Guinea - National Rural Infrastructure Project

Document of The World Bank FOR OFFICIAL USE ONLY 5ZZ g G 61- 67-UI Report No. 8012-GUI STAFF APPRAISAL REPORT REPUBLIC OF GUINEA NATIONAL RURAL INFRASTRUCTURE PROJECT NOVEMBER 27, 1989 Agricultural Operations Occidental and Central Africa Department This document has a residcted dribution and may be used by recipient only In the performe of their offlidil dudtes. Its contents may not otherwise be disclosed without Wi  CURRENCY EQUIVALENTS Currency Unit - Guinean Franc (GF) US$1.0 - GF 590 1/ SDR 1.0 - US$1.28 21 WEIGHTS AND MEASURES Hecric System ABBkEVIATIONS AND ACRONYMS AFVP - Association Frangaise des Volontaires du Progrbs BSD - Bureau de Strat6gie et de Developpement (MARA) BTGR - Bureau Technique du Genie Rural CAF - Cellule Administrative et Financiere (DNGR and SNAPE) CEP.. MkAG - Centre d'Etudes et de Perfectionnement en Machinisme Agricole DNFPR - Direction Nationale de la Formation et de la Promotion Rurale (MARA) DNGR - Direction Nationale du Genie Rural (MARA) FAC - Fonds d'Aide et de Cooperation (France) KfW -. Kreditanstalt fUr Wiederaufbau (Federal Republic of Germany) MARA - Ministare de l'Agriculture et des Ressources Animales MEF - !linistbre de l'Economie et des Finances MPCI - Ministere du Plan et de la Cooperation Internationale MTTP - Ministere des Transports et des Travaux Publics NGO - Non-Governmental Organization PNIR - Projet National d'Infrastructure Rurale PPF - Project Preparation Facility PNRVA - Projet National de Recherche et Vulgarisation Agricoles RWS - Rural Water Supply RWSS - Rural Water Suppl- and Sanitation SAL - Structural Adjustaent Lending SMC Small and Medium Scale Contractor SNAPE - Service National d'Amenagement des Points d'Eau SPTP - Societe Pilote de Travaux Publics UME - Unite Mobile Ecole USAID - United States Agency for International Development WTP - Willingness To Pay FISCAL YEAR January 1 - December 31 1/ Represents the exchange rate at appraisal in June 1989. The exchange rate is established through an auction system and was GF 615 to the U.S. dollar at negotiations in November 1989. 2/ As of November 1989.  FOR 0MCIL USE ONLY REPUBLIC OF GUINEA NATIONAL RURAL INFRASTRUCTURE PROJECT TABLE OF CONTENTS Pge- DOCUMENTS CONTAINED IN THE PROJECT FILE ........................... i CREDIT AND PROJECT SUMMARY ......................................... ii KEY FEATURES OF THE RURAL SECTOR ....... ............................ v I. INTRODUCTION .. ........................................ 1 II. THE RURAL INFRASTRUCTURE SUB-SECTOR ............................ 2 A. Economic and Sector Setting ......... .............. ........ 2 B. Government Strategy and Policies ........................... 3 C. Detailed Features of the Rural Infrastructure Sub-Sector ... 5 1. Institutions ........ ................. * ......... . 5 0*0 . 5 2. Feeder Roads ................. ...................... 7 3. Bottomlands ........................................ 10 4. Rural Water Supply and Sanitation ................... ...... 13 D. IDA's Sector Strategy and Experience ....................... 15 III. THE PROJECT . ........................17 11. TEPOCT................................................... 1 A. Project Objectives ........................................ . 17 B. Summary Description ...................................... . 17 C. Detailed Features ....... ........ . ................ ..... 19 1. Institutional Development ............................ .. 19 2. Investment and Maintenance Program ...... ............... 22 3. Studies and Pilot Programs ............................. 28 D. Project Cost and Financing .................................. 33 1. Cost Estimates ................ .............. 33 2. Financing .......................................... 34 3. Procurement ........................................ . 35 4. Disbursements and Revolving Funds ...................... 38 5. Accounts, Audits and Reporting ....... .................. 40 This report is based on the findings of an IDA Appraisal Mission which took place from May 29 to June 19, 1989, consisting of Messrs Jean-Claude Balcet (Mission Leader, AF1AG), Donald Drayton (AFTAG), Beyene Wolde-Gabriel (RWSG/WA, Abidjan) and Siaka Bakayoko (AF1AA, Abidjan), Karl-Heinz Fleischhaker and Heinz Neubauer (KfW), Iqbal Qazi (USAID), Yves Laurent (FAC), Sven Hertel (ILO), Jacques Bret (consultant road engineer), and Christian Araud (consultant road economist). This document has a restricted distribution and may be used by recipients only in the perfonmance of their oMcil duties. Its contents may not otherwise be disclosed without World Bank authodzation.  Table of Contents (Cont'd) PMs IV. ORGANIZATION AND MANAGEMENT ............... ............................ 41 A. Project Organization .................. .............. .. .. 41 D. Project Implementation ..... . ..................................... 43 C. Personnel Policies . ................... .................... 44 D. Cost Recovery ................................ ... ... 45 V. BENEFITS. JUSTIFICATION AND RISS .............................. 46 A. Project Benefits .................... ..*.0 ... ... *. 46 B. Economic Justification ........ ............................ . .... . 47 C. Budgetary Implications .......... ... .. ......... ............... 48 D. Employment and Poverty Impact .. ..... ........ ......... .... .*.*..... 49 E. Environzental Impact ............ .. ......................... 49 F. Risks ................I.............. 00. ............. 0. SO VI. ASSURANCES. CONDITIONS AND RECOMMENDATION .o.................... 5. ANNEXES 1. Territorial Organization of the Government 2. MARA's Organizational Structure at Central and Territorial Levels 3. MURA's Organigram (Central Services) 4. Organigram of DNGR 5. Organigram of SNAPE 6. RWS Public Investment Program (PIP) up to end-1988 7. RWS Public Investment Program (PIP) (1989-91) 8. List of Key Positions to be Filled by National Staff 9. Resident Technical Assistance (List of Positions and Terms of Refe.ence) 10. Training Program 11. Feeder Roads Program 12. Bottomland Development Program 13. Summary Account by Project Component and by Year, and Cost of Yearly Routine Maintenance of Regional Rural Roads (1990-2000) 14. Procurement of Goods and Services (Main Lots) 15. Disbursement Profiles 16s Economic Analysis 17. Project Budgetary Impact MAPSs IBRD Nos 21840 - Project Investment Program 21841 - Project Implementation Structures  REPUBLIC OF GUINEA NATIONAL RURAL INFRASTRUCTURE PROJECT Documents Contained in the Prolect File A. BACKGROUND DOCUMENTS AND REPORTS 1. Deuxieme Projet d'Appui aux Services Agricoles, G6nie Rural, Ministere du Developpement Rural/SCET-AGRI, Juillet 1987. 2. Rapport Annuel d'Activit6s de la Direction G4ndrale du G4nie Rural et de l'Hydraulique Villageoise pour l'Annie 1987. 3. R6flexion sur l'Appui Technique a la Direction Generale du Genie Rural, Ministire du D6veloppement Rural/BSD, 13 Janvier 1988. 4. