanzania's towns has been the result of migration rather than natural population increase. Dar es Salaam (590,000) is by far the largest city in the country, followed by Tanga and Arusha (94,000 each). Data on the growth of Tanzania's urban centers are presented in Annex 1 and their locations are indicated in Map IBRD 12572. 2.02 Unplanned squatter settlements have proliferated. Official esti- mates of the number of squatter houses in Dar es Salaam imply an average annual increase of 16.6% during 1963-73 and a growth rate of 24% per annum during the last half of the decade. Between 40% and 70% of the residents of Tanzania's main towns now live in squatter settlements. Although these settlements are not dense by international standards and much of the housing is sound or improvable, the infrastructure in these communities is often inadequate or completely lacking, and their residents have less access to community facilities than other urban dwellers. 2.03 Another consequence of the rapid growth of Tanzania's towns has been an increase in urban unemployment. Population growth has considerably exceeded the creation of new employment opportunities. Regular wage employ- ment, which accounted for just over two-thirds of the employment opportunities in the urban sector in 1969, grew less than 6% a year during 1969-1976. Average real earnings for this type of employment declined slightly during this period. Informal opportunities (casual and self-employment) grew at some 11% per annum, but because low-income self-employment accounted for much of the growth, average real earnings in this sector were more than halved. As a result of these changes and growth in the labor force, average per capita real earnings fell by about 50% during this period and unemployment rose from approximately 8% to 17% of the labor force in urban areas. Furthermore, urban dwellers' purchasing power has been sharply reduced by increases in producer prices for key agricultural goods as well as inflation. Data for 1976 suggest that between one-third and two-fifths of Tanzania's urban households have cash incomes below the minimum wage of T Sh 380 (US$46) per month. Urban Sector Initiatives During the 1970s 2.04 Tanzania's socialist strategy of economic development, articulated in the 1967 Arusha Declaration, focuses primarily on agriculture and emphasizes 1/ More information on the sectoral background can be found in chapter II of the appraisal of the first project (Report No. 337a-TA of May 12, 1974).  - 3 - social equality and self-reliance. Implementation of this strategy has led to more equitable income distribution but has also affected the overall rate of economic growth. The villagization program contributed to a decline in agricult:uzal production; commitment to costly social programs such as universal water supply and universal primary education strained public re- sources, while nationalization of key economic sectors limited the flow of private investment funds. Administrative decentralization initiated in 1972 resulted in overextension of scarce administrative resources. These difficulties were exacerbated by the global recession of 1974. Faced with the economic crisis in 1974, the Tanzanian Government took a series of drastic measures aimed at increasing agricultural production, reducing the growth of recurrent expenditures, and freezing wages and salaries. As a result of these measures, agricultural production increased and economic prospects improved. Nevertheless, budgetary and manpower constraints on public policies are likely to remain severe in the foreseeable future. 2.05 The national development strategy assigns low priority to urban areas and the urban sector. Since a comprehensive urban policy has not as yet been explicitly formulated, specific measures and initiatives affecting the urban sector have evolved mainly as a result of broader policy actions. Four aspects of the urban sector have been particularly affected: urban rents and incomes; land development and housing programs; industrial and administrative decentralization; and urban administration and finance. 2.06 Urban Rents and Incomes. In order to reduce the gap between urban wage earners and rural farmers, the Government has tried to restrain wage increases for better-paid workers and raised producer prices for key cash crops. As a part of this policy, since 1969 rents paid by public sector employees living in publicly owned housing have been determined on the basis of their incomes rather then the characteristics of the units. The policy has had the effect of redistributing real incomes more equitably among public sector employees. Since Government is the country's main employer, the policy also has had a considerable impact on the urban income distribution. The nationalization of rental buildings and the imposition of rent controls further equalized the distribution of wealth in urban areas, but also discouraged new private investment in construction and adequate maintenance of existing buildings. As a result of pricing, income and taxation policies, the esti- mated differential in consumption expenditure between the highest and lowest paid public sector employees has been reduced from 60 to 1 at Independence (1961) to 13 to 1 today. 2.07 Land Development and Housing Programs. The Government's present shelter policies 1/ emphasize improvement of existing squatter settlements and development of new residential areas through surveyed plot programs. The 1/ Based on Ardhi papers on National Housing Policy and on the National Sites and Services Program, designed to meet the shelter needs of 13 regional centers over a ten-year period. These papers were endorsed by the Economic Committee of the Cabinet (ECC) in 1972.  -4- improvement program provides security of tenure in existing settlements and basic environmental improvements. The surveyed plots program uses limited capital resources to provide a planned context for self-help residential construction, offering an affordable alternative to squatting. 1/ Although no services (except minimal access to water) are provided initially, the regional authorities are expected to respond to demands for services generated by settlement on the plots. 2.08 Industrial and Administrative Decentralization. The Second Develop- ment Plan (1969-74) designated eight towns outside Dar es Salaam for preferen- tial industrial development. The third plan (1975-80) currently under review will emphasize the development of all 20 regional capitals. However, the lack of transport and communication facilities in peripheral regions continues to favor industrial location in established growth centers. 2.09 Administrative decentralization appears to have been more success- ful. Initiated in 1972, it shifted 40% of the central establishment from Dar es Salaam to the regions. In 1973, the Government decided to move the national capital to Dodoma and the office of the Prime Minister has already been transferred there. Given the huge investment required to develop access and services infrastructure and the severe budgetary constraints, it is doubt- ful whether the original construction program can be completed on schedule. 2.10 Urban Administration and Finance. Far reaching changes in the political-administrative structure have been introduced by the decentraliza- tion reform of 1972. 2/ Independent local authorities were abolished and replaced by districts and subdistricts grouped into regions. Although some districts and subdistricts correspond roughly to urban areas, there is now no urban government per se. While substantial decision-making powers were transferred to the regions, the public finance system was centralized. Since all fiscal revenues are now collected and decisions a-bout regional budgetary allocations are made by the Treasury, the link between revenue collection and expenditure in urban areas has been broken. 2.11 Tanzania's fiscal system uses taxation to redistribute income between income groups and between rural and urban areas. The tax structure is highly progressive and favors general rather than specific forms of taxation in order to minimize differences in tax burdens due to location. An effort to reduce the share of local taxation in total tax intake has been undertaken since 1969. In line with its general policy, in 1974 Government abolished a variety of local property taxes and rates, replacing them with a single levy, the Land Rent and Service Charge. This is an annual charge equal to 10% of the economic value of unimproved land as determined periodically by Ardhi valuers (See para 4.08) 1/ Government owns the land needed for implementation of this program as a result of the nationalization of land in 1963. 2/ For an extensive discussion of decentralization see Bank Report No. 587-TA "Tanzania: Fiscal Aspects of Decentralization." (See also Annex 7).  - 5 - 2.12 Since the revenues from the Land Rent and Service Charge are not earmarked for any specific use, their level is not determined by urban infra- structure expenditure needs but by general budgetary considerations and by the pace of Ardhi's valuation. The latter is slow, hampered by a lack of resources and manpower and somewhat cumbersome valuation methods. There are also delays in transmitting updated valution rolls from Ardhi to Treasury. Furthermore, because of the equivocal wording of the Land Rent and Service Charge Act of 1974, the charge has not been levied at all in squatter areas. Consequently, the potential revenues of the charge are considerably under- estimated. The collection of charges is poor: only slightly more than half of the estimated revenues are being collected and arrears are large (see Annex 7). 2.13 While the 1972 reform specifically sought to transfer public ex- penditures from urban to rural areas, its impact on the former was accentuated by the general budgetary crisis and lower than expected yield of subsequently introduced new levies such as the Land Rent and Service Charge. The result was a heavy decline in the resources available to urban areas in absolute as well as in real terms. The lack of funds for maintenance and new infrastruc- ture investment caused a large-scale deterioration of urban infrastructure, necessitating stop-gap remedial measures. Recently, for example, the Ministry of Works had to undertake a large program (T Sh 25 million or US$3 million) of road repairs in Dar es Salaam. 2.14 In recent months the Government has become increasingly aware of the shortcomings of the present administrative and financial arrangements for urban areas and the Economic Committee of the Cabinet (ECC) has initiated a review intended to produce recommendations to alleviate the problems con- fronting the towns. As a part of the review, Ardhi has undertaken a study of the Land Rent and Service Charge to be financed under the project which will produce recommendations for improving the policy of taxation of urban land, shelter and infrastructure (para 2.21 and 4.08-4.09). B. The National Sites and Services Program Review of the First Project 2.15 The performance of Bank projects in Tanzania has been uneven, hampered by shortage of local administrative and technical manpower and severe budgetary constraints. Given this context, the performance of the first project has been adequate and in many respects quite satisfactory. A workable approach to low-cost, low-standard shelter has been developed. The execution of physical infrastructure for the sites and services and squatter upgrading areas has been carried out on schedule. The design and implementation capa- bility of Ardhi has been strengthened. In the second project, this approach will be continued and the scope of the project will be expanded to include employment and income generation for poor urban dwellers. The design of spe- cific project components will be modififed to overcome deficiencies. To assist Ardhi, greater emphasis will be given to the financial management and social  aspects of project implementation, where problems were encountered in the first project. The issue of plot charges will be addressed in a broader context of public policies of taxation of urban land, shelter and infras- tructure. 2.16 The composition of the second project will emphasize squatter upgrading and provision of surveyed rather than serviced plots. This will permit extension of the benefits of the project to a larger number of house- holds in more towns so that more families in lower income groups can be reached. The proposed project will benefit almost twice as many people as the first project, although infrastructure expenditures in the two projects will be approximately equal in real terms. 2.17 The standards for the education and health community facilitites originally planned for the first project were higher than facilities currently being provided elsewhere. In the second project, plans for the construction of community education centers, including workshops, assembly hall and dis- pensary as well as classroom, have been revised and emphasis is being given to the provision of classrooms (para 3.08). Medical facilities in the first project have high capital and recurrent costs and reinforce the emphasis traditonally given to curative care in Tanzania's urban areas. The proposed project will emphasize preventive care at Mother and Child Health Clinics (MCHs) and support an MCH extension program focused on nutrition and environ- mental sanitation (see para 3.09). 2.18 Early experience with house construction and improvement loans indicated that prdject allotees were discouraged by THB's application pro- cedures and fees and that THB was reluctant to incur the risks associated with lending to relatively poor households with informal sources of income. To reduce bureaucratic barriers to lending THB agreed to open part-time offices in project areas and to reduce valuation fees for project applicants. Furthermore, in order to make loans accessible to applicants from upgrading areas, THB has begun applying lending procedures used in its rural operations which do not require mortgages as security. Another problem confronting pros- pective low-income households desiring to build or improve their houses is limited access to building materials, particularly cement, at reasonable prices. In the second project materials supply through on-site depots will be introduced to complement cash lending (see para 3.10-12). 2.19 Implementation of the first project was handled by Ardhi's Sites and Services Section in collaboration with regional administrations. In the second project, this arrangement will be continued and an effort will be made to improve specific aspects of implementation. The financial man- agement capability of Ardhi will be strengthened and more emphasis will be given to social aspects of project implementation, particularly in community development and assistance to project beneficiaries. A more formal, system- atic approach to coordination between Ardhi, regional administration and local project management will be introduced (see para 5.02-5.08).  2.20 Cost recovery has been a critical problem in the first project. At the time of appraisal of that project Government did not have any coherent policies or guidelines on cost recovery that could be reconciled with the Bank's requirements. A special formula was therefore worked out to be applied exclusively to project areas. Charges calculated according to this formula were intended to recover the full costs of on-site infrastructure over 25 years at 9% interest. This proved difficult to implement because the level of charges set by the formula were much higher than in comparably serviced adjacent areas where plot charges are grossly under-assessed by the Land Rent and Service Charge Act of 1974 (para 2.11). This situation resulted in political controversy and allottees refused to pay the fees required to obtain title to their plots. The Government therefore requested an amendment of the credit agreement in order to bring charges for project areas within the purview of the Land Rent and Service Charge Act. Plot charges resulting from this amendment will allow recovery of 75% of the capital costs assignable to project beneficiaries amortized over 25 years at 9% interest. The proposed amendment was approved by IDA in February 1977 (see Annex 7). 2.21 In the second project, the question of plot charges will be addressed in the broader, sectoral context of a review of fiscal policies toward urban land, shelter and infrastructure. A reform of the Land Rent and Service Charge will be undertaken to improve its effectiveness, equity and yield. The long- term goal of the reform will be to increase the revenues from the urban sector to a level that would cover investment and operating expenditures for urban infrastructure. An in-depth study of the Land Rent and Service Charge was initiated in February 1977. At negotiations the inception report of the study was reviewed and short-term measures as well as guiding principles for the re- form of the Land Rent and Service Charge were agreed upon (see para 4.08-4.11). Selection of Project Towns 2.22 The first project included components in Dar es Salaam and two up-country towns, Mwanza and Mbeya. Ardhi's principal criteria in selecting towns for the second project were the magnitude of the squatter settlements and demand for plots as determined by the number of applications at the District Land Offices. Other factors such as potential compensation costs for displaced squatters, direct requests from regions, geographical balance and availability of assistance from other sources were also given consideration. 2.23 On the basis of Ardhi's recommendations, Dar es Salaam and four more up-country towns are included in the second project: Tanga, Morogoro, Tabora and Iringa. The five towns had an estimated combined population of 860,000 in 1976, 60% of the total urban population of Tanzania. The number of squatter houses in these towns is estimated at 51,500, about 73% of the estimated national total of urban squatter housing. The two projects will serve seven of the ten largest urban centers in mainland Tanzania, containing 70% of the national urban population and 84% of the squatter housing.  -8- III. THE PROJECT A. Objectives 3.01 The project will continue IDA assistance to Tanzania's urban sector started under the National Sites and Services Project (Credit 495-TA). Based on the experience gained in the first project, this project will provide financial and technical support necessary for the implementation of the country's National Site and Services Program over the next four years (1977- 1981). The most important objectives underlying IDA assistance to Tanzania's urban sector are: (i) to demonstrate a replicable approach which responds effectively to the problems of improving existing slums and controlling rapid new growth; and (ii) to strengthen the institutional and financial capability of the Government to implement such an approach. 3.02 The project will provide basic infrastructure and community facil- ities to 160,000 low-income residents of eleven squatter areas in five towns. The program of surveyed plots for low-income housing will provide about 19,000 plots satisfying 75% of the estimated five-year demand for new plots in project towns. House improvement and construction loans, on-site technical assistance for self-help housing consolidation and support for community development will complement the provision of infrastructure and community facilities. The project will also initiate a pilot program of assistance to small-scale industry in the urban informal sector. Technical assistance provided under the project will support detailed engineering and construction supervision; training of Tanzanian personnel; management support for the implementing agencies; preparation of a follow-up project; and monitoring and evaluation. Assistance will also be provided to Government's efforts to rationalize its policies of taxation of urban land, shelter and infrastructure and to increase its revenues from the Land Rent and Service Charge. B. Main Components 3.03 The project comprises the following interrelated components in Dar es Salaam, Tanga, Tabora, Morogoro and Iringa: (a) Squatter Upgrading: Provision of secure tenure and basic urban infrastructure including water supply, drainage, footpaths, roads and street lighting to about 15,800 houses in existing squatter neighborhoods; (b) Surveyed Plots: Preparation of about 19,000 surveyed plots in planned residential layouts to be allocated with legal titles to low-income applicants in the five towns; (c) Community Facilities: Provision of community facilities in- cluding primary schools, health centers, and markets to the upgraded neighborhoods;  - 9 - (d) House Improvement and Construction Loans: Assistance to the THB for continuation and improvement of the lending program established under the first project for building materials for self-help housing improvements in upgrading areas, construction of new housing on surveyed plots, and construction of worksheds for the small-scale industry program; (e) Small-Scale Industry: An experimental program of assistance to existing and new small-scale industries in Tabora and Tanga. The program would provide serviced sites for about 150 workshops, credit for equipment and technical assistance to selected small enterprises, cooperatives and artisans; (f) Equipment and Vehicles: Provision of the equipment and vehicles needed for project implementation, the surveyed plot program, maintenance and garbage collection; and (g) Technical assistance: Financing for consultants' services for detailed engineering and construction supervision; training of Tanzanian personnel in executing agencies; management support; monitoring and evaluation of the project; a study of pricing and taxation of land and shelter services in urban areas; and preparation of a follow-up project. C. Detailed Features Squatter Upgrading 3.04 A simple and inexpensive approach to infrastructure upgrading, developed and tested under the first project, will be applied in the proposed project with appropriate modifications. Selection of viable neighborhoods was determined by several criteria such as location, economic and transport linkages, technical feasibility and need for services (see Annex 6). The sites selected for upgrading are identified on maps of each project town (see Maps IBRD 12573-7). 3.05 Functional standards for infrastructure, described in Annex 6, are as low as possible initially so that the per family costs can be afforded by households at the lower end of the income scale (see Annex 6, Table 4). However, provision is made in the planning for future upgrading of standards as household incomes increase. Layouts and preliminary engineering for all sites have been completed by consultants (see IBRD Map 12595 and Annex 6, Table 5, for typical examples). Each site will have a network of collector, primary, secondary and pedestrian roads. Collector roads will be macadam sur- faced, as required for bus traffic, whereas primary roads will be gravelled, providing vehicular access to the neighborhoods. Secondary and pedestrian roads servicing inner areas will be compacted dirt roads. Carriage-way widths on the surfaced roads will be kept to a minimum, although the rights-of-way  - 10 - will be wider for future upgrading. Water kiosks with two or four standpipes will provide one standpipe for every 50-70 houses within a maximum walking distance of 250 m. Sewage disposal will be provided by owner-installed tra- ditional pit latrines where soil conditions permit. Where groundwater condi- tions are unsuitable, house owners will be provided technical assistance and TUB loans to install alternative drycell systems. The initial power distri- bution system will provide street lighting on the main roads and electricity for the community facilities and small industries workshops. The drainage system, designed to carry storm water and domestic waste water other than from latrines, consists of open earth ditches along roads. Garbage disposal serv- ice will be provided in the Dar es Salaam sites (where the problem is most acute) by trucks financed under the credit. Surveyed Plots 3.06 Under the program for providing new plots for low-income housing Ardhi will prepare sufficient surveyed sites to satisfy 75% of the estimated demand in the five towns (see Annex 1). The sequence of planning and imple- mentation of the program would be as follows: (a) site selection (see Annex 6 for criteria); (b) layout design (see Annex 6 for criteria); (c) site clearance and surveying of plots; (d) installation of skeletal water supply through standpipes; (e) allocation of plots (see Annex 5 for criteria); (f) settlement and house construction by allottees; and (g) provision and upgrading of basic infrastructure in stages. Financial and technical assistance for (a) - (f) will be included under the project; (g) would be initiated by residents with assistance from regional authorities and Ardhi. 3.07 The layouts for surveyed sites will be designed for the same even- tual standards as serviced sites (Annex 6), but unlike the typical sites and services project, all on-site services except water supply will not be sup- plied until after settlement has occurred. This approach will enable the project to reach a greater proportion of the target population because of the lower initial capital requirements. It affords one of the few practical options available for policy action before squatting occurs.  - 11 - Community Facilities 3.08 In each of the communities to be upgraded a careful evaluation of the needs for community facilities over and above those existing was made to determine the range and number of facilities to be included under the project. The criteria used in the evaluation are discussed in Annex 2 and standards and cost estimates are shown in Annex 6. Twelve primary schools will be constructed to meet the needs of project areas in conformity with Tanzania's program of universal primary education. The schools will be built to minimum standards and provide only classrooms in view of the severe problem of recurrent financing and stated preference of local education officers. 3.09 Health facilities under the second project will be designed with a greater emphasis on preventive care and will focus on the groups most in need of it: mothers and small children. Two new Mother and Child Health Clinics (MCHs) and a medical outpost will be constructed and three existing MCHs upgraded. The Ministry of Health will prepare guidelines to develop an urban MCH program similar to the existing MCH program for rural areas. Particular attention will be given to extension work in which nutrition officers and nurses will visit the homes of families with malnourished children and give courses in the community on nutrition and environmental sanitation. The Tanzanian maternal and child health care program includes pre- and post-natal care for the mother, immunization services, and child spacing advice and services. House Improvement and Construction Loans 3.10 The Tanzania Housing Bank (THB) will continue its lending program for house improvement (in upgrading areas) and low-cost housing construction (in overspill and surveyed plot areas) under the second project. The program will incorporate modifications designed to address two major problems encoun- tered in the first project: (i) availability of key building materials to project beneficiaries; and (ii) access to credit for squatter inhabitants. 3.11 In order to improve the access to and supply of building materials to project beneficiaries, THB will establish on-site offices and materials depots. Critical building materials including cement, corrugated iron roofing sheets and joinery will be stocked at the depots by THB. Most of the materials will be purchased locally by THB using its own resources. However, due to a severe shortage, cement will be imported until 1980 when two new cement plants in Tanzania should start production. Foreign exchange required for importing cement for the first three years is therefore included in the project. In order to minimize dependence on imported materials new building materials and techniques will be introduced through demonstration houses (see Annex 3, Figure 3, for alternative housetypes to be developed). 3.12 In order to encourage lending in squatter areas THB will step up publicity on the availability of loans in project areas and establish on-site offices for distribution and collection of application forms. Furthermore, lending criteria and procedures will be less stringent than those normally employed by THB (see para 5.12).  - 12 - 3.13 Materials loans averaging T Sh 4,000 (US$482) for house improvements in upgrading areas and T Sh 13,000 (US$1,556) for construction of new houses in surveyed plot areas will be offered to project participants (see para 4.05 for terms). Borrowers will have the option of taking their loans in cash and/or vouchers redeemable at THB materials depots on project sites. In addition, loans for construction of worksheds up to T Sh 120,000 (US$14,500) will be offered to participants in the small-scale industry program in Tanga and Tabora. THB will finance about 90% of the projected lending out of its own resources (see Annex 3, para 15). Consequently, the Government and IDA contributions in this component are marginal and directed at special problems: about US$1,075,000 for importation of cement, loans for worksheds, development and construction of demonstration houses and staff training and equipment necessary to operate the lending program effectively. Small-Scale Industry 3.14 The Small Industries Development Organization (SIDO) will administer the experimental program of assistance to the informal sector in Tabora and Tanga included in the project. The program, described in detail in Annex 4, will comprise three major elements: (a) expansion of SIDO's management training and technical assistance program at the regional level; (b) financing of SIDO's hire-purchase program for production equipment and critical raw materials for industries selected under the project; and (c) infrastructure for industrial clusters (to be provided by Ardhi) and provision of THB loans for building about 150 workshops and sheds. 3.15 The basic objective of this component is to increase the economic and employment potential of the project areas and to diversify their economic base. To accomplish this, the program will build upon existing technical and marketing skills in the informal sector. The program will provide its partici- pants with (i) assistance in management, accounting, marketing and quality control in order to help them expand their productive and marketing capabil- ities; and (ii) specialization and diversification opportunities through supply of new equipment. A preliminary list of new product lines to be supported by the project has been established. 1/ Their selection was based on an analysis of local raw materials resources, technical skills of the target groups and demand potential within the region. Particular emphasis was placed on strengthening industries that would serve surrounding rural areas and on developing the building materials industry. 1/ This list will be revised on the basis of more detailed feasibility studies to be conducted by SIDO (see Annex 4, paras 6-7, and Table 1 and 2). By September 30, 1977, SIDO will prepare at least ten feasibility studies for review by IDA.  - 13 - 3.16 Grouping of existing enttepreneurs into larger cooperatives to facilitate bulk purchase of in?pts, isternaZ. opecialization and joint market- ing will be encouraged. Industries that wouid benefit from concentration of production facilities and requiring large energy and water inputs will be encouraged to relocate in clusters near the project area. The project will finance basic infrastructure for the clusters and construction of sheds and worksheds (through THB). Low-cost building technique will be encouraged to minimize the repayment burden. Technical Assistance 3.17 The project provides the following technical assistance: (a) consultants' services to Ardhi for (i) detailed engineer- ing and design for the upgrading projects; (ii) tendering and supervision of infrastructure; and (iii) on-the-job training of engineering assistants; (b) eight man-years of management support services to Ardhi including financial management (Financial Analyst), pre- paration of management manual (Management Specialist), community development (Community Development Specialist), training (Engineering Instructor) and monitoring and evaluation (Researcher); (c) five man-years to SIDO for project implementation, management training and preparation of operating manual (Project Technical Adviser); and marketing and subproject preparation (Specialist); (d) three man-months of management support for THB to assist in developing procedures for implementing the materials lending program (Procurement Consultant); (e) eighteen man-months of consultants' services for con- ducting a study of the Land Rent and Service Charge; and (f) six man-years of consultants' services for the preparation of future projects. 3.18 To the extent possible, the use of local consultants from within Eastern Africa will be encouraged. At least 30% of the technical assistance needs are expected to be met through local consultants. However, provision is made in the project to cover potential foreign exchange costs of expatriate consultants. 3.19 Annex 8 contains a detailed breakdown of the technical assistance included in the project. During negotiations assurances were obtained that consultants to be employed would be acceptable to IDA and would be retained on terms and conditions satisfactory to it.  - 14 - D. Implementing Schedule 3.20 Overall project implementation will take about four years (see Chart 1). Detailed design and engineering will begin in the third quarter of 1977. Upgrading works are expected to be completed by mid-1979. The target for the surveyed plot program would be reached by mid-1980. All other construction components fall within this timetable. THB's lending program and project monitoring, both on-going activities, will not require IDA dis- bursements after December 1981. Accordingly, the project's closing date will be December 31, 1981. A Project Completion Report will be submitted to IDA within three months of the closing date (see Annex 10).  - 15 - IV. PROJECT COSTS, FINANCING AND COST RECOVERY A. Cost Estimates 4.01 The project cost is estimated at T Sh 243 million (US$29.3 mil- lion), of which about 36% are foreign exchange costs. The cost estimates are detailed in Annex 6 and summarized in the table on the next page. 4.02 All costs are expressed in mid-1977 prices. Cost estimates for the infrastructure component are based on preliminary engineering designs by consultants (Messrs. J. Burrows and Partners) and include a location allowance for upcountry civil works. 1/ Other costs are appraisal mission estimates based on data provided by the agencies concerned. Physical con- tingencies of 15% on civil works have been included for infrastructure, community facilities and demonstration houses. In accordance with Bank guidelines, price escalation for civil works is based on estimated inflation rates of 10% in 1976 and 9% in 1977, 1978, and 1979. Price escalation for equipment and services is based on estimated inflation rates of 8% in 1976 and 7.5% in 1977, 1978, and 1979. 4.03 Equipment and materials imported directly for the project are assumed exempt from custom duties and taxes. However, some equipment and materials will be purchased in the local market and hence would be taxed. The import duties and sales taxes for locally purchased materials and equipment are estimated at T Sh 5 million (US$0.6 million) or 2% of the total project cost. B. Financing 4.04 Of total estimate project costs, about 40% would be financed with proceeds from the proposed IDA credit of US$12 million to cover 100% of the foreign exchange costs as well as approximately 5% of local costs. The bal- ance of T Sh 143.4 million (US$17.3 million) would be financed by Government, of which T Sh 39 million (US$4.7 million) will be from its normal budgetary 1/ To account for the additional transportation costs and difficult working conditions compared to Dar es Salaam (see Annex 6).  - 16 - resources and T Sh 104.5 million (US$12.6 million) from THB's own resources. The financing plan is summarized in the table below. ---------------------US$ Thousand-------------------- Estimated Government Cost ment % THB % IDA % 1. Compensation 723 723 100.0 - - - 2 Infrastructure 7,615 2,284 30.0 - - 5,331 70.0 3. Community Facilities 1,125 338 30.0 - - 787 70.0 4. Technical Assistance and Consultants 1,593 160 10.0 - - 1,433 90.0 5. Vehicles and Equipment 414 41 10.0 - - 373 90.0 6. House Improvement and Construction Loans (THB) 12,311 - - 11,236 91.0 1,075 9.0 7. Small Scale Industry Development (SIDO) 1,445 145 10.0 - - 1,300 90.0 8. Contingencies 4,054 1,003 25.0 1,350 33.0 1,701 42.0 TOTAL 29,280 4,694 16.0 12,586 43.2 12,000 41.0  - 17 - PROJECT COST ESTDIATES T Sh '000 Us$ '000 Foreign Exchange Local Foreign Total Local Foreign Total as % of Total 1. COMPENSATION 6,ooo - 6,0oo 723 - 723 0 2. INFRASTRUCTURE a. Roads and Surface Water Drainage A 20,295 16,605 36,900 2,445 2,001 4,446 45 b. Water Supply /1 12,226 8,151 20,377 1,473 982 2,455 40 c. Electricity /1 602 1,806 2,408 72 218 290 75 d. Surveyed Plots Infrastructure 1,612 1,074 2,686 195 129 324 40 e. Demonstration Houses 664 166 830 80 20 100 20 Subtotal 35,399 27,802 63,201 4,265 3,350 7,615 44 3. COMMUNITY FACILITIES a. Schools 3,922 1,307 5,229 473 157 630 25 b. Mother and Child Health Centers (MChs) 1,343 447 1,790 162 54 216 25 c. Markets 664 166 830 80 20 100 20 d. Furniture and Equipment 1,115 371 1,486 134 45 179 25 Subtotal 7,044 2,291 9,335 849 276 1,125 25 4. TECHNICAL ASSISTANCE AND CONSULTANTS a. Detailed Engineering 1,644 4,936 6,580 198 595 793 75 b. Urban Finance Study 75 675 790 9 81 90 c. Management Assistance 756 1,134 1,890 91 137 228 to d. Training 250 750 1,000 30 90 120 75 e. Project Preparation 1,500 1,500 3,000 181 181 362 50 Subtotal 4,225 8,995 13,220 509 1,086 1,593 68 5. VEHICLES AND EQUIPMENT /2 a. Ministry of Lands - 640 640 - 77 77 100 b. Surveyed Plots 57 510 567 7 61 68 90 c. Sanitary Vehicles - 1,400 1,400 - 169 169 100 d. Maintenance 623 207 830 75 25 100 25 Subtotal 680 2,757 3,437 82 332 414 80 6. HOUSE I24PROVEMENT AND CONSTRUCTION LOANS (THB) a. Development and Promotion 310 310 620 38 37 75 50 b. Materials Loans 74,100 24,700 98,800 8,927 2,976 11,903 25 c. Worksheds Loans 2,070 690 2,750 250 83 333 25 Subtotal 76,480 25,700 102,180 9,215 3,096 12,311 25 7. SMALL SCALE INDUSTRY DEVELOPMENT (SIDO) a. Management Assistance 1,142 2,218 3,360 138 267 405 66 b. Hire-Purchase Program 1,173 6,647 7,820 141 801 942 85 c. Infrastructure 454 356 810 55 43 98 44 Subtotal 2,769 9,221 11,990 334 1,111 1,445 77 TOTAL 132,597 76,766 209,363 15,977 9,249 25,226 36 8. CONTINGENCIES a. Physical /3 6,358 4,419 10,777 766 532 1,298 41 b. Price /4 15326 7.5 22,874 1,847 909 2,756 33 GRAND TOTAL 154,281 88,733 243.016 18,590 10,690 29 280 36 A Including location allowance at 30% of the basic infrastructure cost for Tabora, 25% for Iringa, 12.5% for Tanga and 7.5% for Morogoro. /2 Excluding THB and SIDO. 3 15% on Items 2 (with location allowance), 3abc, and 7c. , 12% on Items 2abc, 6bc, 7c; 14% on 3abcd; and 9% on 2d, 7b (after physical contingencies).  - 18 - 4.05 The Borrower will be the Government of Tanzania which will pass on the proceeds of the proposed IDA credit to Ardhi, THB and SIDO for their respective project components through arrangements acceptable to IDA. For Ardhi, Treasury will channel funds for the project through Government's normal budgetary procedures as in the first project. Ardhi in turn will make funds available to agencies other than THB and SIDO involved in implementation of sub-components (see para 5.01). The arrangements for channeling funds were finalized during negotiations. For THB and SIDO the terms will be as follows: (a) THB. The funds for purchase of imported cement and workshop loans will be on-lent to THB at 3% interest over 20 years including a five year grace period. THB's terms to bene- ficiaries of materials loans for housing will be at 6% interest up to 20 years with a 1 year grace period, in ac- cordance with current terms for such loans. Workshop loans will be at 8% over 15 years with up to 2 years grace period. Staff development and training expenditures will be made available to THB as a Government contribution. (b) SIDO. The funds for the hire purchase program will be on-lent to SIDO at 3% interest over ten years including a three year grace period. SIDO will recover the funds from project beneficiaries at 8% interest over eight years with up to two years grace period. SIDO's current rate of interest for the program is 5%. The proposed increase will bring the rate in line with the interest rate struc- ture of the National Bank of Commerce (NBC), the main source of working capital credit to small-scale industry. This increase in interest rate was confirmed during negotiations. Funds for management training and assis- tance will be provided to SIDO through usual budgetary procedures, while funds for infrastructure for industrial clusters will be channeled through Ardhi. 4.06 THB will use its own funds to finance the balance of the lending program for house improvement and construction totalling (before contingen- cies) approximately T Sh 93.4 million (US$11.2 million) over five years (see Annex 3). Incremental credit needs for working capital of the small-scale industries, estimated at T Sh 4 million, will be covered by NBC through its normal lending procedures. During negotiations, assurances were obtained from THB and Government that these funds would be made available to project beneficiaries. 4.07 Incremental recurrent expenditures for governmental agencies at the central and regional level (excluding THB and SIDO) are estimated at T Sh. 3.5 million (US$0.4 million) annually. During negotiations, assurances were obtained that these expenditures would be financed through normal budget- ary channels.  - 19 - C. Land Rent and Service Charge 4.08 Experience in the first project demonstrated the infeasibility of cost recovery arrangements in project areas which differ from those in other urban areas (para 2.20). Charges for other areas as applied previously did not meet IDA criteria, which call for consistent fiscal policies to recover the economic cost of providing and maintaining the infrastructure. The Tanzania Government does not have a clearly articulated policy on cost re- covery in urban areas. Until 1974 legislation concerning charges in urban areas was governed by five different ordinances. In 1974 the newly enacted Land Rent and Service Charge Act repealed previous legislation and established a single, consolidated charge for all urban plotholders. The basis for levy- ing charges was to be the economic value of land as assessed by Ardhi. In theory the economic value of land should reflect the full costs of capital infrastructure investment and hence it should be possible to set charges at a level sufficient to recover full costs. In practice, however, this is impos- sible under the present circumstances: valuation rolls used by Ardhi are outdated and new land valuation is seriously behind schedule; the Land Rent and Service Charge Act itself is unclear on many of the issues involved; and the collection machinery is not effective. The Government recognizes these problems and the broader need to rationalize its fiscal policy toward urban land, shelter and infrastructure. It has already engaged a team of consult- ants to carry out an in-depth study of the Land Rent and Service Charge (para 2.14 and 2.21). 4.09 The purpose of the study is twofold: (a) to analyze the existing situation regarding the assessment and collection of the Land Rent and Service Charge; and (b) to propose specific measures to increase its effectiveness, yield and equity. Terms of reference for the study were agreed upon by the Bank and Government during appraisal (see Annex 7, Attachment 1). The study is expected to result in reforms that would introduce valuation methods which relate the level of charges to urban infrastructure investments and maintenance expenditures, and improve valuation, collection and enforcement procedures. The long-term goal of these reforms is to increase the revenues from the Land Rent and Service Charge to a level sufficient to cover invest- ment and operating expenditures for urban infrastructure. 4.10 Detailed cost recovery arrangements for the project will be for- mulated in conjunction with the study )J. During negotiations the inception report for the study was reviewed and agreement reached on short-term measures, guiding principles for the long-term reform of the Land Rent and Service Charge and arrangements for the cost recovery in the project. As a short-term measure, Government agreed to progressively introduce the Land Rent and Service Charge in all squatter areas in Tanzania. This introduction will be linked to the program of house registration (see para 5.09). The charge will be set initially at T Sh 100, which is 33% higher than the current level of charges in areas that have not been revalued since 1969. 1/ The study commenced on February 14, 1977.  - 20 - 4.11 Specific recommendations of the study and the agreement between IDA and the Government on cost recovery arrangements for urban infrastructure expenditure in the project will be guided by three main principles: (a) self-sufficiency of the urban sector; (b) close relationship between expenditure on and revenues from urban land development and service infrastructure; and (c) simplicity of the Land Rent and Service Charge Structure. Recommendations of the Study will be reviewed and agreement between IDA and Government on level and composition of plot (house) charges as well as revision and collection procedures for the project areas reached by December 31, 1977. D. Accounts and Auditing 4.12 Procedures established in the first project have been revised to overcome auditing problems. Separate project accounts would be maintained for all civil works by Ardhi. The Management Section of the Sites and Services Division will be responsible for maintaining accurate records and preparing annual statements. Ardhi's Internal Auditor who reports directly to the Principal Secretary will monitor project accounts. For the THB and SIDO components the two agencies will also maintain separate accounts for all project expenditures in their respective components. Assurances were obtained during negotiations that annually audited records of project accounts would be submitted to IDA no later than six months following the close of the fiscal years of Ardhi, THB and SIDO. E. Procurement and Disbursement 4.13 Infrastructure works totalling about T Sh 59.7 million (US$7.2 mil- lion) before contingencies will be procured through international competitive bidding in accordance with Bank/IDA Guidelines for Procurement. While the relatively small size, scattered locations and staging of the contracts make it unlikely that foreign firms will be attracted, a number of established contractors in Eastern Africa should be interested in bidding. Tenders will be advertised in Eastern Africa and embassies of Bank member countries and Switzerland in Dar es Salaam notified. There will be prequalifications of firms so that the Government can assess the firms' experience, financial standing and performance on similar construction in Tanzania. Local contracts will be accorded a preferential margin of 7.5% over foreign contractors in the evaluation of bids. 4.14 Contracts for construction of community facilities totalling T Sh 7.8 million (US$0.95 million) before contingencies will be awarded on the basis  - 21 - of competitive bidding advertised locally in accordance with standard pro- cedures of the Ministry of Works and satisfactory to IDA. Efforts to attract small contractors from project towns will be made by grouping contracts to be within their range. 4.15 Equipment/vehicles totalling T Sh 16.6 million (US$2 million) will be procured in accordance with Government's standard procedures which were reviewed at negotiations. Cement totalling T Sh 5.4 million (US$0.65 million) will be procured by international competitive bidding in accordance with Bank guidelines. Other building materials totalling T Sh 93.4 million (US$11.2 million) will be procured by local competitive bidding following procedures satisfactory to IDA. SIDO will follow its usual procedures which were reviewed at negotiations for procuring hire-purchase equipment totalling T Sh 7.8 million (US$0.9 million). 4.16 Disbursement of funds from the credit would be made on the following basis: (a) 70% of expenditures for civil works against payment certificates endorsed by consultants and Ardhi; (b) 100% of foreign expenditures if directly imported or 75% of expenditures for locally purchased furniture, equipment materials and vehicles; and (c) 100% of foreign exchange expenditures and 75% of local expenditures for consultant services. 4.17 In view of the urgency and importance of the Land Rent and Service Charge Study, Government requested approval to appoint consultants to begin the study in February 1977. Approval of retroactive financing is recommended for consultants' fees paid prior to credit signing. Retroactive financing is also recommended to cover consultants' services for detailed design and engineering from May 1, 1977. A total sum of up to US$200,000 would be required to cover expenditures incurred for these two items from February 1 to the date of signing.  - 22 - V. ORGANIZATION AND MANAGEMENT 5.01 Responsibility for project execution will be shared between Ardhi, the principal implementing agency, and the decentralized administrations in each project town. The specialized components involving power supply, housing loans and small industry will be handled by Tanzania Electric Supply Company (TANESCO), THB and SIDO, respectively. Because of Tanzania's severe manpower constraints, adjustments in central and local project management to overcome problems experienced in the first project will emphasize more effec- tive use of existing personnel rather than new recruitment. A. Ministry of Lands, Housing and Urban Development (Ardhi) 5.02 Ardhi's Sites and Services Unit will have operational responsibility for the planning and implementation of the project, as it did for the first project. The unit will be restructured to strengthen its project management and community development capabilities and its status within the Ministry upgraded. Government has already approved the proposed reorganization which will become effective at the beginning of the fiscal year, July 1, 1977. The project includes T Sh 1.5 million (US$0.2 million) for technical assistance to Ardhi for the development of financial and management systems and community development capability. 5.03 The physical planning, detailed engineering and supervision of con- struction contracts for the infrastructure and community facilities in the project areas will be the responsibility of the unit's Planning and Engineer- ing Sections with assistance from consultants. The community' facilities will be planned and designed in consultation with appropriate ministries and regional officials. Consultants were hired to prepare preliminary engineering designs, and detailed engineering and some supervision will also be handled by consultants. As in the first project, the engineering consultancy contracts will supplement on-the-job engineering training through secondment of trainees to work with the consultants. The capabilities developed through the training program will enable the Engineering Section to assume responsibilities, such as the engineering for community facilities and the supervision of contractors at some sites, handled by consultants in the first project. 5.04 The Sites and Services Unit will be responsible for the overall scheduling and coordination of the surveyed plot program. The Planning Section will liaise with zonal and regional planning officials to select sites and prepare layouts. The section will prepare detailed site selection and planning guidelines as well as model layouts and as much of the planning work as local staffing permits will be done locally. The surveys will be carried out by an Ardhi mobile survey crew based in Dar es Salaam. As part of the reorganization of the Sites and Services Unit, the crew will be transferred to the Surveys and Mapping Division, which is better able to provide the manage- rial and professional support it requires.  - 23 - 5.05 Management and community development capabilities relatively ne- glected during the first project will be expanded under the proposed project. Financial management will be strengthened by recruitment of a Financial Analyst to be financed by project technical assistance funds. In order to familiarize local administrators with the process of project implementation, the unit will prepare a management handbook and conduct an orientation seminar in Dar es Salaam for local officials. Community development and land officers in the unit will be responsible for the non-physical aspects of the project, including publicity, allocations, registration of squatter houses and pre- paration of titles, on-site assistance to residents in project areas, and the formation of housing cooperatives. The unit will prepare guidelines and information materials for the social components of the project and work with local officials during their implementation. B. Regional and District Administrations Local Project Management 5.06 Teams of local officials under the direction of Project Coordinators designated by the Regional Development Directors (RDD) will be responsible for local implementation. The Coordinator's responsibilities will be to plan local project implementation with assistance from the Sites and Services Division, to coordinate the activities of the various local officials involved in the projects, and to monitor progress and submit periodic reports to Ardhi. One Site Officer (a junior official with technical skills in construction) will be appointed in each project town to assist the Coordinator and to insure a continuous administrative presence in the project communities. Other local officials such as the relevant Land Officers and Engineers will be responsible for their normal administrative functions in the project areas. 5.07 Considerable attention will be given to explaining the project to community residents and to providing on-site assistance. The Site Officers will work closely with the new ward councils mandated by the Urban Wards (Administration) Act, 1976. Demonstration houses built in the project areas will be used as offices by both THB and the Site Officers as well as to demonstrate low-cost building materials and techniques. Transportation will be provided for households displaced by the improvement works to help them relocate. 5.08 The regional and district administrations will assume responsibility for the recurrent operations of the community facilities and the maintenance of infrastructure. Budgetary allocations for these services will begin in fiscal 1978/79. Road maintenance has been a problem in Tanzanian cities due to budgetary constraints. An experiment in community maintenance of roads and drains using manual equipment to be supplied through the project will be conducted in a few project sites to be selected by Ardhi. The equipment will be given to the local engineers who will loan it to the ward councils and provide technical supervision.  - 24 - Allocations and Tenure 5.09 In order to provide security of tenure in the upgrading areas, titles based on house rather than plot occupation will be introduced. The title form was reviewed during negotiations and revisions were agreed. The titles will be issued for five year periods and renewed automatically on expiration. They will also specify the level of Land Rent and Service Charge, which can be revised with six months advance notice. The registration of existing houses in the squatter communities will be the responsibility of the District Land Development Officers (DLDO's). Maps based on the most recent aerial photographs available showing the location of existing units will be prepared by the Sites and Services Division. The DLDO's will update these maps and work with the Site Officers and local CCM leaders to identify the house-owners. 5.10 The DLDO's will also be responsible for soliciting and processing applications for the surveyed plots. A points system taking into account household income, number of dependents, employment status and length of time on the waiting list will be used to rank applicants. The system will be weighted to give preference to households with monthly incomes below T Sh 1,000 (US$120). However, because of the Government's policy of encouraging mixed-income communities and the scarcity of other housing opportunities, there will be no income eligibility limit. At negotiations the allocation system was reviewed and agreed. The existing District Allocation Committees consisting of administrators, elected representatives and CCM officials will select the allottees after review of ranked application files prepared by the Land Offices. Allottees will be issued long-term leases (normally for 33 years) which can be mortgaged and serve as security for THB loans. C. Tanzania Housing Bank (THB) 5.11 The Tanzania Housing Bank (THB) will administer the house improve- ment and construction lending component. As in the first project, the lending program will be run by the Sites and Services Section of THB's Operations Department. The day-to-day administration of the program will be handled by regional offices in Tabora, Tanga, Morogoro and Iringa. 1/ Expansion of THB's lending operations in squatter areas and establishment of the on-site materials depots will necessitate increases in THB staff at headquarters and in the regions. The addition of a Procurement Officer to manage the materials opera- tions will be particularly important. A Procurement Consultant will be appointed to design and oversee the establishment of material loans operations (see Annex 3, Attachment 1, for the terms of reference). THB has agreed to hire the additional staff necessary to implement the program. Project funds will be provided for short-term training for the staff involved in the program. 1/ The THB Sites and Services Section will assume the day-to-day management of Dar es Salaam program.  - 25 - 5.12 For the loans in the surveyed plot areas, THB will apply its normal lending procedures. For loans in the squatter upgrading areas, lending procedures will be patterned after THB's rural lending program. The major differences between the two types of loan are: I/ (a) loans in squatter areas do not have to be secured by mortgages; (b) THB does not charge valuation fees for such loans; and (c) loans in these areas can be made for smaller amounts for partial home improvement. In order to reach the target population more effectively, part-time site offices will be established in project areas for distribution and reception of loan applications. Depots for storage and distribution of materials will be built in Dar es Salaam, Tabora and Morogoro. In Iringa and Tanga, storage space will be leased. THB will offer standard house-type plans (see Annex 3, Figure 3) to prospective borrowers and build demonstration houses in the upgrading areas. D. Small Industries Development Organization (SIDO) 5.13 The small-scale industry (SSI) program will be implemented by SIDO, the parastatal responsible for assisting SSIs in Tanzania. Established in 1973 under the Ministry of Commerce and Industry, SIDO has set up an extension service in each region, usually staffed by a Promotion Officer, an Economist and an Engineer. SIDO operates a hire-purchase program for equipment and raw materials, eight production-cum-training centers and an industrial estate. 5.14 SIDO will hire a local Project Implementation and Training Coordina- tor and a Project Economist for each project town. The Coordinator will be responsible for selecting and providing assistance to participants in the hire-purchase program and for coordinating and supervising the industrial cluster and network schemes. Cooperatives, partnerships, sole artisans and District Development Corporations would be eligible for SIDO support. The Coordinator will also organize and implement apprenticeships, exchanges, courses and individual training in accounting, technical skills, market organ- ization and cooperative organization. The Project Economist will prepare feasibility studies and financial packages and insure links between project participants and lending institutions--NBC (working capital), THB (sheds) and SIDO's hire-purchase program (equipment)--facilitating the processing and repayment of credit. 1/ See Annex 3.  - 26 - 5.15 Consultants, probably from outside Tanzania, will assist in develop- ing management training programs, initiating the industrial clusters and net- works and, if necessary, assisting in subproject preparation. Two Project Technical Advisers will spend up to two years in Tanga and Tabora assisting in the initiation of the clusters and networks and development of training programs. The consultants will also develop an operating manual for this kind of program which could then be applied in other regions. Additional consulting assistance will be made available for subproject preparation, market organiza- tion and financial management as needed.  - 27 - VI. PROJECT JUSTIFICATION AND ECONOMIC BENEFITS 6.01 The project is expected to generate a wide range of direct benefits to more than 315,000 low-income urban dwellers, some 26% of the total urban population of Tanzania. Provision of essential infrastructure, regulariza- tion of tenure and supply of credit for house improvement aim at improving living conditions for some 160,000 inhabitants of squatter settlements. Additional schools, market and health facilities will raise the standards of their community facilities. The project will also meet the demand for new shelter by providing 19,000 new surveyed plots. Institutional and Sectoral Objectives 6.02 In addition to direct benefits to a significant proportion of low-income urban dwellers of Tanzania, the proposed project will also con- tribute to the development of more effective urban shelter policies and to strengthening public institutions responsible for their implementation. The project will further demonstrate the feasibility and replicability of a low-cost approach to upgrading of squatter settlements. The surveyed plot program of the project will introduce and test a novel response to the demand for new plots, which combines minimal infrastructure investment with a planned residential framework for building low-cost shelter. If successful, the program will reduce the proliferation of squatter settlements and allow more rational planning and provision of urban services. A broader approach will also be followed in the implementation of the health community facilities. In addition to the provision of facilities, guidelines will be developed for an urban HICH program, similar to the existing MICH program for rural areas, with a strong emphasis on community extension work focussed on preventive medicine and nutrition education. 6.03 Particular attention in project implementation will be paid to the broader fiscal and administrative contexts of urban policies. The project will help define a program to reform the Land Rent and Service Charge, the principal urban land use and shelter tax (para 4.07-4.10). In connection with the reform, a dialogue with Government on other aspects of the present system for financing urban investment and maintenance and of the administra- tion of urban areas will be pursued. The technical assistance component of the project has been designed to strengthen existing institutions responsible for different aspects of urban policies at both central and regional levels. Institution-building assistance will also be provided to SIDO and THB. Squatter Upgrading and Surveyed Plot Program 6.04 A cost-benefit analysis based on comparison of "with project" and "without project" situations has been used to evaluate the economic rate of return for squatter upgrading and surveyed plots program. 1/ The analysis 1/ The shadow prices used in the analysis were derived from a number of key primary factors, the most important of which are (i) tradable goods; (ii) taxes and Government revenues; (iii) labor, skilled, semi-skilled, unskilled urban and unskilled rural; and (iv) non-traded goods including foodstuffs.  - 28 - takes into account only those benefits that are easily quantifiable in monetary terms such as higher rental income from better quality housing. However, the project will also generate less tangible external benefits such as better health, lower infant mortality, and a higher level of education. The quantification of these benefits is generally difficult due to the lack of adequate comparative data on the levels of health, education and related variables. Therefore, the analysis based on higher rental income provides only a partial indication of the economic return on the project. 6.05 Costs. Project costs used in the economic analysis included all construction, equipment and maintenance costs of infrastructure for squatter upgrading and surveyed plots net of taxes and duties. Physical contingencies, administration and detailed engineering were also included as well as house improvement, new construction and resulting maintenance expenditures incurred by project beneficiaries. The cost of schools, clinics and health programs were omitted (see para 6.04 above). 6.06 Benefits. Project benefits were measured by the increase in total rent payments generated by the project. Rents may increase because existing houses are improved, new ones are built or higher rents are paid for the same accommodation as a result of improvement of urban services and security of tenure. Values for differentials in the levels of new construction, house improvements and rents between "with-the-project" and "without project" sit- uations were based on survey evidence collected by Ardhi. 6.07 Internal rates of return for the squatter upgrading and surveyed plot program are as follows: Dar es Salaam 24% Tabora 19% Tanga 22% Morogoro 22% Iringa 28% All towns 23% All towns 20% (squatter up- grading only) All towns 35% (surveyed plots only)  - 29 - 6.08 Sensitivity Tests. The results of the evaluation were relatively insensitive to changes in assumptions used in the analysis. In the evalua- tion of squatter upgrading and surveyed plots, the critical parameters were the differential in rent levels and housebuilding and improvement rates between improved and unimproved areas. The total value of benefits would have to decrease 60% for the internal rate of return to fall below 10%. Small-Scale Industry Development 6.09 Urban unemployment and limited industrial activity are major prob- lems of Tanzanian towns which the proposed SSI component would address. The component would bring direct and tangible benefits by creating about 870 new jobs, including 350 skilled and semi-skilled workers at a unit in- vestment cost of US$650 (US$1,000 if working capital needs and technical assistance are included). The potential annual output of the 150 new pro- duction units is valued at about US$2.7 million. The component would also demonstrate a feasible and low-cost approach to assisting the urban informal sector and strengthen SIDO's financial, management and technical assistance programs, especially at regional levels. Risks 6.10 Risks involved are primarily institutional and financial. The project's scattered locations combined with a general shortage of qualified personnel contribute to the institutional risks. Specific provisions for strengthening the management capability of the institutions implementing the projects and for improving coordination between them are made in the project to limit this kind of risk. Financial risk is associated with the reform of the Land Rent and Service Charge. While the Government expressed its willingness to undertake the reform, its complete implementation may prove difficult in the context of continuing budgetary constraints and scarcity of qualified personnel. Flexibility, careful timing and continuing monitoring will be required. However, even partial implementation of the reform will represent a significant improvement over the existing situation. Financial risk is also involved in THB lending program in the squatter areas, given its novelty and lending criteria and procedures less stringent than those normally employed by THB. During negotiations, assurances were obtained from the Government that the Treasury would closely monitor the progress of the program and, if required, take the necessary steps to protect THB's financial integrity.  - 30 - VII. AGREEMENTS AND RECOMMENDATIONS 7.01 The following agreements were reached with the Government during negotiations: (a) Land Rent and Service Charge. Assurances were obtained that the Land Rent and Services Charge Study will be completed in accord- ance with the inception report and an implementing plan including short-term measures agreed. Agreement on final cost recovery arrangements for the project will be reached by December 31, 1977 (para 4.10-4.11); (b) Allocation and Tenure (i) the legal title form to be used in squatter upgrading areas was agreed (para 5.09); (ii) assurances were obtained regarding the collection of Land Rent and Service Charge in squatter areas (para 4.10 and 5.09); and (iii) the allocation system to be used for the surveyed plot program giving preference to the target population was agreed upon (para 5.10). (c) Ardhi. An approved plan and details for the reorganization of Ardhi's Sites and Services Section were submitted at negotiations (para 5.02); (d) Community Facilities. Assurances were obtained that the guidelines for the urban MCH program will be prepared by the Ministry of Health by December 31, 1977 (para 3.09); (e) House Improvement and Construction Loans Program (i) assurances were obtained that THB will appoint the additional staff required to implement its lending program by September 30, 1977 (para 5.11); (ii) THB confirmed Government approval for importation of cement for the materials loans (para 3.11); (iii) THB submitted approved proposals for meeting its share of project costs for the lending program (para 4.06); and (iv) interest rates and terms of lending to beneficiaries proposed were agreed (para 4.05).  - 31 - (f) Small-Scale Industry Development (i) assurances were obtained from SIDO on the timely appointment of additional staff and consultants re- quired (para 5.14); (ii) procedures and criteria for selecting SSI project participants were agreed with eligibility open to cooperatives, partnerships and individuals in Tanga or Tabora (Annex 4); (iii) ten feasibility studies will be completed and sub- mitted for IDA review by September 30, 1977 (para. 3.15); and (iv) interest rates and terms of lending to beneficiaries proposed were agreed (para. 4.05). (g) Recurrent Expenditures. Government agreed to provide all recurrent costs for components financed under the project (para 4.07); (h) Financial Records and Auditing. Government agreed to main- tain separate records and accounts for the project (including THB and SIDO components) which will be audited (following procedures acceptable to IDA) annually and submitted to IDA within six months of the close of the financial years of Ardhi, THB and SIDO (para 4.12); and (i) Project Completion Report. Government agreed to submit a completion report within three months of the project's closing date (para. 3.20). 7.02 Subject to the above assurances and agreements, the project is suitable for an IDA credit of US$12 million.   ANNEX 1 Page 1 TANZANIA SECOND NATIONAL SITES AND SERVICES PROJECT Squatter Settlements and Project Towns 1. This Annex describes the characteristics of the housing and residents of squatter communities in two project towns and provides a brief sketch of each project town. Its final section summarizes the analysis used to determine the demand for surveyed plots in each town. A. Characteristics of Squatter Housing and Residents 2. Detailed studies of housing characteristics in project communities are available for Dar es Salaam and Tabora (Table 4). The characteristics of squatter housing in the two towns are quite similar, and analogous studies in the other three towns are likely to produce comparable results. Local variations in typical building materials are discussed in the profiles of the project towns. 3. In both cities the model squatter dwelling had four rooms and housed four to eight people, although over a quarter had ten or more residents. Squatter houses often follow the traditional Swahili house plan, illustrated in Figure 1, which is characterized by a front verandah, a central corridor and four to six rooms.. The rooms often have external access and, therefore, can be rented easily. Monthly room rents in Dar es Salaam tended to fall in the T Sh 20-29 range; those in Tabora were somewhat lower. 4. Physically, the model squatter house type is a mud and pole structure with a corrugated iron roof; roughly half have cement floors. Two-thirds of the houses in Dar es Salaam had no foundations, compared to 39% in Tabora (a high water table in one of the Tabora settlements may encourage residents to install foundations and floors). Pit latrines were by far the most common form of sanitation in both cities. The more detailed information available for the Dar es Salaam settlements indicates that 54% of the houses had open pits or pits with only temporary structures. 16% of the houses in Tabora and 4% of those in Dar es Salaam had piped water on the plot. In Tabora, where a more detailed question was asked, almost 20% of the houses had no access to local standpipes, and another 45% were more than 100m from a standpipe. Data about effective plot size coverage and use wer! also collected in Tabora. Seventy percent of the plots were 200 to 300 m ; 81% of the informants felt that their plots were large enough. Slightly more than half of the plots had 50% to 74% coverage, with the majority having around 50%. Almost 90% of the plots had gardens and 9% had businesses. 5. A large survey of Dar es Salaam residents in 1971 permits comparison of the characteristics of people living in squatter settlements and planned areas of the city (Table 5). The survey indicated that 69% of the squatters were migrants, compared to 66% of the non-squatters. There were also no major  ANNEX 1 Page 2 differences between the groups in the reasons given for coming to Dar es Salaam. Residents in squatter areas were somewhat more likely to be home- owners, 17% claiming to own a home in Dar es Salaam compared to 7% of those living in planned areas of the city. 6. Residents in the planned areas tended to have more education and higher incomes than the squatters. Half of the squatter respondents had no formal education, compared to just over a quarter of the non-squatters. At the other end of the educational spectrum, only 5% of the squatters had reached Form I or higher, compared to 23% of the non-squatters. Income is closely related to education, and the squatters had lower incomes. A third of them had incomes below T Sh 200 (US$24) a month, while only 15% of the other respondents had comparably low incomes. At the other end of the income distribution, 41% of the non-squatters had monthly incomes of T Sh 400 (US$48) or more, compared to only 15% of the squatters. Educational attain- ments were associated with income among both squatters and non-squatters; however, even when education was statistically controlled, the non-squatters had somewhat higher incomes. Self-employment and casual labor were somewhat more common among residents of squatter areas. Unemployment rates were comparable for the two groups, with about 20% of each group indicating that they were unemployed or seeking work. 7. These comparisons indicate that differences between the residents of squatter settlements and planned residential areas in Dar es Salaam are relatively small. Far from being marginal areas populated by economically marginal residents, the squatter settlements are an integral part of the city. Other squatter settlements chosen for upgrading in the proposed project are comparably integrated in the socio-economic structure of their cities. B. Project Town Profiles 8. Maps IBRD 12573-7 show land-use patterns and the locations of the project sites in each town. The criteria used to select the sites and the chracteristics of each site are discussed in Annex 6. Dar es Salaam 9. Dar es Salaam is by far the largest city in Tanzania, more than six times larger than the second biggest city, and accounts for 42% of the national urban population. As the colonial capital and principal port, Dar es Salaam was the administrative and commercial hub of the Tanganyika Colony. During the past decade the Government has endeavored to decentral- ize both the economy and the administration. The regionalization of the Tanzanian Government during the early 1970's led to the dispersion of over 40% of the Government's central establishment from Dar es Salaam, a trend which will continue with the development of Dodoma as the new national capital. Economic decentralization has been far less successful, and the completion of the TAZARA railway line from Dar es Salaam to Zambia has re- inforced Dar es Salaam's importance as a transportation center.  ANNEX 1 Page 3 10. Dar es Salaam's 1977 population is estimated to be 630,000, based on an average growth rate of 8.5% a year during the past decade. At its current estimated annual growth rate of 7%, the city will reach one million inhabitants by 1985. Approximately two-thirds of the population now lives in squatter housing. During 1963-73, a period for which reliable data are available, squatter housing expanded at an average rate of 16.6% a year. Mianzese and Mtoni-Tandika, upgrading areas in the first and second projects respectively, are the largest squatter settlements in the country. (The characteristics of the housing in Manzese are described in para 3-4 and Table 4). Although Dar es Salaam provides over 40% of the country's urban wage employment, unemployment and low incomes are major problems for its residents. For example, a recent survey in Manzese suggested that 10% of the household heads and much larger proportions of other household members were unemployed. A Bureau of Statistics budget survey in Manzese and Mtoni- Tandika during 1975-76 indicated that almost a quarter of the households had incomes below the minimum wage (TSH380 or US$46 a month). Tanga 11. For the past two decades Tanga has experienced relative stagnation compared to other towns and is being eclipsed by Arusha as Tanzania's second largest city. Prior to World War II, Tanga prospered as a port serving the rich northern agricultural region and its settler population. Its last spurt of growth occurred during 1952-57 when the population nearly doubled, largely because of rises in world market prices for sisal. Since then, however, its economy has expanded sluggishly and the population has grown more slowly than in other principal cities. Efforts to divert port traffic to Tanga from Dar es Salaam may provide some economic stimulus in the future. 12. In 1976, Tanga's population was estimated to be 90,000 based on a stable growth rate of 5% a year. The town's economic problems are indicated by the fact that wage labor grew at an average rate of only 1% a year between 1962 and 1972. The Tanga Master Plan 1975-1995 pointed to unemployment and underemployment as severe problems. A survey conducted in 1974 as part of the master planning exercise indicated that the bottom 50% of the households had monthly incomes of T Sh 300 (US$36) or less. According to budget surveys conducted by the Bureau of Statistics during 1975-76, over half of the house- holds in selected low-income communities had incomes below the minimum wage. Housing conditions in Tanga generally appear to be worse than in other project cities. The Master Plan categorized 45% of the housing as in "bad" condition in 1974. 1/ Squatter housing has been expanding rapidly particularly along the main southern access route. Ardhi estimates that 40% of the population currently lives in squatter housing. The prevalence of thatch roofs which catch fire frequently suggests that the availability of materials loans for corrugated iron sheets should be stressed in the upgrading program in Tanga. 1/ Ardhi's Urban Planning Department uses a standard house rating system. The lowest rating is assigned to units "built of temporary or semi- temporary materials and those incapable of rehabilitation." Master Plan for Morogoro (1974), p. 6.  ANNEX 1 Page 4 Morogoro 13. Morogoro, located 180 km southwest of Dar es Salaam, has promising growth prospects. It is an important transportation center, situated on the Tanzam Highway, the Eastern African Railways central line and the road from Dar es Salaam to Dodoma. Morogoro was selected as a site for a Bank- financed industrial estates project which will provide 3,000 new jobs. It is also an educational center with four secondary schools, a teacher training college, the National Institute of Development Management, and the agricultural faculty of the University of Dar es Salaam. 14. Morogoro's 1977 population is estimated at 45,000 and the town is growing at an estimated rate of 6% a year. Table 6 indicates that Morogoro has the highest wage employment per capita and, proportionately, the largest manufacturing sector among the project towns. However, sisal and tobacco processing industries account for much of this emp'loyment, and average wage earnings per job are below the national urban average. More than 40% of the population lives in squatter settlements. The Master Plan for Morogoro prepared in 1974 classified 80% of the housing in Msamvu, a project upgrad- ing site, as "poor", but the proportion of "good" housing in the town's other squatter settlements (36%) was comparable to the rest of the town. The potential for house improvement in Morogoro's squatter settlements thus appears high. Iringa 15. Iringa is located about 400 km southwest of Dar es Salaam on the Tanzam Highway. It is a commercial and trading center for one of the most fertile agricultural areas in the country. Although it is strategically located for the development of agricultural processing industries, the city currently does not have any major industries, and the tertiary sector is proportionately more important than in the other cities (Table 6). A short- age of piped water is an impediment to economic expansion as well as a poten- tial health hazard. 16. The city's population of 56,000 is estimated to be expanding at 8% a year. During 1967-75, planned areas grew at an average annual rate of 6% compared to 14% for the squatter areas; the proportion of squatters in the total population rose from 26% to 49%. Soil conditions are appropriate and firewood is available for burnt bricks in Iringa. Therefore, much of the housing in the squatter settlements is structurally very sound. The Bureau of Statistics' pilot household budget survey in 1974 included a small sample (59) in Kihesa, one of the communities to be upgraded. Approximately 45% of the households sampled had incomes below the minimum wage. Tabora 17. Tabora, 520 km west of Dar es Salaam, was an important Arab trading town which was developed as an administrative center by the Germans. The East African Railways line from Dar es Salaam and Dodoma branches in Tabora and the EAR has a servicing yard there. Tabora is also a national educational center, with six secondary schools, teacher training and secretarial colleges, a SIDO  ANNEX 1 Page 5 cottage industry school, and EAR vocational training school. However, the town lacks major manufacturing and processing industries, and its narrow economic base poses a major problem for its future growth. 18. Tabora's estimated 1977 population is 69,000; it is growing at the comparatively low rate of 5% a year. Data presented in the Master Plan for Tabora indicate that in 1967-72, the population of squatter areas grew at 6% a year, compared to 3% per annum in other areas. Ardhi estimates that about 70% of the population now lives in squatter areas. The characteristics of the housing in two squatter settlements are described in paras 3-4 and Table 4. The regional administration undertook an improvement program affecting 500 houses in one squatter settlement on its own in 1972. The gravity of the employment situation in Tabora is indicated by Table 6; among the project towns Tabora had the least employment per capita and proportionately the smallest secondary sector (manufacturing and construction). Average incomes in Tabora are substantially below the national urban sector average. C. Size of the Surveyed Plot Program 19. The analysis used to determine the number of plots to be surveyed in the project is summarized in Table 7. Detailed data about population characteristics, household formation and effective demand for housing in Tanzania's cities are not available. Therefore, a simple approach based primarily on projected population growth was adopted. The population and growth rate estimates presented in Table I were used as the starting point for the analysis. It was assumed that a population equivalent to the increase due to natural growth would be accommodated in existing housing. The estimated growth rates were therefore deflated to remove the impact of natural increase, estimated at 1.8% a year based on analysis of data from the last census. The resulting rates were then used to project the number of people who will need new housing during the project implementation period, 1977-1981. The result- ing figures were reduced by 25% to reflect infilling and extension of houses in existing squatter communities and the very limited availability of other housing opportunities for better-off households (e.g., NHC and THB houses and medium-and low-density plots prepared by the regions). 20. The estimated net demand for new housing was then divided by the average number of people per plot in low-incomes communities to determine the required number of plots. Although Ardhi uses seven people per plot as a planning guideline, nine was chosen as a more realistic figure reflecting prevailing conditions in existing high-density settlements. The estimated total need for surveyed plots for all five towns is 18,977. 1/ It should 1/ This is 19% fewer plots than Ardhi originally estimated would be needed. Ardhi's projections took into account the number of households on the district allocation waiting lists as well as population growth over a three year period. The mission chose not to take the waiting lists into account because they are not consistently kept up to date.  ANNEX 1 Page 6 be noted that this estimate does not take into account the backlog of demand due to the small size of the surveyed plot program in the past. Because of the unreliability of the data and parameters on which the estimates are based, the phasing of allocations of surveyed plots will be planned to begin slowly and to increase in relation to the volume of applications received. If the application rate is lower than anticipated, the allocations can be phased over a longer period.  ANNEX 1 Table 1 TANZANIA SECOND NATIONAL SITES AND SERVICES PROJECT Growth of Urban Centers in Mainland Tanzania Urban Centers Growth 1/ Estimated Estimated 4/ (With Populations of Rate 1967 2/ Current 1975 5,000 or More in 1967) 1948-67 Population Growth Rate 3/ Population Dar es Salaam 7.5% 272,821 7% 550,000* Tanga 5.9% 61,058 5% 89,000* Mwanza 6.1% 34,961 n.a. 60,000 Arusha 10.0% 32,451 n.a. 89,000* Moshi 6.5% 26,864 n.a. 69,000* Morogoro 6.1% 25,262 6% 40,000* Dodoma 4.9% 23,559 n.a. 50,000 Iringa 7.3% 21,746 8% 56,000* Kigoma 7.3% 21,369 n.a. 30,000 Tabora 2.7% 21,012 5% 62,000* Mtwara 2.7% 20,413 n.a. 30,000 Musoma 9.1% 15,412 n.a. 30,000 Lindi 2.4% 13,352 n.a. 20,000 ,beya 7.5% 12,479 n.a. 50,000 Singida n.a. 9,478 n.a. 20,000 Wete n.a. 8,469 n.a. n.a. Bukoba 5.0% 8,141 n.a. 10,000 Mwadui n.a. 7,383 n.a. n.a. Korogwe n.a. 6,675 n.a. n.a. Songea n.a. 5,430 n.a. 10,000 Shinyanga n.a. 5,135 n.a. 10,000 Bagamoyo n.a. 5,112 n.a. n.a. Total 6.8% 658,483 7% 1/ Claes-Fredrik Claeson and Bertil Egero, Migration and the Urban Population, University of Dar es Salaam, Bureau of Resource Assess- ment and Land Use Planning, 1972. 2/ 1967 Population Census, Volume 2, Statistics for Urban Areas, Dar es Salaam, 1970. Tables la and lb. 3/ Mission estimates. 4/ Ministry of Lands, Housing and Urban Development or mission (*) estimates.  ANNEX 1 Table 2 TANZANIA SECOND NATIONAL SITES AND SERVICES PROJECT Squatter Housing in Urban Centers of Tanzania Estimated 2/ Estimated 1/ Existing Population Squatter Squatter Houses per City/Town 1975 Houses 1,000 Population Dar es Salaam 550,000 40,000 73 Tanga 89,000 3,000 34 Arusha 89,000 6,000 67 Moshi 69,000 2,000 29 Tabora 62,000 4,000 65 Mwanza 60,000 3,000 50 Iringa 56,000 3,000 54 Mbeya 50,000 5,000 100 Morogoro 40,000 1,500 38 Mtwara 30,000 500 17 Musoma 30,000 500 17 Kigoma 30,000 1,000 33 Bukoba 10,000 1,000 100 1/ Table 1. 2/ Ministry of Lands, Housing and Urban Development.  ANNEX 1 Table 3 TANZANIA SECOND NATIONAL SITES AND SERVICES PROJECT IDA - Financed Sites and Services and Squatter Upgrading Projects First Project Second Project Serviced Houses in Up- Surveyed Houses in Up- Cities Plots grading Areas Plots grading Areas Dar es Salaam 6,182 7,600 14,150 9,138 Mwanza 1,900 - - - Mbeya 850 1,200 - - Iringa - - 1,770 1,088 Morogoro - - 810 2,069 Tabora - - 925 2,784 Tanga - - 1,330 732 TOTAL 8,932 8,800 18,985 15,811  ANNEX I Table 4 TANZANIA SECOND NATIOIIAL SITES AND SERVICES PROJECT Characteristics of Squatter Housing in Dar es Salaam and Tabora Dar es Dar es Characteristic Tabora Salaam Characteristic Tabora Salaam Number of Rooms Floor I - 2 9% 11% Cement 61% 50% 3 - 4 60% 59% Earth 37% 50% 5 - 6 19% 28% Cow Dung Plaster 2% - 7+ 12% 2% Walls Number of Occupants Grass and Poles 1% 1- 5 27% 35% Mud and Poles 68% 67% 6 - 9 46% 30% Mud Bricks 28% - 10 - 13 14%) Soil-Cement 2% 2% 14 - 17 11%) 35% Sand-Cement - 30% 18+ 2%) Concrete Blocks 1% - Other - 1% Number of Rooms Rented Roof 0 42% 44% 1 6% 6% Corrugated Iron Sheets 94% 83% 2 15% 16% Beaten Tin 5% 10% 3 15% 11% Grass 1% 7% 4 13% 15% 5+ 7% 10% Toilet Average Rent per Room (T Sh per Month) W.C. 4% 2% Pit Latrine 92% 1 - 15 1 - 14 28% - Perm. Structure 28% 15 - 19 ) 16% Temp. Structure 23% 20 - 24 ) 53% 33% Open Pit 31% 16 - 30 25 - 29 ) 32% Dry System 1% 30 - 34 ) 11% None 3% 14% 31 - 45 35 - 39) 15% 4% 40 - 49 ) 3% Water 46+ 4% - Tap on Plot 16% 4% Foundation Tap < 100 m) 20% Tap > 100 m) 45% 96% None 39% 66% None ) 18% Stones ) 39%) Bricks ) Yes 15%) 34% Electricity Concrete) 8%) On Plot 33% 3% None 67% 97% Source: Ardhi Sites and Services and RLDO. Tabora.  ANNEX 1 Table 5 TANZANIA SECOND NATIONAL SITES AND SERVICES PROJECT Characteristics of Households in Squatter and Non-Squatter Areas, Dar es Salaam, 1971 Respondents in: Respondents in: Squatter Non-Squatter Squatter Non-Squatter Areas Areas Areas Areas Reasons Cited for Migrating 1/ to Dar es Salaam Nigration Status Employment-Related Migrant 69% 66% To look for work 24% 19% Non-Nigrant 31% 34% To take job obtained prior to arrival 2% 4% House Ownership Government transfer 2% 3% Employer Transfer 1% 2% House Owner 17% 7% 29Z 28% Non-Owner 83% 93% Accompany family Employment Status To accompany parents 17% 16% To accompany spouse 23% 20% Casual/"kibarua" 9% 5% 40% 36% Regular Wage 26% 19% Salaried Staff 21% 38% To Attend School 3% 6% Senior Management 0% 2% Self-Employed/ Miscellaneous Non-Wage 21% 16% To visit 10% 9% Unemployed or Other 5% 4% seeking work 22% 20% Do not know/not stated 14% 17% 29% 30% Monthly Income (T Sh) 2/ Level of Education Under 100 10% 4% 100- 199 24% 11% None 50% 28% 200- 299 34% 26% Standards I-IV 21% 18% 300- 399 17% 18% Standards V-VIII 23% 31% 400- 499 5% 8% Forms I-IV 5% 21% 500- 749 6% 13% Forms V-VI 0% 1% 750- 999 1% 4% University 0% 1% 1000-1499 1% 7% 1500 and over 2% 9% 1/ More than one response possible. 2/ Includes wage and non-wage income. Data Source: National Urban Mobility, Employment and Income Survey. Sample Size: 1,873 in squatter areas; 1,417 in non-squatter areas. Source: Richard E. Stren, Urban Inequality and Housing Policy in Tanzania (Berkeley: Institute of International Studies, University of California, 1975), Tables 11, 13, 15, 16, 19, and 23.  ANNEX 1 Table 6 TANZANIA SECOND NATIONAL SITES AND SERVICES PROJECT Wage Employment in Project Towns, 1974 Dar es Salaam Iringa Morogoro Tabora Tanga Total Employment 96,053 7,960 11,131 8,179 13,332 Employment per 1,000 Population 187 154 295 139 147 Sectoral Distribution of Employment Agriculture 4% 5% 5% 5% 0% Manufacturing 23% 5% 30% 16% 24% Public Utilities 3% 6% 3% 12% 5% Construction 11% 20% 16% 7% 15% Commerce 8% 13% 3% 3% 12% Transport 25% 14% 18% 32% 21% Finance 5% 2% 1% 1% 1% Services 21% 35% 24% 24% 22% Source: Bureau of Statistics.  ANNEX 1 Table 7 TANZANIA SECOND NATIONAL SITES AND SERVICES PROJECT Surveyed Plot Requirements Dar es Salaam Tanga Tabora Morogoro Iringa 1. Estimated 1976 Population 1/ 588,500 93,450 65,100 42,400 60,480 2. Estimated Annual Growth Rate 1/ 7% 5% 5% 6% 8% 3. Estimated Growth Rate 2/ Due to Migration 5.2% 3.2% 3.2% 4.2% 6.2% 4. Population to be Housed 1977-81 169,772 15,940 11,104 9,684 21,222 5. .75 x #4 127,329 11,955 8,328 7,263 15,917 6. Average Number of People Per Plot 3/ 9 9 9 9 9 7. Number of Plots Required 1976-81 14,148 1,328 925 807 1,769 Sources: 1/ Table 1. 2/ Demographic analysis of the 1967 Census indicated that the rate of natural increase in urban areas was 1.8% per annum. This rate has been applied to all five towns. Claes-Fredrik Claeson and Bertil Egero, Migration and the Urban Population (Bureau of Resource Assessment and Land Use Planning, University of Dar es Salaam, 1972), p. 20. 3/ Estimate based on Tanga Master Plan, 1975-95 and housing surveys in squat- ter communities in Dar es Salaam.  ANNEX 1 TANZANIA Figure 1 SECOND NATIONAL SITES AND SERVICES PROJECT A TYPICAL TRADITIONAL HOUSE W.C. BATH SR KITCHEN _______ Front Elevation 18.29 Plot Boundary ROOM ROOM X 4.16x3.35 3.66x3.35 3 ROOM ROOM 3.66x3.35 ; 4.16x3.35 3 t l I n ,////////////////I C/// 0 CL < l I tsw t o /Z/v///////zoo v 20 -t I I z m o Xo//z////v//v/s X0 I I | L_lc | X zov///vovowoI - ROOM 6 ROOM 44.16x3.35..I 3.66 '/ '/9.14'.-'-",/ 2.73 1 7 1 VERANDAH \_1 Entrance Plot Boundary L _ ____ .-----. "-II STREET Ground Plan Site Plan Scale: 1: 100 Scale: 1: 200 World Bank-16745  ANNEX 2 TANZANIA SECOND NATIONAL SITES AND SERVICES PROJECT Community Facilities Program INTRODUCTION 1. To raise the level of community services in each site to national standards, the project will provide primary schools, Mother and Child Health (MCH) Clinics, and markets. The number and location of these facilities is summarized in Table 1. This Annex reviews the sectoral context, first project experience and the proposed approach in the second project for each category of community facility. SCHOOLS 2. Background. In 1974, the Tanzanian Government established an ambitious program of universal primary education aiming at 100 percent enroll- ment by 1977. Despite considerable progress, it has since become apparent that achievement of this target would entail severe pressure on the national budget and take a longer time. 1/ 3. Due to the severe shortage of funds, school construction in the regions has been falling behind schedule. The cost of classrooms constructed by contractor is currently estimated at T Sh 20,000 for upcountry towns and T Sh 40,000 in Dar es Salaam. However, on the assumption self-help is avail- able, Government allocates each region amounts sufficient to buy only building materials. Over the past two years, this has been cut back from T Sh 8,000 per classroom to T Sh 5,000 for upcountry towns, and from T Sh 20,000 to T Sh 5,000 in Dar es Salaam. Local officials have repeatedly commented on the difficulty of obtaining self-help labor in urban areas. As a result, funds to finance classrooms are transferred from other budgets and/or the funds pooled and fewer classrooms built. In the Dar es Salaam region, for example, only 74 of the 101 classrooms scheduled to be constructed in 1975/1976 were actually completed. In the town of Morogoro, seven of the ten classrooms planned were constructed. Typically, a one stream (seven years) school is opened with one classroom for the entering Standard I pupils. A class room is added each year to allow the intake of new students as the first year students advance. The 1/ For detailed information see "Fiscal Implications of Universal Primary Education and Universal Rural Water Supply", report of World Bank Special Mission, March 1976. The achievement of UPE would require almost doubling the number of classrooms in order to accommodate the increase of school population from 433,000 in 1975/1976 to 841,000 in 1977.  ANNEX 2 Page 2 school is _orirleted at the end of six years. Building standards are low: concrete floors, unpainted cement block or burnt brick walls, corrugated iron roofing, no ceilings, and screens for windows. 4. To date, urban schools have had no shortage of staff and sufficient funds for the teachers' salaries have been provided by the Government. How- ever, severe shortages exist for supplies. In 1975 the Ministry of Education recommended an annual average expenditure of T Sh 80 per primary school pupil for supplies. In fact, the Dar es Salaam region received only about T Sh 15 per pupil in 1975-76. 1/ 5. First Project Experience. Phase I primary schools were originally designed as complete "Community Education Centers", with six classrooms, teachers' offices, kitchen, dining hall, craft workshops and dispensary. Their cost is presently estimated at T Sh 1.2 million per school. In light of the financial constraints, the proposed standards are unrealistically high. Furthermore, the provision of workshops and technical workshops attached to primary schools is still an experimental program, aimed at a small number of schools, funded by Danish aid and therefore not generally applicable at this time. 6. Second Project. Phase II schools will basically provide classroom space. As shown in Table 8, Annex 6, a one stream school will include six classrooms, 2/ a small administration block, and a toilet block. Building standards will be as low as possible: concrete floors, burnt brick or cement block walls, corrugated iron roofing, and screens for windows. Ceilings will be included if funds allow. Space will be provided on the site for additional streams and workshops that could be constructed in the future. The cost estimate of T Sh 43,000 (US$5,180) per classroom in Dar es Salaam is based on the 1975 contract price of a classroom updated to mid-1977 level. For upcountry towns, the cost of a classroom is estimated at T Sh 33,000 (US$3,976). 3/ The project will also provide T Sh 5,500 (US$663) per classroom for furniture and T Sh 40 per pupil to partially cover initial expenditure on books and supplies. Anticipated recurrent staffing and supply expenses are presented in Table 4. 7. To determine the number of schools to be provided (see Table 2), the school-age population and existing facilities in and adjacent to the 1/ The Dar region received T Sh 944,000 for supplies in 1974-75, 850,000 in 1975-76, and is expected to receive only 835,000 in 1976-77. Due to inflation and the increase in enrollment both current and real per pupil expenditures have decreased substantially. 2/ As Standard I and II operate on double shifts, only six classrooms are needed for the seven standards. 3/ On the basis of information supplied by regional education officers, it is assumed that a 25 percent saving can be achieved in the construction of upcountry classrooms with contracts tendered locally.  ANNEX 2 Page 3 project areas were taken into consideration (see Table 2). Most of the existing schools are not completed because of the staged expansion described in para 3. It was assumed, however, that the regions will find the resources, as they have in the past, to add classrooms each year and gradually equalize the number of streams per standard. 8. The Bank supports the goal of UPE, although it remains concerned with the overall costs of the program and would prefer to see the target reached over a longer time span. On the basis of a one stream school for 315 children or approximately 225 families and in conjunction with planned regional efforts, the project would help to meet the demand for new class- rooms required for close to 100% enrollment of the current school-age popu- lation in project areas by 1981. HEALTH AND NUTRITION 9. Background. Tanzania's health problems are severe. In 1967, infant mortality was at 165/1,000, with life expectancy at birth only 41 years. Most sickness results from poverty-linked infectious and parasitic diseases, such as malaria, measles, and gastroenteritis, and from various anemias and protein malnutrition. These diseases affect both urban and rural areas. 10. To combat these problems, Tanzania has been committed to a rapid extension of its health care system, particularly over the past six years. Foreign assistance has been crucial in this expansion; 90% of the 1975/76 health development budget of T Sh 97 million was supplied by outside donors, up from 70% the year before. For fiscal year 1977/76, the recurrent public sector expenditures, financed from internal sources, were about T Sh 400 million (US$48 million), an expenditure of T Sh 27 per capita. At 2.4% of GNP, this represents a stronger public sector commitment to health care than is shown by other countries of similar per capita incomes. The 1975/80 Five Year Plan calls for public health care to expand to 4.5% of GNP. Such an expansion, however, would generate further pressure on the public budget. In view of present budgetary constraints its achievement appears unlikely. 11. Most of the health sector investment is going to the rural areas. By 1980, the Five Year Plan calls for one 4-room rural dispensary for every 8,000 people and a larger rural health center for every 50,000. The emphasis is placed on preventive care to the two groups in greatest need--pregnant women and children under the age of five. One of the rooms of the rural dispensary serves as a Mother and Child Health (MCH) Clinic, staffed by a specially trained MCH aide and providing basic pre-natal and post-natal examinations, innoculations, medicine, nutrition education and food supple- ments if necessary. 12. The program has been well established in rural areas, following Ministry of Health guidelines. With the cooperation of USAID and UNICEF,  ANNEX 2 Page 4 twenty centers to train MCH aides have been set up which will be supplying 450 MCH aides per year by 1980. The standards for the program have been kept low: a simple MCH aide kit is distributed to all facilities and the aides are given only a basic 18-month course. 13. Urban Health and First Project Experience. Urban populations have traditionally had greater access to medical care than rural families, with most cities served by a district or regional hospital and outpatient clinics or dispensaries. However, the availability of care is unevenly distributed, with facilities adjacent to squatter areas severely overcrowded. 1/ Designed to serve 20,000-30,000 people, urban dispensaries are usually large (about 400 square meters), with high physical standards. Curative treatment is emphasized, although some preventative care, such as innocula- tions, is given. These facilities have high recurrent costs, due particularly to the large expense for drugs and other supplies. 2/ Recurrent funding has been inadequate and severe shortage of simple equipment, such as sterilizers, clamps and beds, often prevents the staff from carrying out simple operations which the facilities are designed to handle. As a result, many patients must still be referred to the hospitals. 14. Designs of health facilities in the first project, based on model Health Ministry plans, conformed to the traditional pattern of urban medicine and emphasized curative care and high construction standards. 15. Second Project Program. The second project will extend the emphasis on preventive medicine from rural to urban areas by financing the construction and/or expansion of MCH clinics. The program should help hold down recurrent expenses, relieve the overcrowding of existing facilities, and improve the access to health care for pregnant mothers and small children in squatter areas. 16. In selecting the location and type of MCI facility, complete clinic or extension, existing medical facilities were taken into account (see Table 3). Although a few clinics already operate in some towns, no specific guidelines have been established to indicate the changes required for the large-scale implementation of the MCH program in urban areas. The Ministry of Health has agreed to prepare such guidelines which will specify the serv- ices to be offered, the number of patients to be served, organization of the facility, staffing, equipment, layout and estimated capital and recurrent 1/ Dispensaries near the project sites in Dar es Salaam were designed to handle 500 patients a day; they currently handle 1,000-1,500 per day. 2/ In the current Five Year Plan, it is projected that 40% of a clinic's total expenses, including salaries, should be allocated for supplies. However, in two Dar es Salaam dispensaries, supplies equal 60-70% of total costs.  ANNEX 2 Page 5 costs. Emphasis will be placed on the efficient utilization of staff, reducing physical standards, and limiting recurrent expenses. 17. An urban MCH should serve about 30,000 people and accommodate a daily flow of 250-350 mothers and children. Table 8 of Annex 6 pre- sents a tentative list of functions and specifications for a complete MCH, based on a well-functioning clinic in Dar es Salaam. A type design for the clinics will be included in the guidelines. To lower capital costs, two construction standards should be used: a low standard, similar to that used for school classrooms, for the waiting and registration areas; and a higher standard for medical facilities, i.e., the rooms where actual treatment or injections are given. Recurrent costs and staffing level are given in Table 4. 18. The guidelines for urban MCH program will also include a community extension program focussing on nutrition education, and which should become one of the standard functions of the clinic. The current MCH nutrition education program offers short lessons on nutrition and other subjects while the women are waiting for consultation; however, mothers' attendance is only sporadic. There is little effort to follow up on children identified in the clinic as malnourished. Mothers are encouraged to return to receive further guidance, but home visits are rarely made. In an attempt to over- come these problems, nutrition and MCH aides based at the clinic will be assigned to work in adjacent communities, visiting houses of particularly malnourished families and providing regular courses in family hygiene, nutrition and environmental sanitation at community centers. UNICEF guidance and assistance will be sought in improving the nutrition education program. 19. The program should begin with at least two extension workers assigned to each clinic, with the possibility that more may be added as the country meets its health manpower requirements. The project will finance bicycles, motorbikes, and other equipment necessary for starting the program. The guidelines for the extension service will be prepared with the assistance of the local UNICEF office, which has agreed to finance a field work orientation seminar providing MCH aides. The exten- sion workers will be placed under the supervision of the MCH clinic director. MARKETS 20. Design, construction and management of markets is the responsibility of the regions. Most towns have one or two large markets and a few smaller neighborhood markets. These smaller markets may have some high standard, formally constructed stalls, but are primarily made of informal wooden stalls, built by the vendors themselves and frequently strung out along the roadside. It has been difficult for the regions to raise funds to improve these markets, as all revenues collected from stall rents are channeled to the central Treasury.  ANNEX 2 Page 6 21. In the first project, markets were designed to provide some higher quality, more hygienic stalls, particularly for meat and fish selling. It is anticipated they will serve a population of 15-20,000. Standards for the second project will be similar. As outlined in Table 8, Annex 6, the proposed markets would include eight vegetable stalls, built to low standards of unfinished concrete block walls and no water, and four meat stalls, built to higher standards of fully enclosed space, washable concrete surfaces, and piped water. A porch in front of the stalls and toilet block are included. Recurrent expenses for the market are shown in Table 4.  ANNEX 2 Table 1 TANZANIA SECOND NATIONAL SITES AND SERVICES PROJECT Community Facilities Program No. of Medical Facilities Number of Schools MCH Extension/ Number of Markets city Site 1 Stream 2 Stream MCH Clinic Outpost New Extension Tabora Isebya 1 1 1 Kiloleni 1 1 Tanga Gofu Juu 1 1 I Mwakizaro 1 Morogoro Kichangani 1 1 1 Msamvu 1 Iringa Kihesa 1 Dar es Salaam Mtoni 2 1 1 Tandika 2 1 1 Hanna Nassif 1 1 Total 6 6 2 4 4 2  TANZANIA SECOND NATIONAL SITES AND SERVICES PROJECT Community Facilities - Projected School Enrollments and Classroom Demand Estimated School Streams Required 2/or Existing Streams Streams Provided City Site Age Population 100% Enrollment - (completed or begun) In Project Dar es Salaam Hanna Nassif 1,590 5 4 1 Mtoni 7,000 21 8 (11) 3/ 4 Tandika 4,200 14 0 (15) 4 Morogoro Msamvu 1,620 5 0 (5) 1 Kichangani 630 2 1 1 Tanga Gofu Juu 590 2 1 1 Mwakizaro 560 2 0 (1) 1 Tabora Isebya 2,380 8 5 3 Kiloleni 1,680 5 2 (1) 2 Iringa Kihesa 1,540 5 6 0 18 1/ 20% of the estimated total population in the project areas. 2/ Based on 315 students per one stream school, 45 students per standard for seven standards. 3/ Figures in parenthesis represent school streams in adjacent areas that children in project sites attend. tD >4  ANNEX 2 Table 3 TANZANIA SECOND NATIONAL SITES AND SERVICES PROJECT Existing and Planned Medical Facilities Existing Project to Provide Dar es Salaam Hanna Nassif 1 dispensary, small MCH near Extension of MCH Mtoni ( Both sites, plus the planned MCH 1/ ( area of Temeke, served by Tandika ( 1 MCH, and 1 dispensary MCH Morogoro Msamvu Downtown MCH Kichangani No services Medical Outpost (small dispensary/MCH) Tanga Gofu Juu No'dispensary for west side MCH Extension, with of city, existing small MCH some dispensary ser- vices Mwakizaro Dispensary nearby Tabora Isebya Existing 2-room MCH Expansion of MCH Kiloleni Nearby dispensary now being constructed Iringa Kihesa 2 dispensaries 1/ Will also serve overspill area of Mbagala - 1,000 houses.  ANNEX 2 Table 4 TANZANIA SECOND NATIONAL SITES AND SERVICES PROJECT Community Facilities - Staffing and Recurrent Expenses (in TSh) Salary/Month Facility Category Per Position Total Per Year School 1. Personnel 1 Principal 1,500 18,000 6 Teachers 700 50,400 2 Watchmen, custodians 500 12,000 Subtotal 80,400 2. Supplies 1/ 12,600 Total 93,000 MCH Clinics 1. Personnel 1 Staff Nurse 2,500 30,000 2 Grade A 1,500 36,000 5 Grade B/Midwife 800 48,000 3 Grade C 600 21,600 2 Nutrition Extension Officers 800 19,200 Subtotal 154,800 2. Supplies 200,000 Total 354,800 Markets 1 Watchman 600 7,200 1 Custodian 600 7,200 Total 14,400 1/ TSh 40 per pupil.  ANNEX 3 TANZANIA SECOND NATIONAL SITES AND SERVICES PROJECT House Improvement and Construction Loans Program 1. Under this project, house improvement and construction loans to project beneficiaries will be made by the Tanzania Housing Bank (THB). This will be an extension of the program established under the first project with modifications designed to address problems encountered. This annex discusses recent developments in the organization and management of THB, 1/ reviews the experience of the first project and describes the proposed lending program for the second project. A. THB--Recent Developments 2. Since 1974 THB has embarked upon an ambitious program of regional- ization. It currently has ten Regional Branches as well as a Zonal Office in Moshi (covering Kilimanjaro, Arusha and Tanga regions). In the near future THB plans to establish branches in eight remaining regions of the mainland, where Regional Cooperative Unions presently act as THB agents. In early 1976 the heads of Regional Branches were promoted to Manager levels. However, all lending operations still have to be approved by the Head Office. THB Manage- ment is currently considering an experimental delegation of lending responsi- bility to the Zonal Office in Moshi. 3. The THB staff has grown rapidly from 100 persons in mid-1974 to 270 in mid-1976. As a parastatal, THB offers better promotion and career develop- ment possibilities than most other government employment and is therefore able to attract qualified personnel. However, difficulties are experienced in two areas: accounting and architecture. These are due to the general shortage of qualified Tanzanians and the length of formal training in those fields. Currently ten members of THB staff are being trained locally and abroad. 2/ 4. THB has diversified its operations. It administers the Workers' and Farmers' Housing Development Fund, created in mid-1974 and financed by a monthly levy of 2% of the wage bill of every employer of ten or more persons. The Fund is likely to become one of the main sources of 1/ For a detailed description of THB's functions, purpose, organiza- tion and management see the appraisal report for the first project (Report 337-TA). 2/ One architect, six accountants, one insurance specialist, one banking specialist and one statistician.  ANNEX 3 Page 2 THB's financing for low cost housing in both urban and rural areas. On the other end of its lending activities spectrum, THB set up a subsidiary, THB Estates, which builds and sells high standard housing estates and commercial buildings. 5. The balance sheet of THB as of 31 December 1975 is shown in Table 1. Since its creation THE has considerably improved its financial situation. Between 1974 and 1975, and despite a 58% increase in its admin- istrative and overhead expenses due primarily to regionalization, profit (before taxes) increased from T Sh 0.4 million to T Sh 2.4 million. The volume of lending increased 58% to T Sh 177.6 million (US$21.4 million). The majority of those loans was made for medium-cost housing, priced at T Sh 35,000 or more. The average interest charged is 7% per annum. The lending policies of THB are sound, if anything somewhat conservative, stressing the link between the income of prospective borrowers and their housing plans. THB's building inspectors follow-up the loans in order to ensure that the loan money is not misused. The rate of default is low and the enforcement is fairly rigorous. 6. A vigorous savings mobilization campaign across the country in 1975 was very successful and resulted in a 71.5% increase in deposits over the 1974 level from T Sh 74.6 million to T Sh 127.8 million. Within those deposits the share of fixed term two-year deposits increased from 7.3% in December 1974 to 24% in December 1975. The average interest rate on public deposits is 5.4%. 1/ 7. Despite the spectacular increase in public deposits, there was still need in 1975 to find complementary sources to finance THB's lending operations. The main source was the Workers' and Farmers' Housing Develop- ment Fund which provided T Sh 31.5 million at 3.5%. THB management believes that it will be able to finance the future demand for mortgages out of its own resources (including Workers' and Farmers' Rousing Development Fund). The present spread between the average cost of money from all sources and income from lending operations -- estimated at 2% -- is judged satisfactory. There is a concern however that a greater share of fixed terms deposits which are paid a higher interest rate than the short term deposits, combined with an increased emphasis on low-cost housing loans, which carry lower interest rates than medium and high-cost housing, might result in a reduc- tion of the spread. The interest rates on deposits and loans have not been adjusted since early 1974. An upward adjustment appears desirable in the middle term, but it should take place in the context of an overall review of the interest rate structure of the banking sector. 1/ The effective rate is lower (about 5%) due to the short maturity (less than a year) of the majority of deposits.  ANNEX 3 Page 3 B. Experience of the First Project 8. It is still too early to draw definitive conclusions about the effectiveness of the lending program in the first project. There are, how- ever, indications that the program did not work as well as expected. In the site and services areas, the proportion of project beneficiaries who applied for loans was lower and the loan amounts higher than originally anticipated. Until mid 1976 no loans have been made in improvement areas. This was due mainly to the fact that in dealing with project beneficiaries, THB was apply- ing its normal lending procedures which are rather stringent and not geared toward the means and expectations of the target population. There was also a lack of coordination between THB and Ardhi. 9. A number of steps were taken during recent supervision missions to increase the effectiveness of the program. Formal coordination of THB and Ardhi programs has been established. THB has agreed to adopt a more aggressive posture in project areas. A publicity campaign advertising the availability of loans has been undertaken. Lending criteria have been relaxed somewhat and application procedures simplified. A demonstra- tion houses program was started in late 1976 and part-time site offices opened in the project areas. 10. In the second project further effort will be made to simplify lend- ing procedures and establish THB's presence in the project areas (see para 16 below). This should help to overcome the reluctance of the inhabitants who often tend to perceive THB as a remote, somewhat aloof and hardnosed institu- tion lending mostly for large projects. In addition, the problem of access to building materials encountered in the first project will be addressed more directly in the second project. 11. The problem of access to building materials has two facets. First, building materials, particularly cement, are scarce; second, the existing distribution system is inefficient and does not reach the target population on a systematic basis. Therefore, in the absence of specific measures aimed at ensuring an adequate supply of building materials to project areas, many project beneficiaries are unlikely to obtain them at reasonable prices. C. The Lending Program in the Second Project 12. Under the project, three categories of loans for house improvement and construction will be made: (a) Loans for improvement of existing houses in upgrading areas. The average size of these loans, which will be used primarily for building materials, is estimated at T Sh 4,000 and the estimated take-up rate for this category is 20% of the eligible population.  ANNEX 3 Page 4 (b) Loans for construction of new houses for households to be relocated as a result of squatter upgrading works. The average size of these loans is estimated at T Sh 14,500. The estimated take-up rate for this category is 60%. (c) Loan for construction of new houses in surveyed plot areas. The average size of these loans is estimated at T Sh 13,000. The estimated take-up rate for this category is 30%. 13. Given the novelty of lending in upgrading and surveyed plot areas, no firm data exist on the take-up rate for these categories. The proposed rates are mission estimates based on discussion with THB and Ardhi staff. The average size of loans is based on estimated costs of improvement and construction of houses (see Annex 9 for details). It was assumed that the houses in surveyed plots areas, where the level of services will be very low initially, will be of somewhat lower quality than new houses in better serviced areas. The size and volume of the lending program is summarized below: House Improvement and Construction Lending Program--1978-1982 No. of Take-up No. of loans (Amount Category Households rate to be made T Sh '000) 1. Upgrading Areas 14,293 20% 2,860 11,440 2. Relocation 1,519 60% 920 13,340 3. Surveyed Plots Area 18,977 30% 5,700 74,100 Total 34,797 9,480 98,880 14. In addition to the house improvement and construction loans, THB will also make loans for construction of worksheds to the participants in the small-scale industry programs in Tanga and Tabora. The maximum size of these loans will be T Sh 120,000 (US$14,500). The size of the workshed construction program is estimated at T Sh 2.8 million (US$0.34 million). 15. Under the first project, IDA and the Government were to supply funds to THB for the total estimated demand for loans from project beneficiaries. In the second project, IDA and Government contributions will focus on those aspects of the program that appear as likely impediments to its effective im- plementation. No major difficulties are anticipated for financing of the lending program for house construction and improvement. IDA will provide T Sh 8.9 million (US$1.075 million) to cover the foreign exchange costs for importing cement for the first three years (see para 19 below), loans for work- sheds, staff training and equipment necessary to operate the lending program effectively. THB will finance the balance before contingencies of T Sh 93.4  ANNEX 3 Page 5 million (US$11.2 million). Assuming that the lending program will be spread over five years, the average annual lending volume between 1978 and 1982 would be T Sh 18.6 million (US$2.2 million). This represents 10.6% of current THB lending. In the past, THB has been able to finance considerably larger expansion of its operations. Assurances were obtained at negotiation that THB will supply the funds for the local currency balance of the lending program. Lending Procedures 16. The lending procedures and terms for surveyed plots areas will be similar to those normally applicable to low-cost housing. Loans will be secured with a first mortgage which can be obtained only for plots held under long-term right of occupancy. Letters of intent from Ardhi indi- cating that preparation of a long-term right-of-occupancy title has been initiated are to accompany the loan application. The applicant has to show his ability to repay and is required to put down 5% of the total construc- tion cost. This sum however includes different fees to be paid in connec- tion with the loan. 17. For loans in squatter upgrading areas, THB will use procedures similar to those used in its rural lending program. These loans do not have to be secured with mortgages and fees are considerably lower. The application process for the loans will be administered from offices housed in THB demonstration houses to be constructed in the upgrading areas. The application form will have to be signed by the District and TANU officials in order to confirm the information supplied by the applicant. Rental income will be considered in determination of the applicant's ability to pay. Loan applications will be sent to THB's Sites and Services Section for final approval. 18. Procedures and lending terms (interest rates, grace period, and down payment requirements) for loans to small-scale industry program participants for workshed construction will be less stringent than those normally applied by THB in its lending for commercial operations. SIDO's Project Economist will play a crucial role as a link between program participants and THB. He will assist the participants in preparing the loan applications. His feasibility studies will be used by THB to determine applicant's ability to pay. Materials Loans 19. Borrowers will have a choice of taking their loans in cash or vouchers redeemable against specific quantities of cement, corrugated iron sheets or other building materials. These materials will be available to project borrowers at THB materials depots to be built on project sites. THEB will procure locally available materials using its own resources and maintain adequate stocks at on-site depots. Cement, however, is a scarce commodity in Tanzania; the current production of cement by the Wazo Hill factory in Dar es  ANNEX 3 Page 6 Salaam satisfies only about two-thirds of the existing demand. The shortage is unlikely to be alleviated before 1980 when two new plants under construc- tion in Tanga and Mbeya will begin production. Thus, it is expected that the incremental demand generated by the project estimated at 12,360 tons over three years will have to be covered by imports. Foreign exchange to cover the needs of the first three years are therefore included in the project. Table 2 provides an estimate of the cement requirements for the program. 20. Figure 1 presents a simplified flow chart for cement procurement. The purchase, storage and transportation arrangement will be the responsi- bility of the Procurement Officer in THB's Sites and Services Section. In order to design and oversee the implementation of the materials loan opera- tions a Procurement Consultant will be hired for three months (see Attachment I for the Terms of Reference). Bulk purchases from the Saruji Corporation, which is responsible for importing cement will be made on a periodic basis. The materials will be transported by truck or railway--whichever is cheaper-- and by hired rather than purchased vehicles. The initial distribution of materials among project towns and sites will be based on the estimates of demand shown in Table 2. Subsequent shipment will be adjusted to reflect the actual pace of lending. Depots for storage of the materials will be built in three towns where the demand is likely to be the greatest, Dar es Salaam, Tabora and Morogoro. In Tanga and Iringa storage space will be leased. The price for materials charged to borrowers will reflect the full cost of pur- chasing and transporting the materials and operating the depots. THB Staffing Requirement for Project Implementation 21. In order to implement this component THB will have to increase its staff both in its Sites and Services Section in Dar es Salaam and in the other project towns. Table 3 indicates the existing staffing pattern and the new staff necessary to carry out the program. A Procurement Officer responsible for ordering materials and arranging for their transportation will be added to the headquarters staff in addition to the existing positions. A Building Inspector and Bank Clerk will be added to the operational staff in Tabora, Tanga and Morogoro, and a Bank Clerk added in Iringa where the project lending operation will be modest. Staff to manage the materials depots in Dar es Salaam, Tabora, and Morogoro will also be needed. The estimated annual salary costs for the new staff are approximately T Sh 355,000. No serious difficul- ties are expected in finding qualified Tanzanian nationals for the new posts. As part of the loan agreement THB will undertake to hire the additional staff necessary to implement and operate the program.  ANNEX 3 Attachment I TANZANIA SECOND NATIONAL SITES AND SERVICES PROJECT Procurement Consultant - Draft Terms of Reference Job Description The Procurement Consultant will be appointed for a period of three months by THB. He (she) will be responsible for establishing the material loan operations of THB to be implemented as a part of the Second National Sites and Services Project. In particular he (she) will: (a) determine the location of material storage facilities (wholesale and retail), and the appropriate form of establishing these facilities (new construction, or renting of existing facilities), (b) review the procurement arrangements for different materials, 1/ determine specific procedures 2/ to be used in the program, and indicate the necessary legal and administrative steps to initiate the procurements, (c) review the transportation requirements of the program, determine the appropriate mode of transport (truck or railway) and determine the necessary administrative and legal arrangements to ensure an efficient and cheap transportation of materials, (d) prepare an operating manual for the material loan operation which would detail procurement, transportation, storage and disbursement procedures of the program, and (e) organize the necessary training activities for THB staff (field and headquarters) which will operate the materials loan program. The Procurement Consultant will report to the Managing Director of THB. A counterpart Procurement Officer from THB staff will be assigned to work closely with the Consultant. The Procurement Officer will be responsible for overseeing the implementation and operation of the material loan operation. Qualifications An MBA with experience in procurement and transportation planning and operations. Familiarity with building materials industry, prior experience in east Africa, preferably in Tanzania. Proven ability to communicate quickly and to devise solutions in situations of scarce resources. 1/ Cement, corrugated iron, bricks, joinery products. 2/ Consistent with Bank procurement guidelines and legal agreements between IDA and the Government regarding the project.  Table 1 TANZANIA SECOND NATIONAL SITES AND SERVICES PROJECT THB Balance Sheet as of December 31, 1975 /1 Authorried Issued & Paid 1974 Cod Acc. Net Book 1974 Depreciation Valus 7. Sb.. T. She. T. Sb.. - ~~~~~~~~ ~~~~~T. Shs. T. Shs. T. She. T. Shs. T. She. T. Sbs. T. Shs. FIXED ASSETS SHARE CAPITAL Lsnd *nd Buildings 8,247.796 972.087 7.276.708 3.733.835 100 Ordinary Shares of Shs. 1 million eXch 10000,000 28,33.178 18.720000 Pictures. Fumiture and Equipment 1.161,329 521.378 639.951 571.999 - -rSroS Imlo c 100 2837 7 Motor Vehicles 393.967 167.890 226.077 119.668 -3,803.091 1.661355 8141.736 4.425,502 RFSERVES Contribution to I.F.M. C~pital 15.225 15.225 Capital Development Programme 900.000 90.000 General 887.585 646,683 Proltt rnd Loss Account 1.278.579 240.902 MORTGAGE AND OTHER LOANS Tanzania Mainland 177.610.341 112.331.497 2.181,389. 902,810 Zanzibar 2.756,463 2.537.466 GRANTS: Government Revolving House Loans 4.196.055 5.001.171 German Government 450,.000 4,500.000 184,562.859 115.870.134 SIDA 2.0000D0 2,000.000 Less Provision for louses for Zanzibar Mortgages 2.020.309 1807.081 6.5D0000 5 650.000 182 542,55 11&0D3053 LOAN CAPITAL 193.549.550 S 8%. Commonwalth Development Corporation 3.388.725 4195,50S THB Houahn Estats z Sal ft.109,35't 21% Cnadman Commodity 14000.000 14.000000 FNxed Term end Short Temm Deposits - 5,000.000 3j-- Workers and Fa mers Housing DevelopmentFund 31.500.000 - Debtors and Prepayments 1.491,717 566.388 Treasury (World Ba-k. credit) B00.000 - Cash at Bank and in Hand 18.764.989 477.574 Government Revolving House Loan Fund 855.800 7.852,800 Tanzania Government (Unsecured) 1,013.789 1.013,789 20.256,706 6.043.962 - LESS CURRENT LIABILITIES 51.554.314 27.062.09 Creditors and Accrued charges 1,068.960 1.42328 Public Deposits 127,838,777 74.614.771 Provision for Cortoration Tax 1.110.703 210,514 Housing Deposits 4,339,023 - 2.179.663 , 1.632,842 - - -Net Current Assets 18.077.043 4.411.120 220,770.682 127.799,675 220.770.i82 127.799.675 8ibi Z. Fadiili. MIP. Director L. A. Sazia, M.P. Exc. Chairman /1 Source: THB Annual Report, 1975  ANNEX 3 Table 2 TANZANIA SECOND NATIONAL SITES AND SERVICES PROJECT Home Improvement and Construction Loans - Cement Requirements 1/ Number Estimated Total Amount 2/ Yearly Cement 2/ of Number of Cement Requirements Citv Loan Category Households of Loans (in tons) (in tons) Improvement Loans 8,113 1,625 1,950 390 Dar es Salaam) Relocation Loans 1,024 615 2,310 462 Surveyed Plots Loans 14,150 4,245 10,190 2.038 Subtotal 2,890 Improvement Loans 2,600 520 625 125 Tabora ) Relocation Loans 180 110 415 83 Surveyed Plots Loans 925 280 675 135 Subtotal 343 Improvement Loans 1,860 375 450 90 Morogoro ) Relocation Loans 215 130 490 98 ) Surveyed Plots Loans 810 245 590 118 Subtotal 306 Improvement Loans 1,060 215 260 52 Iringa ) Relocation Loans 30 20 75 15 Surveyed Plots Loans 1,770 530 1,275 255 Subtotal 322 Improvement Loans 660 135 165 33 Tanga ) Relocation Loans 70 45 170 34 Surveyed Plots Loans 1,330 400 960 192 Subtotal 259 Summary Improvement Loans 14,293 2,870 3,450 690 Relocation Loans 1,519 920 3,460 692 Surveyed Plot Loans 18,985 5.700 13,690 2.738 Total 34,797 9.490 20.6j0 4.120 1/ Figures rounded to nearest five. 2/ See attached explanatory note for details on estimation of cement requirements.  ANNEX 3 Table 2(a) Explanatory Note on Cement Estimation 1. The figures are derived by the following calculation: (a) in order to build a Swahili house of 100 m with concrete floor 3.75 tons of cement are required. This is obtained as follows: (i) floors - 100 m x .15 m = 15 m of concrete mix; assuming 1:123mix cement/sand and specific gravity of 1.6 tons/m = 2.0 tons (15 x 0.083 x 1.6); (ii) walls = (10 m x 2.52m) = 25 m - 3 m (for doo5s and windows) = 22 m x 4 = 88 m x .15 = 13.2 m; assuming mix 1:4:8 cement/sand/gravel and specific weight as in (i) = 1.75 tons (13.2 x 0.083 x 1.6); and (iii) floor and walls = 3.75 tons of cement. (b) the amount of cement required for improvement is assumed to equal 1/3 of the amount needed for new conssruction or 1.2 tons (based on the assumption th5t only 5 m of concrete will be required instead of 15 m in the new house); and (c) the amount required for surveyed plots is assumed to equal 2/3 of the amount required for the new house or 2.4 tons (it is assumed that, in the absence of services, inhabitants of surveyed plots will not proceed immediately to build high- standard houses). 2. It is assumed that the total amount of cement will be supplied over a 5 year period. Hence column (4) = column (3)/5.  ANNEX 3 Table 3 TANZANIA SECOND NATIONAL SITES AND SERVICES PROJECT House Improvement and Construction Loans - Staff Requirement Salaries for Existing New New Staff City Position Staff Staff (In T Sh/Month) Dar es Salaam Senior Operations Officer 1 1 2,500 Operation Officer 1 1 2,000 Procurement Officer - 1 2,000 Valuer 1 2 1,130 Store Manager - 1 1,260 Assistant Store Manager - 1 750 Handler/Watchmen - 4 1,800 Subtotal 11 11,440 Tabora Regional Manager 1 - - Building Inspector 1 1 1,500 Bank Clerk 1 1 1,130 Store Manager - 1 1,260 Assistant Store Manager - 1 750 Handlers/Watchmen - 4 1,800 Subtotal 8 6,440 Morogoro Regional Manager 1 - - Building Inspector 1 1 1,500 Bank Clerk 1 1 1,130 Store Manager - 1 1,260 Assistant Store Manager - 1 750 Handlers/Watchmen - 4 1,800 Subtotal 8 6,440 Tanga Regional Manager 1 - - Building Inspector 1 1 1,500 Senior Operation Clerk 1 - - Bank Clerk - 1 1,130 Subtotal 2 2,630 Iringa Regional Manager 1 - - Building Inspector 1 1 1,500 Senior Operations Clerk 1 - - Bank Clerk 1 1 1,130 Subtotal 2 2,630 Total 31 29,580  Figure 1 TANZANIA SECOND NATIONAL SITES AND SERVICES PROJECT Flow Chart for Cement StePs 1. Procurement order from THB to Saruji Co. 2. Delivery of cement by Saruji Co. To THB's Central Depot. 3. De;ivery ot cement to on-site depots in project towns. 4. Customer signs loan comnnittment and obtains cement vouchers. 5. Customer takes possession of cement. 6. Voucher are forwarded to THB's regional branch. 7. Vouchers are forwarded to THB's headquarters. 8. Adjistments ir shipment to different project towns are made. World Bank-16847  TANZANIA HOUSING BMINK HEADQUARTERS ORGANIZATION CHART PREPARED IN 1976 I EXECUTIVE CHAIRMAN GENEA MANGE OPERATIONS ADMINISTRATION TECH. & PROMOTION FINANCE | E?ARTfvM U:NT DEPARTMENT DEPARTVMENT DEPARTMENT PROi CFSS;ING l DIVISIONnGENERAL SERVICES VALUATION BANKING ECONOMIC | INSTI F:JTl5NIAL LOAN5 _ PERSONNEL INSPECTION ACCOUNTS INTERNAL AUDIT l)VISION _ DIVISION 1 DIVISION _DIVISON DIVISION PLANS & DESIGNS MORTGAGE ACCOUNTS T.HB ESTATE CO. DIVISION DIVISION | li SUi; k \l c r Dik;-un1 1 N sl - 11VIES I El. World Bank-16707  SEcarnD NATION" sin AMn suRIFIGMu PjA ALWURUAIWS UOIMSE-?WP VW* S 4 U 1 1 iSW 5T 2 0 mevSuw Bi t Lte TpV Sb 5 (2 Ri 4ul AIM 4*2 AND 402ev' AMe: 42A4_ 1T 34 a TIs SS6 1(4 _om lIp. SbS5 (4 _b _ AMe 7.U1 Agm: 7*2ev Awe: "31P A,@@: *S.1s PA - -e-7arAm li-I ei floom ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~~~~Ro5 g.Uo _1lie/ _.IV552( 3( * . 1 5 3 -~~~~W If Pe SS I tS nd K SSims I iSa J4tlG _I2 IiK40 Type SS. 1(3 RooPmsL)w PoRooTwo sih lxPd sk Am: 44 AEas 5= - Too: S i- (4 Rooms) TWp: 96-248 PAs) T 6 S-3( PRxo) Awn. 54 4ev A.: 61.2ev A', Type SS I SRPAM id Ales: 64.Pass Rusis 6~~~~~~~~~~~~~~~~~S I- 3 Dosusos isv B.Odok. ftsa then As* - B naps~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~S n.db - & B 9 No= PAM~~~Nos SloTgm wS.&mH fkxm ~ ~ ~ ~ ~ ~ ~ ~ ~~Ro Is 2 I 2W 2 VososnidS g ~ ~ ~ ~ ~ ~ ~ *IdB*-161  ANNEX 4 TANZANIA SECOND NATIONAL SITES AND SERVICES PROJECT Small-Scale Industry Program 1. The second project will include an experimental program of assis- tance to the small-scale industry sector in Tabora and Tanga. This program will be implemented by the Small Industries Development Organization (SIDO). It would be the Bank group's first venture into the urban informal sector in Tanzania and its first involvement with SIDO. The objectives of the program are two-fold: first, to directly increase production and employment in the project areas, diversifing the economic base; second, to strengthen SIDO's programs and implementation capabilities in the urban sector, particularly at the regional level. Total costs of the recommended program including loans for worksheds and cluster infrastructure will be T Sh 14.8 million or approx- imately US$1.8 million before contingencies. The program includes management training and technical assistance, funds for hire-purchase of equipment, in- frastructure for small industry clusters, and loans for construction of work- sheds. Approximately 75% of project costs will be required in foreign exchange since nearly all the equipment would have to be imported. This annex discusses the program concept and institutional arrangements, program components, expected benefits and SIDO's performance and prospects. A. Program Concept and Institutional Arrangements 2. SIDO will be the major intermediary for the small-scale industry program, building upon its present role as the government parastatal providing integrated support to small industries, including management and technical assistance through its regional staff and financial support through its equipment hire-purchase scheme. The program will focus on increasing SIDO's regional implementation and management capabilities, and developing project promotion, preparation, technical and financial assistance capability that, once refined, could later be applied in other regions throughout Tanzania. 3. The National Bank of Commerce (NBC) will not receive project finance; rather, it will continue in its present role, offering com- plementary working capital finance once SIDO has prepared, appraised, and approved a small industry as eligible under SIDO's hire-purchase scheme. The small industry clients will continue to deal almost exclusively with SIDO staff in acquiring equipment, construction, and working capital loans since NBC and THB would rely on SIDO's feasibility studies and financial packages. Since working capital funds required would be about T Sh 4 mil- lion, no special Bank-NBC agreement will be necessary; this initial par- ticipation equals less than .01% of NBC's annual lending, and NBC's head- quarters management has stated that there would be no difficulty in extending  ANNEX 4 Page 2 this amount of working capital in Tanga and Tabora nor need for a special project agreement. The project will fund appointment of a Regional SIDO Credit Officer to do the promotion and preparation work for SIDO hire-purchase, NBC working capital and (when necessary) THB construction loans; this person could be transferred to NBC at the point that a specialized field staff were organized within NPC. However, all indications from NBC Headquarters and Regional Managers are that NBC will continue to rely on SIDO to do the promo- tional, technical and supervisory work. NBC has strong interest in sim- plifying procedures for lending to small industries and would be willing to introduce new procedures as part of the project in Tabora and Tanga Towns. 4. Regional Focus. Since this project would represent the Bank group's first experience working with SIDO and on small industries in Tanzania, the small industry component will limit its scope to two project towns, Tanga and Tabora. A limited regional approach is necessary and appropriate at this stage. Decentralized government administrative structure and the poor internal communications make it necessary to improve regional programs and staff. In SIDO as in most other government agencies, the regions constitute the basic operating unit. The small-scale industry program is designed to increase SIDO's regional and implementation capacity and develop procedures which later would be applied throughout Tanzania. Experiences gained from specific training and extension methods developed and used in Tanga and Tabora will be transferred through an operating manual developed as part of the program, to be used by all the regions. Improved procedures for supervision, collections and financial controls for the hire-purchase scheme will be installed and later implemented in other regions. Low-cost workshed models will be designed by Ardhi in collaboration with SIDO for SSIs on and off the clusters in Tanga and Tabora; these low-cost models could be copied in other zones to transform the costly industrial estate approach into an econom- ically viable scheme for work space. 5. In urban areas, financial and technical assistance under the program will be available to cooperatives, associations, sole artisans, and District Development Corporations. The project would emphasize cooperative formation building upon existing technical and marketing skills in the project areas. Cooperative groups and individuals from other areas will also be eligible. The main emphasis of the program will be on the upgrading of managerial, accounting and marketing skills. The project will also provide new specialization and diversification opportunities through the supply of equipment under hire- purchase program. Grouping of artisans and productive units into larger partnerships and cooperatives that would facilitate bulk purchase of inputs, internal specialization and joint marketing will be encouraged. 6. Priority Industries. In selecting the priority subsectors, preference has been given to industries utilizing local resources; employ- ing labor-intensive technology; building upon existing technical skills  ANNEX 4 Page 3 while increasing specialization and/or market reach; developing forward and backward linkages with existing larger industries in project towns; and meeting regional demand for items traditionally imported into the region. Particular emphasis was placed on industries that would enhance urban-rural links and seLve surrounding rural areas and on the development of the building materials industry. 7. The detailed list of selected industries is shown in Tables 1 and 2 of this Annex. Selection was done in cooperation with SIDO's staff and is based on an analysis of the regional potential of Tanga and Tabora. By September 30, 1977, SIDO will prepare at least ten feasibility studies for review by IDA. These feasibility studies will give special emphasis to market potential and channels, raw material sourcing, criteria for selection of participants, and organization aspects. Consultants support for additional project preparation is available under technical assistance (see para 29 below). 8. Some of the priority industries will be located in clusters near the project areas; in Tanga the designated small industrial zone is between the two project areas of Gofu Juu and Mwakizaro, while in Tabora the small industry area is in the project area of Kiloleni (see para 18-20 below). Other selected small industries will be scattered throughout the wards of each town, but organized into cooperatives or joint ventures, forming networks with common facilities for storage of raw materials and finished products as well as for specialized machinery which individual small industry cooperatives cannot afford. 9. The priority industries provisionally selected for Tanga will: (a) use waste or by-products from the parastatals (e.g. shorts from the steel rolling mill for nuts, bolts, hinges, and structural steel; gypsum board and plaster of paris from wastes of the fertilizer plant; and edu- cational equipment from the shorts from the sawmill; (b) process some agricultural products (e.g. salt refining, coir and sisal mats or ropes) and other natural resources (wood industries, lime-based construction materials); (c) manufacture products and implements needed by agriculture, e.g., sisal knives, husking blades; and (d) provide construction material and hardware materials, e.g., nuts, bolts, hinges, lime building material, joinery products.  ANNEX 4 Page 4 10. In Tabora, the proposed priority industries would: (a) specialize in products utilizing local resources of cotton, wood and lime; (b) supply needed services such as general engineering workshops and garage; (c) respond to the local demand for educational supplies, agricultural implements and packaging. B. Program Components 11. The small scale industry program for Tabora and Tanga will involve three major components: (a) expansion of SIDO Management Training and Tech- nical Assistance programs and staff at the regional level to increase imple- mentation capacity in the towns without diverting resources from the develop- ment of rural industries; (b) funds for SIDO's program for hire-purchase of production equipment and purchase of critical raw materials for participating small industries; and (c) infrastructure for the industrial clusters as well as provision for credit from THB for loans for low-cost industrial structures and sheds. The costs of the program are summarized in Annex 6, Table 12. Management Training and Technical Assistance 12. Management and technical apprenticeships and training materials for short, practical courses, e.g., on accounting or financial management, will be funded under the project. In Management Apprenticeships, a prospective industry manager or cooperative chairman could spend one to three months with a successful industry with similar product lines and ownership structures to learn the general management methods and problems in getting raw materials, finding markets, managing money, organizing the work, costing products, etc. The project will pay the participants' transportation and subsistence costs as well as a stipend to the industry offering the the training. 13. Project Implementation and Training Coordinators in each town will be responsible for identifying and organizing productive groups which would participate in the industrial clusters and networks. The Coordinator will also be responsible for the start-up of the industries and for on-going manage- ment assistance to the industries in Tanga and Tabora towns promoted under the project. Working with Ardhi and appropriate regional and national administra- tions, he will oversee the orderly settlement on the clusters and insure that the service and facilities provided are properly maintained. In addition, he will work with the Technical Adviser developing training programs and super- vising management apprenticeships, short practical courses and exchanges among artisans. Project Technical Advisers will be hired (probably from outside Tanzania) in both Tanga and Tabora towns for a period of up to two years. They will assist the Coordinator in starting new production cooperatives and  ANNEX 4 Page 5 other industries and will focus on developing management assistance and training programs for cooperatives and SSI networks. In addition, they will be responsible for developing an operating manual for technical assistance and training program which could be used in other regions to assist small-scale industry activities. Hire Purchase of Equipment 14. This component will build upon SIDO's existing Equipment Hire-Pur- chase and Raw Material Bank Programs (see para 27(d) below). Over 85% of T Sh 7.8 million worth of equipment and raw materials made available under the project will require foreign currency. About 60% of these funds have been allocated to specific priority industries within the industry clusters and networks (see Section A), while 20% will be available for incremental equip- ment for existing small industries and for other subprojects (whose feasibil- ity would be determined during the course of project implementation) and 20% for critical raw materials difficult to obtain through normal distribution channels. In addition, funds will be provided for contracting to install equipment and provide critical technical training on the more unfamiliar machinery. While most of the industries involve labor-intensive technologies, nearly all of the equipment is imported and about half of the industries in the clusters would require assistance by the equipment suppliers in installing and technical training. SIDO has positive experience contrating foreign machinery suppliers for this role. 15. SIDO and its predecessor, the National Small Scale Industry Cor- poration, have had hire-purchase schemes. With the one to two years grace period, collections to date have been limited; however, analysis of the portfolio shows that arrears rates are below 10%. Funds for the hire-pur- chase program have recently been made available from various outside sources (see para 28 below), and there is a need to formalize and standardize lending and repayment procedures and terms and to establish a revolving hire-purchase fund within SIDO. The project will provide assistance for this task. 16. Project Coordinators in each town will be responsible for iden- tifying and organizing new productive groups which would participate in the Industrial Clusters and networks. They will also be responsible for coordi- nating the start-up of the industries and providing on-going management assistance to the new industries promoted in Tanga and Tabora under the project. They will be assisted by Project Implementation Advisers in starting new production cooperatives and other industries. 17. Project Economist within SIDO's Tanga and Tabora offices will be responsible for preparing feasibility studies and financial packages including necessary equipment, credit and construction financing. The Project Economist will be the project participants' link with NBC and THB, facilitating the processing and repayments of credit.  ANNEX 4 Page 6 Infrastructure for Industrial Clusters 18. To provide the needed space and infrastructure for small indus- tries, three hectares (adjacent to the project areas) have been set aside as small industries zones; in Tanga the designated small industrial zone is between the two project areas of Gofu Juu and Mwakizaro, while in Tabora, the allotted small industry area is in the project area of Kiloleni. The project will provide funds for basic infrastructure--preparation of land, roads, electricity, water supply and sewerage. Cost per hectare would be about T Sh 110,000 (US$13,000) before contingencies. The infrastructure will be supplied as a part of the upgrading works in project towns. Services on and off clusters will be maintained by appropriate regional and national agencies (Ardhi, Tanesco, Comworks, Regional Water Supply). 19. Traditional industrial estates often involve expensive buildings with costs exceeding the price of machinery and equipment. It is estimated that building costs could be reduced to T Sh 700 (US$84) per square meter by using a combination of traditional building materials and earthern floors, with cement floors as needed for machinery. In several industries planned for industrial clusters, basic infrastructure, combined with construction of basic workshops and sheds with security are all that are needed. In order to minimize costs, project participants will assume the responsibi- lity for actual construction of sheds. 20. T Sh 2.8 million (US$330,000 million) will be available to finance low-cost sheds in Tabora and Tanga towns, on and off the clusters. Loans of up to T Sh 120,000 will be made to program participants at interest rates of 8% with maturities of 15 years and up to two years' grace. Security and down payments will be lower than THB's normal requirements for commercial under- takings given the reduced size and new nature of the enterprises. Ardhi will provide low-cost designs to meet the needs of various types of industries. Local Project Coordinators will supervise construction with the assistance of SIDO's staff. C. Expected Benefits 21. The small industry component will provide benefits in increasing employment, incomes and production in Tanga and Tabora Towns. The component's incremental employment of 870 people over 3 years would cost US$650 in fixed investment per position created and less than US$1,000 including initial working capital needs and technical assistance. This employment figure includes those who were previously seriously underemployed. The majority of the priority industries represent new product lines in these regions and often in Tanzania as a whole; however, they build upon existing skills and resource base. Many of the industries use local resources as inputs process- ing them into finished products which were formerly imported; therefore, the program will result in savings of foreign exchange for inputs and increase local value added.  ANNEX 4 Page 7 22. One major objective of the program is to strengthen SIDO's regional implementation capacity and to develop methods in Tabora and Tanga which can be applied elsewhere. The institution building benefits would include devel- opment of: (a) new, improved training and extension methods, transmitted to other regions through manuals and seminars; (b) regional management and implementation capabilities in urban areas through addition of staff to perform key functions; (c) financial management under the hire-purchase scheme; (d) low cost workshop clusters and networks to replace the capital intensive industrial estate approach formerly favored; (e). coordination among SIDO, ARDHI, NBC and THB. D. SIDO's Performance and Prospects 23. SIDO - Background and Organization. The small scale industry program will be implemented by the Small Industries Development Organization (SIDO). SIDO is a parastatal under the Ministry of Commerce and Industry. It replaced the national Small Industries Corporation (NSIC), established in 1966 as a subsidiary of the National Development Corporation (NDC). According to a 1974 directive on small-scale industries, SIDO should not run or own any industries. Rather, its role should be to render services to new and existing small scale industries. Consequently, SIDO attempts to assist all forms of the small-scale industry from simple self-employed artisans to small factories, during all stages of their activities. 24. SIDO is headed by a Director General who reports to a Board of Directors headed by a Chairman who at present also serves as the Junior Minister of Industries. The Director General is assisted by three Depart- mental Directors of Research and Planning, Training and Marketing and Admini- strative and Account Section. A small extension service has been estab- lished in each region, staffed by one Small Industry Promotion Officer (SIPO) and one Training Officer. Complementing the regional staff, there are Zonal Economists covering three to four regions. The organization of SIDO is shown in Figure 1. 25. SIDO's Performance. SIDO has been in operation for too short a period to permit a definitive judgement on its performance and effectiveness. Assistance to the small-scale industries is an experimental area in which few countries have accumulated solid and transferable experience. Establishment of programs and staff to assist small-scale industries is a long process.  ANNEX 4 Page 8 At present, large variations exist among regions in the quality of programs and considerable improvements in SIDO strategic planning and staff are needed to increase its impact on SSI development in urban and rural areas. Until recently, physical development has been emphasized with 70% of SIDO's budget in 1974/75 allocated to construction of industrial estates and urban workshops in some regions. Additional attention is needed in initial organization of industries, their management, quality control and marketing of final products. Coordnation with other agencies and regional authorities is often insufficient. Project preparation skills are generally satisfactory, although more attention should be given to problems of marketing and organization. 26. These problems can be partially attributed to the scarcity of quali- fied manpower and resources. They also reflect the ambiguities of the original strategy of SIDO and its evolution. One of the main ambiguities concerns the ownership of industries. In ujamaa villages small industries can only be operated by cooperatives, while in urban areas individuals and small partner- ships as well as cooperatives are eligible for assistance. A related problem is the role of small-scale industry in the national development strategy; large parastatals are often promoted to meet the same demand as are the small industries promoted by SIDO. Furthermore, no clear choice appears to have been made between the view of small-scale industry as a social service which provides employment and education, but whose economic viability is secondary and which is to be subsidized; and a view of small-scale industry as an economic viable entity, which has to satisfy the vigorous test of competi- tion and which is not to be subsidized beyond a short start-up period. 27. The services SIDO is providing include: (a) Industrial consultancy and marketing services provided by Regional SIDO staff. (b) Training. Eight training-cum-production centers are presently in operation, including one in Tabora, Muhezg and Tanga. They serve as training and industrial extension nuclei, while producing goods for local consumption. In addition SIDO provides basic training of long duration and short refresher courses for its staff. Until 1975, 55 engineering staff have been sponsored abroad (mainly in India) for specialist training. (c) Industrial Estates and Clusters. SIDO is presently engaged in the implementation of several industrial estate and production- cum-training projects and has plans to take up five more projects during the Third Five-Year Plan. A T Sh 4 million estate in Dar es Salaam has been recently completed, as well as a smaller estate in Songea. In Tanzania's Five-Year Plan, SIDO is committed to developing industrial clusters of estates in each regional center. These and other expansion plans of SIDO would require continued external as well as internal financing.  ANNEX 4 Page 9 (d) Hire-Purchase Programs for Machinery. SIDO has a small hire-purchase program. SIDO's budget for 1975/76 included about T Sh 20,000 per region for hire-purchase to small-scale industry but internal and external funding has since increased. An initial deposit of 10% is required prior to delivery. The balance is lent at 5% for three to seven years (with up to one year grace period) depending on the size of the loan and type of machines. The repayment record appears satisfactory with arrears to date less than 10%. (e) Government Purchase Program. Government departments and parastatal organizations are the largest buyers of indus- trial products. Following a 1974 directive, SIDO arranged a program in which government agencies are encouraged to give preference to purchases from small industry producers. In 1974, over T Sh 3.2 million worth of goods produced by small-scale industries were purchased through SIDO. (f) Raw Materials Supply. At present SIDO supplies its work- shop network with raw materials and plans to extend this service to other small-scale industries, using the outlets of the Regional Trading Companies as well as developing one or two Raw Material Banks for scarce items. 28. SIDO has a variety of funding sources (including Treasury, India and Scandinavian countries), each with a different perception of needed programs of small industries and making funds available at different costs. SIDO faces a major challenge in developing an integrated strategy and standardized interest rates for its hire-purchase program. 29. While the proposed program will be on too limited a scale to allow for short term, direct impact on SIDO's national strategy, policies and organization, the program will be used as an instrument of dialogue with SIDO on these matters and regional approaches can form the basis for overall reform in policies and procedures. Efforts will be made to strengthen SIDO's capability in these areas where weaknesses have been acknowledged. The project will focus on the regional level of SIDO operations and place strong emphasis on managerial assistance and the process of project identification and selection of participants. Assistance will also be provided to SIDO's headquarters: (a) to develop effective methods of management assistance to the small-scale industry especially to cooperatives, on industrial clusters, and off clusters, to decentralized networks of industries; (b) to improve the procedures, structure and performance of the hire-purchase program and to formulate guidelines on lending to small-scale industry; the interest rate will be increased from 5% to 8% in order to bring it in line with interest rates charged by NBC;  ANNEX 4 Page 10 (c) to formulate a policy on the development of ancillary industries; and (d) to reinforce its marketing and management assistance capabilities. 30. Toward a Small-Scale Industry Project in Tanzania. SIDO would almost certainly be the prime mover if the Bank group eventually decided to fund a nationwide small industries project in Tanzania. The major financial intermediary could be NBC with SIDO providing project promotion, preparation, evaluation, technical, management, and marketing advice and organization. SIDO's management has stated that it would be amenable to the eventual estab- lishment of a small industry fund within NBC. In this case, SIDO's hire- purchase would most likely be maintained for use by small industries which require small amounts of equipment to complement new product development or product upgrading efforts. 31. Several major changes would be required in NBC's strategy, staffing and procedures before it could be expected to function as the major financial intermediary for small industries. Until now, NBC has little commitment to and experience in lending to small industries. Until 1975, NBC had lent less than US$10,000 to small industries; in 1975-76 NBC's lending to small indus- tries increased, but the amount was about US$1 million, less than .O1X of its total annual lending. Most of NBC's small enterprise clients are retail outlets which require standard working capital, usually keep books and require little active involvement by NBC. 32. A major limitation of NBC is its charter to lend only for working capital needs; there was a small hire-purchase fund, which has been discon- tinued in the regions visited due to lack of promotion. The headquarters and regional managers claim that at present they would not be able to lend for equipment or worksheds. Board-level discussions had been held on the possible request for a change in NBC's mandate to include investment and working capital, but no quick decision was foreseen. 33. NBC has no specialized small industry staff or program; the accounts department at both headquarters and regional branches handles agriculture, commerce, large and small industry loans. In small industries, NBC relies almost completely on the project appraisals of SIDO staff, with NBC providing complementary working capital to SIDO's hire-purchase clients. Recognizing their lack of personnel with experience in small industry lending, NBC's management recently sent one member of their technical staff to India to study small industries. This lack of specialization precludes NBC's active promotion or information dissemination about the availability and terms of credit to small industries. It also hinders the development of the project preparation, technical and management assistance and loan supervision capa- bilities that are needed to develop small industries in Tanzania. While SIDO can do most of the extension work, some specialized small industry unit would be needed within NBC to appraise projects SIDO prepares, provide some super- vision and act as a link between SIDO and NBC.  ANNEX 4 Table 1 TANZANIA SECOND NATIONAL SITES AND SERVICES PROJECT Small Scale Industry Development--Some Selected Priority Industries TANGA No. of Equipment Incremental Units Investment Employment (T Sh '000) CLUSTERED INDUSTRIES Metal workshop, Agricultural Implement Specialization 1 220 14 Nuts and Bolts 1 250 10 Building Hardware (e.g. hinges) 1 110 20 Steel Structures 1 150 18 Gypsum Board 1 120 15 Plaster of Paris 1 100 10 Chalk 1 110 20 Pencils 1 170 12 Salt Refining 1 110 15 Joinery products, furniture 1 90 15 Subtotal 10 1,430 149 INDUSTRY NETWORKS AND OTHER DECENTRALIZED INDUSTRIES Readymade childrens' clothes 10 200 50 Coir ropes, sisal mats etc. 20 270 100 Blacksmith units specializing in sisal knives, husking blades 6 180 30 Building materials 3 100 30 Subtotal 39 750 210 INCREMENTAL EQUIPMENT FOR EXISTING INDUSTRIES Leather 5 110 20 Mechanics 4 120 16 Carpenters 5 110 20 Others 15 550 60 Subtotal 29 890 116 TOTAL 78 3,070 475  ANNEX 4 Table 2 TANZANIA SECOND NATIONAL SITES AND SERVICES PROJECT Small-Scale Industry Development--Some Selected Priority Industries TABORA No. of Equipment Incremental Units Investment Employm-ient (T Sh '000) CLUSTERED INDUSTRIES General Engineering + Garage 1 220 12 with foundry Sawdust Briquettes 1 100 15 Poultry feed 1 100 12 Wooden Bobbins 1 220 20 Wooden Products - Ed. Supplies 1 70 15 Joinery Products 1 70 15 Cotton Surgical Bandages 1 40 12 Lime Kilns, Lime Building materials 1 70 20 Groundnut Oil 1 110 20 Building Hardware (hinges, ect.) 2 120 25 Sheetmetal - tins, stoves, 1 130 20 (and others) Thread Reels 1 40 5 Subtotal 13 1,290 191 NETWORK AND HOME INDUSTRIES Handloom weaving 10 220 50 Powerloom 5 300 25 Hand Knitting 5 130 25 Readymade Garments 10 230 50 Woodworking specific products (e.g., Beehives) and educational supplies 10 160 60 Subtotal 40 1,040 210 TOTAL 53 2,330 401  TANZAN IA SECOND NATIONAL SITES AN. SERVICES PROJECT Sn'2a1-Scale Indust-' DeveloDmer.t Organizetion of SIDO BOARD OF DIRECTORS Administretive Manager __ D I RECTOR:5ENERAL tAccountant DIRECTOR OF DIRECTOR DIRECTOR OF RESEARCH AND ACCOUNTS OF MARKETING PLANN I NG DEPARTMENT TRA IN ING SERVICES Manager F anager Manager Marae r cManager Manager Ecornomi c Technical Development Recurrent Staff Client Handicrafts of of Services Services Accounts Acccunts Developnt Development Development Marketing Supplies REGIONAL EXTENSICI SERVICES (20) (SllALL INCUSTRIES PROMOTIGN GFFICES) SIFO) (DI _ _ _ _ _ . _ . . _ _ _ . . . _ _ . . _~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~H  ANNEX 5 TANZANIA SECOND NATIONAL SITES AND SERVICES PROJECT Implementing Agencies and Project Organization and Management 1. This Annex provides a brief overview of the principal implement- ing agencies and describes the administrative arrangements for the proposed project. The distribution of responsibility for various administrative functions described in the Annex is summarized in Figure 1 and an implemen- tation schedule is presented in Figure 2. 2. Responsibility for project implementation will be shared between the Ministry of Lands, Housing and Urban Development (Ardhi), which will be designated as the Executing Agency with overall responsibility for the project, and the decentralized administrations in each project town. A. Central Project Administration Ministry of Lands, Housing and Urban Development 3. Ardhi has overall responsibility for formulating and implementing policies for urban development, particularly with regard to land use and housing. It also assists the Prime Minister's Office in administering land development in rural areas. In 1973 Ardhi was reorganized and some of its staff decentralized as part of the regionalization of the Tanzanian Govern- ment. Ardhi's current organization is indicated in Figure 3 and recent ex- penditure patterns are shown in Table 1. 1/ 4. The ministry is confronted with a severe shortage of trained staff (see Table 2). Less than two-thirds of the ministry's approved non-adminis- trative posts in the technical grades are currently filled. The Ardhi Institute, which began as a Survey Training Center, has been expanded to provide training in urban and rural planning and land economics as well as surveying. The institute currently has 21 faculty members. In 1976 approximately 75 students received professional diplomas after three years of study, while another 100 earned subprofessional certificates in surveying after a two-year course. The first diplomas and certificates in housing design, construction and management will be granted in 1979. The institute plans to train 342 diploma graduates and 520 certificate-holders by 1980. The first project includes US$280,000 for expansion of the institute's facilities. 1/ The reorganization of the ministry and its current organization are discussed in Annex 2 of the appraisal report for the first project.  ANNEX 5 Page 2 Sites and Services Unit 5. The Sites and Services Unit will have operational responsibility for the planning and implementation of the project, as it did for the first IDA-financed project. Although it has handled the design and construction of the physical components of the first project satisfactorily, there have been deficiencies in other aspects of project implementation. First, the social aspects of the project have been given inadequate attention. At least par- tially because they were not given assistance in relocating, a substantial proportion of the families whose units were demolished because of upgrading works have squatted elsewhere or become renters rather than moving to the serviced plots allocated to them. 6. Secondly, the Project Manager has been over-burdened. Although the unit is responsible for almost half of the ministry's development expenditure, it had only sectional status and was headed by a fourth-tier manager. The Project Manager was frequently given assignments not related to the project and did not have authority for independent action commen- surate with his units' responsiblities. 7. Thirdly, insufficient attention has been paid to financial management. Records have not been adequately maintained. The turnover of accountants has been high and currently the position is filled by an accounts assistant, who lacks qualification. 8. Fourthly, insufficient guidance and support has been given to local project administrators. A systematic approach to local project management has not been developed, and there are no well established lines of communication between Ardhi and local administrators. Consequently, problems have had to be dealt with on a sporadic, ad hoc basis. 9. Finally, although the Monitoring and Evaluation Team has provided useful feedback about project implementation, problems of communication and unclear responsibility have arisen because the team is supposed to report to the Principal Secretary rather than to the Project Manager. Furthermore, they cannot act as the Manager's representatives to recom- mend remedial actions when problems are identified during site visits. 10. In order to overcome these deficiencies in the proposed project the Sites and Services Unit will be restructured and upgraded to divisional status within the ministry (see Figure 4). Project management and community develop- ment capabilities will be strengthened primarily by shifting established posts into the new sections and converting a few of them into more senior positions. The reorganization was approved by the Ministry of Manpower Development for the beginning of the 1977/78 fiscal year. The organization plan adds only three new technical positions to the unit, but will increase recurrent expen- diture by 15-20% due to the upgrading of some posts.  ANNEX 5 Page 3 11. The Division's Planning Section will function much as the existing unit does. The section prepared preliminary site layouts and has been review- ing the consultants' preliminary engineering and making appropriate modifica- tions. Plans for the community facilities in the upgrading areas will be prepared in consultation with the appropriate ministries and regional offi- cials. The section will also work with local officials to select sites and prepare layouts for the surveyed plot program. 12. Because of the Engineering Section's limited capability, consult- ants (John Burrow and Partners) were hired to prepare preliminary engineer- ing designs for the proposed project, and the detailed engineering and some supervision will also be handled by consultants. The engineering consult- ancy contracts have been used to supplement on-the-job training begun under the first project. Five of the first six trainees were seconded to work on preliminary engineering and supervision with consultants, and the contract for detailed engineering will incorporate similar opportunities for some of the twelve remaining trainees. The capabilities developed through the training program will enable the Engineering Section to assume some of the responsibilities, such as the engineering for communities facilities and the supervision of contractors at a few sites, handled by consultants in the first project. 13. Management and community development functions relatively neglected during the first project will be strengthened under the second project. Three management functions will be given particular attention. First, financial management will be strengthened through recruitment of a Financial Analyst using project technical assistance funds. The Financial Analyst will be responsible for the maintenance of project accounts and preparation of reports, budgets and cash flow projections and will be supported by a Tanzanian Accounts Assistant. 14. Secondly, more attention will be devoted to systematic monitoring of project implementation and identification of problems needing special attention. (see Annex 11) Effective performance of this function will depend on the development of a simple management information system to provide information to Ardhi and guide local administrators in planning and implement- ing the project. In order to familiarize local administrators with project implementation, the section will prepare a brief management handbook, includ- ing standardized planning and reporting forms, and conduct an orientation seminar in Dar es Salaam for local project administrators. The first draft of the handbook was prepared between the appraisal mission and negotiations. The unit will prepare periodic reports on the progress of project implementation in each city based on standardized reports prepared by local administrators and occasional visits to the cities. 15. Finally, research studies will be conducted on specific topics which cannot be monitored adequately through normal procedures. In order to carry out these studies funds will be provided for hiring temporary interviewers, reproducing questionnaires, and processing data.  ANNEX 5 Page 4 16. The community development and lands staff will be responsible for the social aspects of the project and will work closely with local officials to ensure their effective implementation. Their specific responsibilities will include publicity for the project, allocations and titles for the sur- veyed plots, registration of squatter houses and preparation of titles, the various forms of on-site assistance to be provided to residents in the project areas, and the formation of housing cooperatives. During the planning phase of the project, they will develop guidelines and prepare information materials for the social components and work with local officials to plan their imple- mentation. They will also visit the towns periodically to monitor progress and provide assistance. 17. On-Site Assistance. Considerably more attention will be given to explaining the project to community residents and to providing on-site assistance than in the first project. The Sites and Services Unit will prepare guidelines and standard information materials and work with local officials to develop integrated service programs consisting of some or all of the following elements: (i) publicity about the project, emphasizing provision of compensation and relocation assistance to households displaced by the improvements, procedures to be followed in the registration of houses and the availability of THB house improvement loans; (ii) type plans for house improvement and construction, which will be sold at the site offices at prices set to recover reproduction costs; (iii) three to five demonstration houses, including scale models showing different stages of construction, in each city; (iv) technical assistance in house improvement and construction to be provided by artisans; (v) relocation assistance to households displaced by upgrading works, including early payment of compensation and selection of an alternative site, assistance in applying for.a THB loan, and transportation for moving possessions; 1/ and (vi) assistance in the formation of housing cooperatives, to be provided by local Cooperative Officers following guidelines developed by Ardhi. These forms of assistance will be available through site offices in the demonstration houses. 1/ A truck to be purchased by Ardhi for transporting the Mobile Survey Crew from town to town will also be used to relocate displaced households.  ANNEX 5 Page 5 18. S;veying for the surveyed plot program will be done by an Ardhi Mobile SUGVz7 Crew consisting of three surveyors and ten technical assis- tants. Field assistants will be supplied by the Regional Surveyors or hired locally, and survey students from Ardhi Institute will gain field experience working with the program. The crew will be organized in two teams, each consisting of 5 technical assistants and 20 field assistants under a sur- veyor. It will alternate between trips of two to three months to the regional capitals and periods of work in Dar es Salaam. As indicated in para 5.04 of the report the Sites and Services Unit will be responsible for the overall scheduling and coordination of the surveyed plot program. B. Local Project Management 19. In 1974 Tanzania's administrative and political structures were decentralized in an effort to shift important decision-making power to local bodies. The structure now consists of the Central Government, 20 regions, over 70 districts and sub-districts in some localities. Among project cities, Dar es Salaam is a separate region with three districts, Tanga is a district, and Tabora, Iringa and Morogoro are sub-districts. 20. Figure 5 provides a schematic overview of the national administra- tive structure. At both the regional and district levels teams of functional officers operate under the leadership of Development Directors. Sub-district teams are led by Assistant District Development Directors (ADDD's) and have fewer functional officers. Planning Officers responsible for financial planning and administrative coordination assist the Development Directors. Although the functional officers follow policies set by their ministries, they are responsible to their Development Directors for local administrative matters. Local Project Management: Overview 21. In the first project primary responsibility for local project ad- ministration was assigned to the Regional Land Development Officers (RLDO's). These offices are part of the regional administrations under the Regional Development Directors (RDD's) and are responsible for physical planning, surveying, land valuation, and allocations. Experience with the local administration of the first project has indicated two general shortcomings of this arrangement. First, no single local official has explicit overall operational responsibility for the local implementation of the project, resulting in ambiguities about responsibility for solving problems and communication difficulties locally and with Ardhi. Second, there is no formal mechanism for ensuring that the various local officials who need to be involved in the project make the required inputs on schedule.  ANNEX 5 Page 6 22. A more formal, systematic approach to local project management will be employed in the proposed project. Teams of local officials under the direction of Project Coordinators designated by the RDD's will be responsible for local implementation. The Assistant Development Officer, District Planning Officer or one of Ardhi's decentralized staff members will be designated Project Coordinator. The Coordinators' responsibilities will be to plan local project implementation with assistance from the Sites and Services Unit, to coordinate the activities of the various local officials involved in the project, and to monitor progress and submit periodic reports to their RDD's and Ardhi. The Coordinators will either establish formal Project Committees or rely on less formal coordination mechanisms. 23. The Project Coordinators are not expected to devote full time to the projects; their involvement is likely to vary from a few hours to several days a week. A Site Officer (a junior official with technical skills in construction) will be appointed in each project town to assist the Coordinator and to ensure a continuous administrative presence in the project communities. The Site Officers (as well as other officials) will work out of demonstration houses to be built on the sites. The Site Officers represent the only increment to the local administrations planned for the local management of the projects. Other officials will participate in project execution as part of their regular administrative duties. 24. The success of this approach to local project administration will be heavily dependent on the guidance and support provided by the Sites and Services Unit. The orientation seminar in Dar es Salaam for local project administrators will provide an opportunity to familiarize them with adminis- trative requirements of the project. The management materials prepared for local administrators will identify the major tasks in project execution, indicate their interrelationships, and suggest simple management tools such as planning and monitoring charts. A project implementation schedule worked out between the local administrators and Ardhi will indicate specific targets which can be used to plan local activities and to monitor progress. The management assistance to be provided to Ardhi under the project is described in Annex 8. Upgrading Program 25. Tenure. House-owners in the upgrading areas will be issued titles based on house rather than plot occupation. 1/ The titles will be issued for five years and normally be renewed automatically. They will specify 1/ Tanzania's land tenure system is described in Annex 6 of the appraisal report of the first project. The most salient features of the system are that land was nationalized in 1963 and that land use is normally governed by short (usually annual) or long-term (usually 33 years) rights of occupancy issued by Ardhi.  ANNEX 5 Page 7 annual charges which can be revised with six months advance notice. Title- holders will be obligated not to erect additional buildings or transfer their holdings without permission from the District Land Office. The titles will be convertible to standard rights of occupancy if the holder arranges for the effective plot on which the house stands to be surveyed and registered. House-owners will be required to pay fees of approximately T Sh 90 for title preparation and registration and stamp duty in order to obtain a title. The legal title form to be used to grant tenure in the upgrading areas was re- viewed and revisions agreed at negotiations. 1/ 26. The Site Officers will work closely with local leaders to ensure community involvement in the project. CCM is organized in ten-house cells at the local level and is particularly strong in some of the project communities. In 1977 Ward Assemblies and Councils will be elected according to the Urban Wards (Administration) Act, 1976. These bodies are intended to complete the process of decentralization by creating popular bodies which can participate in decision-making at the grassroots level in the urban areas. Either the Construction and Transportation Committee which each Ward Council must establish or a specifically constituted project subcommittee will serve as the principal formal link between the community and the project administration. Working relationships will be developed with key local leaders. Surveyed Plot Program 27. Allocation of Plots. The RLDO's are responsible for the allocation of surveyed plots. Arrangements for allocation of the plots included in the program will begin before completion of the surveying. Because the program will involve a substantial number of new plots, their availability will be publicized in existing low-income communities, and information about house- holds on the waiting list will be updated. Final decisions about alloca- tions are made by District Allocation Committees consisting of administrators, elected representatives and CCM officials on the basis of files prepared by the RLDO's. 28. Allocations will be made on the basis of a point system weighted to favor households with monthly incomes below T Sh 1,000 (US$120) a month. It is Government policy to promote mixed-income communities and housing opportunities are scarce for even better-off households in Tanzania because of the weakness of the private sector. Therefore, although applicants in the 1/ The legal form will be called an "Offer of a Right of Occupation and Use". This type of instrument is normally used for land which can be adequately identified even though it has not been surveyed. R.W. James notes that "the expression 'offer of a right of occupancy' is misleading because the document is not a mere offer, but it is a final document embodying agreed terms" (Land Tenure and Policy in Tanzania, p. 122).  ANNEX 5 Page 8 target group will be given preference, households with monthly incomes over T Sh 1,000 will not be absolutely excluded. Other criteria to be considered, but given lower weightings in the point system will include number of dependents, employment status and length of time on the waiting list for plots. Assurances were obtained from Government at negotiations that the allocation procedures will give strong preference to households in the target income group. 29. Tenure. Following Government policy, long-term rights of occupancy issued for 33 years will be employed in the surveyed plot areas. Responsibil- ity for the issuance of rights will be shared between the Land Offices and Ardhi following standard procedures. Allottees are informed of their selec- tion and receive a Letter of Offer specifying fees which must be paid to obtain title. Based on annual Land Rent and Service Charges of T Sh 75 for unserviced plots, these fees typically amount to about T Sh 130 plus the Land Rent and Service Charge due for the remainder of the year (Table 3). However, Ardhi has agreed to revise the survey fee schedule during 1977 to recover the actual average cost of surveying plots and the review of the Land Rent and Service Charge is likely to result in higher annual charges. Therefore the level of the initial fees will also increase. If the fees are not paid within a month, the Government is entitled to reallocate the plot. 30. After being notified that an allottee has paid the required fees, Ardhi's Land Development Division in Dar es Salaam assigns a Land Office Number to the newly opened plot file, secures a deed plan from the Survey Division and prepares a Certificate of Occupancy. The certificate is then sent to the Land Office and the allottee is notified to report to sign it. The signed certificate is then returned to Dar es Salaam for the signature of the Director of Land Development Services and forwarded to the Registrar of Titles who registers it and returns a copy to the allottee.  ANNEX 5 Table 1 TANZANIA SECOND NATIONAL SITES AND SERVICES PROJECT Ministry of Lands, Housing and Urban Development Recurrent and Development Expenditure (TSh'000) 1973-74 1974-75 1975-76 1976-77 1/ Recurrent Expenditure 1. Administration and General 2/ 2,334.7 3,401.1 4,588.1 6,299.5 2. Surveys and Mapping 4,139.1 4,969.4 4,396.2 4,265.7 3. Land Development Services 2,800.3 3,283.1 3,430.3 3,715.4 4. Urban Planning 601.0 1,031.1 1,475.3 2,082.8 5. Housing Development 256.6 459.8 862.6 1*441.0 6. Housing and Building Research Unit 183.7 452.8 582.0 783.4 TOTAL 10,315.4 13,597.3 15,334.5 18,587.8 Development Expenditure 1. Surveys and Mapping 15,455.2 10,007.7 4,601.6 9,000.0 2. Economic Infrastructure, Dar es Salaam & Arusha 11,526.0 2,041.8 4,993.0 13,550.0 3. Urban Planning 171.6 286.1 330.4 1,400.0 4. National Sites and Services Program 3,409.1 10,397.5 27,009.0 33,120.0 5. Expansion of Ardhi Institute 1,031.5 2,837.0 9,219.9 3,040.0 6. Housing and Building Research Unit 785.0 729.1 827.8 940.0 7. National Housing Corporation 5,120.0 - 6,144.6 7,100.0 8. Registrar of Buildings - - - 5,800.0 TOTAL 37,498.4 26,299.3 53,126.2 73,950.0 1/ Approved expenditure. During 1973-76 the average deviation between approved and actual recurrent expenditure was 1%; actual development expenditure averaged 16% less than planned. 2/ Includes Ardhi Institute, which accounted for 80% of the increase in recurrent expenditure for this category between 1973-76. Source: Ardhi.  TANZANIA SECOND NATIONAL SITE AND SERVICES PROJECT Ministry of Lands, Housing and Urban Development Staffing Patterns, 1976 ~~~~~~~~~~~Snon 8 Senior TechnicalOteTos tion mlitB Ap~~~Aproved Actual Approved &tI Approved Actual Approved Actual Approved -Actual koproved Actual Administration and Ceneral 1I114 126 Planning Unit I 1 4 3 Manpower Development 3 2 6 5 4 4 2 1 Otber 1 1 10 7 25 23 87 78 Survey and Mapping 1 1 2 1 82 61 176 168 113 87 374 318 Land Development Services 1 1 309 210 Legal Section 2 1 42 9 31 20 26 24 Land Economics Section 1 1 18 3 31 19 19 18 Land Registry 1 1 15 13 17 11 19 15 Customary Land Tribunal 1 1 2 3 5 2 30 22 Rental Tribunal 3 1 4 4 41 41 Urban Development 1 1 179 132 Urban Planning Division 1 1 2 2 37 34 6 8 29 15 Housing Development Division 1 1 3 1 44 32 17 9 38 28 TOTAL 6 6 17 12 263 171 316 268 404 329 1,006 786 Actual as I of Approved 100Z 71% 65% 85% 81% 78Z Note: Ardhi. Institute and Housing and Building Research Unit excluded. Source: Ardhi..  ANNEX 5 Table 3 TANZANIA SECOND NATIONAL SITES AND SERVICES PROJECT Fees for Rights of Occupancy for Surveyed Plots (T Sh) Fee Method of Calculation LR & SC Fee Preparation Fee < 250 40 251 - 500 100 501 - 1,000 300 > 1,000 500 Duty for Duty for Each LR & SC Original Duplicate (4 Req.) Stamp Duty < 100 5 2.5 101 - 200 8 4 201 - 400 10 5 Plot Size (Sq ft) Fee Survey Fee < 7,500 25 7,501 - 20,000 125 > 20,000 225 Fee for Deed Plans 7 per copy (6 copies required) Registration Fee Round [(LR & SC x 20)/1,000] to whole number; multiply result by 5 Advance LR & SC Advance payment for remainder of first year (LR & SC = Annual Land Rent & Service Charge.)  TANZANIA SECOND NATIONAL SITES AND SERVICES PROJECT RESPONSIBILiTY MATRIX: PROJECT MANAGEMENT AND SOCIAL COMPONENTS TRAL ADMINISTRATION LOCAL ADMINISTRATION AROHI SITES & SERVICES ~~~r _ l __ _ __ __s"-~ I7:~:~7~A::v~FUNCTION*S c 'i ~ . O ~& ? , ARDHI: SITES AND SERVICES Preparation of title form for upgrading areas 0 Preparation of management handbook 1 0 0 * 0 Preparation of community development guidelines & materials O Planning for & condu^t of orientation seminar 00 0 0 0 Review of type plans for demonstration houses 0 0 Preparation of periodic progress reports 00 0 Conduct of small research studies - LOCAL ADMINISTRATION Designation of Project Coordinators Appointment of Site Officers * 0 Preparation of local implementation schedule l * I O 0 0 0 00 0 UPGRADING COMPONENT Publicity in communities . 0 0 Registration of squatter houses o * O Preparation of titles 0 0 0 0@ 0 Compensation of displaced households 0 0 * 0 Relocation assistance to displaced households 0 * 0 Management of site offices * 0 0 0 Construction of demonstration houses * 0 0 Sale of demonstration houses * O Technical assistance (artisans) 0 * 0 Formation of cooperatives _ SURVEYED PLOT COMPONENT Site selection *0 O Base map and layout preparation * O Planning review Allocations survey * Title preparation O Construction of demonstration houses - COMMUNITY FACILITIES Operation of schools e Operation of MCH's Operation of markets 0 Budgetary planning for recurrent expenses _ - MAINTENANCE Maintenance of roads and drains Experimental community maintenance program o World e15748 * Primary direct responsibility C Secondary responsibilitY or responsibility for inputs  TANZANIA SECOND NATIONAL SITES AND SERVICES PROJECT IMPLEMENTATION SCHEDULE: PROJECT MANAGEMENT AND SOCIAL COMPONENTS 1976 1977 1978 1979 1980 1981 A1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 3 Preparation of title form for upgrading areas _ Restructuring & upgrading of Sites & Services Section _ Preparation of management handbook Preparation of community development guidelines & materials Preparation for and conduct of orientation seminar Review of type plans for demonstration houses Preparation of periodic progress reports _ - _ - _ _ _ _ _ LOCAL ADMINISTRATION Designation of Project Coordinators Appointment of Site Officers Preparation of local implementation schedules _ UPGRADING COMPONENT Publicity in communities Registration of squatter houses & preparation of titles _ _ _ _ _ Compensation & relocation of displaced households Construction of demonstration houses Operation of site offices - _ _ _ _ Sale of demonstration houses Technical assistance (artisans) _ _ _ _ _ _ THB lending program _ _ _ _ _ - _ _ _ _ _ - SURVEYED PLOTS Allocations _ _ _ _ _ _ _ _ _ _ Preparation of titles _ _ _ _ _ _ _ _ _ _ _ _ = = _ THB lending program Construction of demonstration houses COMMUNITY FACILITIES Budgetary planning for recurrent operations _ Recurrent operations MAINTENANCE Maintenance of infrastructure Experiment in community maintenance of roads & drains _ _ _ L - World Bank-16746 CHART No. World Bank-16747 shows the implementation schedule for the physical components of the project.  Li r~~~~~~~~~~~~~~~~~~~~~~~~~~~ Tv~~~~~~~~~~~~~~~~  ANNEX 5 CURRENT SITES AND SERVICES SECTION Figure 4 Housing l ~~~~~~~~Devlelopment Monitoring & Evaluation Physical Survey Grew | Egineering Estate Housing Plan -ing Trainees Management Cooperatives PROPOSED SITES AND SERVICES UNIT Principal Secretary Project Manager Planning Engineering Finance Section Section Section  TANZANIA SECOND NATIONAL SITES h SERVICES PROJECT OVERVIEW OF TANZANIA'S ADMINISTRATIVE STRUCTURE PRESIDEN NTj -- - -- - -- - -- - -- - -- - -- - -- - -- -I r-I i____UR 5 E ~l MIITROF FINANCEj RIME MINISTER ...M.NSE OF MAR. ANOSPLANNING E POWE DEVELOLME, [ I_ I1 OF MINISER OFMINISER OfMINISER OfMINISMINISTER INISTER OF M NISERO MI-NIST' OF INITERO MMINISTER . OFMINSTER OF [AGRIUTUECMMREINDUSTES, hEALTH ORSAERLFQEOINGNOM AOl&SOILCULTURE INISTER OF INATURL FOUCATION RESOURC6S REGIONAL F- - - - --'-AI --- I DEVEPENT COMMISSIONER I IEGIONAL * PLANNING & . ||NANCE OFFICER I H~~~~~~~~~~~~~~~ I - COOPERATIVES 7l NATURA. C L .L S W OCRLzE S M. UR _ MARC O SUBDISI A.DISRIT PANTN V ILLAGES \.n II,.II.1,.,.I ,,4A,.I ,IIIIWI~II~MRICIWIISAI~RW~*~I*.VI lAA I  ANNEX 6 TANZANIA SECOND NATIONAL SITES AND SERVICES PROJECT Design Standards and Cost Estimates A. Physical Planning Squatters Upgrading 1. Settlement selection. A detailed evaluation of existing squatter areas in the five project towns was carried out before selecting the nine settlements for upgrading under the project. The selection criteria included: (a) Socio-economic Characteristics of the Community and Housing Conditions - socio-economic vitality of the community; - permanency, quality of building materials and general condition of dwellings; and - potential for upgrading and housing improvement. (b) Availability of Off-Site Infrastructure - proximity of existing trunk roads, primary and secondary distributors; - proximity of existing water supply reticulation, storage and treatment works; - proximity of existing storm culverts, streams and ditches; and - proximity of existing power and telephone lines. (c) Extent and Condition of Existing On-Site Infrastructure - pattern of existing circulation network and access reserves, if any; and - types of sanitary facilities, water supply and drainage. (d) Physical Characteristics of the Site and Land-Use Patterns - topography, ground conditions and natural drainage of site; - scale or size of the site and patterns of land use: space available for infrastructure and community facilities; - density of dwellings.  ANNEX 6 Page 2 (e) Access to Employment Opportunities - extent and potential of informal sector within site; and - distance, availability and cost of transport to existing employment areas. (f) Relationship to General Planning Framework - conformity with existing patterns of development; and - development potential and constraints within planning context for future growth. 2. Squatter settlements were ranked according to these criteria and areas most in need of improvement and with the best opportunities for successful implementation were identified. 3. Site Plans. For each of the selected settlements Ardhi planners carried out surveys and designed circulation layouts on aerial photographs, conforming as far as possible to existing patterns of development. Adequate provision of space has been made for community facilities at convenient locations in each settlement. The consultants prepared engineering designs for infrastructure on the basis of these site plans using standards described in Section B of this Annex. Surveyed Plots 4. Site Selection. The process of selecting suitable sites for surveyed plots will be critical to the program's success. A systematic and rational approach to the evaluation of available sites will be developed and employed by Ardhi. Table 1 lists the criteria to be considered in the evaluation. Sites with the greatest potential will be identified through this process. 5. The Physical Planning Section of Ardhi's Sites and Services Divi- sion will be responsible for providing guidelines and assistance to regional urban planning offices for preparing site plans for the selected sites. The site plans will incorporate the following planning and design considera- tions: (a) adequate land use planning to ensure maximum locational convenience of housing areas, community facilities areas and circulation networks in terms of access, safety and social interaction and to minimize costs; (b) adequate provision of land for community facilities, i.e., schools, clinics, community centers, markets, day care centers, etc.;  ANNEX 6 Page 3 (c) grouping of plots for community facilities to form an attractive, viable nucleus for community activities and to permit multiple uses of building and play areas; (d) provision of a safe, convenient system of pedestrian circulation linking lots to future bus lines, play areas and community facilities; (e) provision of vehicular circulation within the area which will allow access to residential lots, commercial and public areas and permit an easy flow of traffic while protecting the neighborhoods from through traffic; (f) provision of limited car parking and access for garbage pick-up facilities; (g) designing of lot layouts to reflect local social preferences and life styles, and to minimize disturbances such as traffic noises, industrial activities and natural nuisances; (h) efficient physical planning to comply with desirable ranges of population density 1/ and land use percentages 2/ for circulation, public areas, private areas and lot coverage-- all features of a sound physical layout which is the primary determinant of subsequent commitment in terms of cost of land, infrastructure networks, administration and maintenance; (i) convenient location of the initial standpipe cluster for provision of potable water; and (j) planning and design provision for staged installation of basic services in the future. The requirements of each utility will be considered individually (i.e., roads, footpaths, stormwater drainage, water supply, power supply and security lighting) and allowance made for incremental upgrading from initial levels of service which may be lower than desirable minimums. 1/ Defined in this instance as the number of persons divided by total usable area. Desirable/effective gross population density range in such projects is 250-500 persons per ha. 2/ Based on standards developed by several international agencies, academic institutions and public planning agencies, these ranges are: 20-30% for circulation areas, 50-60% for private areas, and 15-30% for other public areas.  ANNEX 6 Page 4 B. Infrastructure Standards 6. In designing the infrastructure, standards employed in the first project (see Annex 5 of Report 337a-TA) were utilized with appropriate modifications. As in the first project, minimum functional requirements were considered in determining initial levels of services and provision was made in the engineering plans for future upgrading as needed. Infra- structure cost estimates for the various sites are summarized in Tables 3 and 4. Detailed breakdown of cost estimates for a typical site, Kiloleni (Tabora), are shown in Table 5 for reference. 7. Roads. Each site will have a network of collector, primary, secondary and pedestrian roads. To the extent possible, layouts of the networks conform to existing patterns of circulation and housing. Design specification for the four different roads will be as follows: (i) Collector Roads (Class A) are designed to serve as the main access routes linking the sites with the urban road network. They would accommodate heavy vehicles including bus traffic which will be provided after upgrading works are completed. Class A roads will have an overall reservation of 36 meters with an initial carriageway width of six meters. The carriage- way construction comprises a treated dry macadam surfacing (20 mm double surface dressing on 150 mm base course) on mechanically stabilized gravel base (200 mm sub-base). The edges of the pavement will be strengthened with concrete border curbs. 4.5 m wide shoulders are provided on both sides for pedestrians and cyclists and for future expansion of the carriageway. (ii) Primary roads (Class B) will serve as the main vehicular access to dwelling groups for taxis, garbage trucks, cesspit emptiers, fire engines as well as private cars. These roads will have an overall reservation of 24 meters with a 4.5 m wide carriageway of rolled gravel on mechanically stabilized gravel (150 mm base course). 3.25 m shoulders are provided on both sides. (iii) Secondary Roads (Class C) are dirt roads which will serve as vehicular access to inner areas for light traffic and occasional service and emergency vehicles. The road has a 12 m reserve with a 5 m carriageway of compacted earth. (iv) Pedestrian Paths (Class D) will function as pedestrian and cyclist accesses to houses as well as emergency vehicles. The D-roads will have a road reserve of six meters.  ANNEX 6 Page 5 8. vat2r Supply. 1/ Potable water will be supplied through standpipes 'n all project areas. The design of on-site distribution systems for squatter upgrading areas is based on the assumption that the percentage of population served through public standpipes will decrease over time from 100% in 1976 to 70% in 1986. Accordingly, the primary distribution reticulation comprising pipes of diameter 75 mm or more, is designed to meet the water demand of 1986. Water velocity is about one m/sec in the primary distribution pipes. The secondary distribution lines comprising pipes of diameter less than 75 mm are designed to meet the demand from standpipes. The number of taps in one standpipe is limited to four to avoid crowding. A 4-tap standpipe is expected to serve 50 plots. Fire hydrants are provided only around community facilities and are spaced at 300 m intervals. 9. The specific water demand criteria adopted were 40 liters per capita (lcd) for people served through pjblic standpipes, 100 lcd for people served thrSugh private connections, 30 m /ha/day for community facilities and 17.5 m /day per school. Water losses totalling 12% of consumption have been taken into account to calculate the total water demand. Peak water demand factors have been estimated at 1.20 for the peak day and 2.9 for the peak hour. The future annual population growth adopted in the squatter areas was assumed to be 5%, except in Mtoni-Tandika and Hanna Nassif in Dar es Salaam where 2% growth only was adopted. These figures are based on growth records of the various towns and particularities of each project area. 10. For the surveyed plots the standards of water supply distribution systems would be limited to the simplest. Full on-site water supply reticulation would not be laid. Provision is made in the project for water mains, designed for an average per capita consumption of 40 liters/day, of 0.5 mile maximum length which would supply water to each project site through public standpipes only. The number of standpipes would be limited to one per 350 plots. The other criteria used to design the water main would be the same as for squatter areas. 11. Surface Water Drainage. The drainage system is designed to carry storm water and domestic waste water other than sewereage from latrines. The system is designed for peak storm frequency of five years and consists of unlined open ditches along the roads and through culverts at vehicular crossings. On steep gradients lining will be provided to prevent erosion. 1/ The status of off-site water supply networks and general water infrastructure for each town is summarized in Table 2 of this Annex.  ANNEX 6 Page 6 12. Sanitary Sewerage. As in the first project, the traditional pit latrine will be used for sanitary sewerage. In the upgrading areas, pit latrines already exist in the majority of houses. Construction and maintenance of pit latrines is the responsibility of project bene- ficiaries. Technical assistance is provided by Ardhi for this. When the pit latrine is full, it either has to be emptied or a new one built. Cesspit emptiers provided under the project would empty pit latrines in project areas in Dar es Salaam and Tabora. Where soil conditions and ground water tables are unsuitable, house-owners would be encouraged to install an alternative drycell system for which technical assistance and loans from THB would be provided. 13. Power Supply. A system of street lighting and security lighting at major communal areas will be installed. The center-to-center spacing for street lighting will vary within a maximum spacing of about 50 m on Class A and B roads. The distribution system will be designed to allow for future individual metered connections which would be installed by TANESCO on application. Cost Estimates 14. Detailed cost estimates for infrastructure, community facilities, vehicles and equipment, technical assistance, house improvement and con- struction loan program and the small-scale industries component are shown in Tables 3, 7, 9, 10, 11 and 12, respectively. Table 4 describes unit costs for infrastructure works for squatter upgrading. Table 5 contains the detailed costs of infrastructure works for a typical upgrading site (Kiloleni, Tabora--the total project infrastructure costs for squatter upgrading were aggregated from similar data on each site). Table 6 shows cost estimates for basic infrastructure (water) for surveyed plots and Table 8 describes unit costs and space standards for community facilities. All costs are expressed in mid-1977 prices.  ANNEX 6 Table 1 SITE EVALUATION CRITERIA FOR SURVEYED PLOT PROGRAM CRITERIA EVALUATION FACTORS RATING A. Land Availability - size (at least 50 ha. to accommodate suitability for proposed the larger plot sizes and pit latrines use traditionally used in Tanzania) - acquisition costs (compensation for exisitng crops and buildings) B. Physical Characteristics - topography (ridge, valley, slope) suitability for proposed development - ground conditions (soil type, rock) - natural drainage and groundwater table (swampy, well drained grasslands or high groundwater table) C. Off-Site Infrastructure - water supply reticulation, storage availability and proximity water treatment works of existing and feasibility for required trunk infra- - trunk roads, primary and district structure distributors - trunk sewers, pumping stations and sewerage treatments works (for sites where applicable) - storm culverts, streams and ditches - power lines, transmission plants - telephone lines D. Employment Linkages - accessibility of the site to existing Access to job opportunities employment sources (distance, availability and cost of transport) - future employment potential in the vicinity of the site (committed within timeframe of proposed development) E. General Planning Framework - conformity with existing development relationship of site with planning context of its - potential for expansion to accommodate urban area future growth (short term and long term) - relationship to proposed or adopted development plans or structure plans - relationship to other similar areas including upgrading projects.  ANNEX 6 Table 2 TANZANIA SECOND NATIONAL SITES AND SERVICES PROJECT General Water Supply Infrastructure for Project Sites Project Town Status of Water Supply Dar es Salaam: The lower Ruvu River scheme financed by the Canadian International Development Agency (CIDA) has increased the production and transmisSion capacity of t1e water supply system from 95,000 m /day to 263,000 m /day. The new facilities were commissioned at the end of 1976. In addition, KfW has recently financed the expansion of distribution reservoir capacity and transmission mains in the town. No off-site mains would have to be laid to supply the project areas except for the (Mbagala) area where an off-site water main has been included under the project to strengthen the supply of the on-site distribution system. Tanga: The Siggi River scheme financed by KfW will make up for the deficit in water production capacity of the bore- holes presently supplying the town. The new T Sh 165 (US$20) million scheme consists of a dam, raising mains, a treatment plant and new storage tanks. There will be more than sufficient capacity to supply the project areas. The scheme is presently under imple- mentation and will be substantially completed before construction of the reticulations of the present project begins. Tabora: A new T Sh 83 (US$10) million extension scheme financed also by KfW will make up for the deficit in water pro- duction capacity of the Iyembe River Scheme presently supplying the town. The new scheme includes capacity for development in upgrading areas of the project. It is under implementation and will be completed before commissioning the reticulations to be built under the project. Morogoro: New production, transmission and distribution facilities have just been financed mainly by the Bank (Urban Water Supply Project). The new T Sh 125 (US$15) million scheme will increase the peak production capacity of the  ANNEX 6 Table 2 (Cont'd) Project Town Status of Water Supply water supply system from 2,300 m /day to 27,300 m /day. It includes the construction of an 1,650 m long earth dam across tqe Ngerengere River for storage of up to 10 million m of water, transmission pipelinS a treat- ment plant and a pumping station of 25,400 m /day capacity. It also includes the 3construction of dis- tribution reservoirs of 5,450 m total capacity and the extension of the main distribution networks to bring it up to an efficient state by early 1979. The new facilities are expected to be commissioned by October 1979. Iringa: A feasibility study for expansion of the water supply system in Iringa has been completed (and appraised by -the Bank in conjunction with six other urban water supplies) but financing has not been found to date. Existing production facilities are incapable of meeting present needs 3(water production has levelled off at 2,200-2,500 m /day since 1971) and average per capita consumption in the town is very low (45 lcd). The present system is therefore incapable of serving addi- tional reticulation without the risk of reducing pressure in the pipes which would cause network contamination and health hazards. The squatter settlement included for upgrading under the project, Kihesa, has some water service -- through 93 private connections and 7 public standpipes. Therefore, a limited strengthening only is proposed under the project until funds (US$4.7 million) are available to finance increased production and transmission facilities in the town.  TANZANIA SECOND NATIONAL SITES AND SERVICES PROJECT Cost Estimates - Infrastructure (in '0000 T Sh) (1) (2) (3) (4) (5) (6) (7) (o) (9) (10) (11) (12) No. of Houses To No. of Houses Rods Water PysiaL rie /5 City Site Area (ha) Re Upgraded To Re Upgraded SWD) X Supply Electricity Preliinaries Subtotal Allouance Cantingeneies CcttinPSncies Total Tabora Isebya 114.5 1,445 140 4,693 1,351 264 631 6,939 2,082 1,353 1,245 11,619 Kiloleni 72.9 1,159 40 2,394 1,269 74 374 4,111 1,233 802 737 6,883 Taizga Gofu Juu 21.6 397 43 1,372 271 81 172 1,896 237 320 294 2,T47 Mwakizaro 21.8 265 27 942 115 53 111 1,221 153 206 190 1,770 Morogoro Kichangani 16.8 413 63 894 1,224 63 218 2,399 180 387 356 3,322 Msamvu 72.7 1,443 150 2,170 821 188 318 3,497 262 564 519 4,842 Iringa Kihesa 66.4 1,058 30 2,075 146 91 231 2,543 636 477 439 4,095 Dar es Salsa. Mtoni/Tandika 470.6 7,115 885 L 12,383 10,084 704 2,317 25,488 - 3,823 3,517 32,828 Ranna Nassif 43.3 998 140 888 1,091 180 216 2,375 356 328 3,059 Mbagala 50.0 _ - 3,046 668 316 403 4,433 - 665 612 5,710 TOTAIS 950.6 14,293 1,518 30,857 17,040 2,014 4,991 54,902 4,783 8,953 8,237 78,875 /1 SWD - Surface Water Drainage. Z2 10% of the sum of (4), (5) and (6). L Experience of the First Project indicated significant differences in the cost of infrastructure works between Dar es Salsm and upcountry towns. These difference reflect higher transportation costs and generally higher costs of construction in upcountry towns -- lack of skilled manpower, inadequate maintenance facilities and poor cammunication with Dar es Salaam. In the absence of reliable data or unit construction costs in upcountry towns, a location allowance expressed as a percentage of infrastructure cost for Dar es Salaam was derived for each of the upcountry towns (30% for Tabora, 12.5% for Tanga, 7.5% for Morogoro and 28% for Iringa) on the basis of consultants' and Ardhi's experience. 4 15% of the amef (8) and (9). / 12 of the s of (8), (9) a (10). 6 To be resettled in lbagala alte.  TANZANIA SE(XED NATI(NAL sITES AND SERVICES PJEcT Cost Estimtes - Unit Costs for Infrastroctbre (In T Sh'000) Area Number of Cost 2ar Nectare _oCt per city Site (ha) Households Total Cost Road & SW]h / Water Supply Eleotricity Total Household Tabora Ieebyn 114.5 1,585 6,939 (9,021) 45 (58.5) 13 (16.8) 2.6 (3.14) 60.5 (78.7) 4.4 (5.7) Kiloleni 72.9 1,199 4,111 (5,3114) 36.1 (47 ) 19.1 (24.9) 1.1 (1.4) 56.3 (73.3) 3.4 (4.4) Tanga Goru Juu 21.6 440 1,896 (2,133) 69.8 (78.5) 13.7 (15.5) 4.2 (4.7) 87.7 (98.7) 4.3 (4.8) )wakJzaro 21.8 292 1,221 (1,374) 47.5 (53.5) 5.8 ( 6.6) 2.6 (3) 55.9 (63.1) 4.2 (14.7) Morogoro Klcehangani 16.8 476 2,399 (2,579) 58.5 (6Q,9) 80.2 (86.1) 4.1 (4.4) 142.8 (153.4) 5 (5.4) Msamvu 72.7 1,593 3,497 (3,759) 32.8 (35.3) 12.4 (13.3) 2.9 (3.1) 48.1 (51.7) 2.2 (2.3) Iringa Xlhesa 66.4 1,088 2,543 (3,179) 32 (12.!5) 2.4 (3.) 1.9 (2.4) 36.3 (147.9) 2.3 (3) IDr as Salaam xtod/Tndikab/Magala 520.6 8,000 29,921 32.6 22.7 2.2 57.5 3.7 Hans Nasif U5 1 .1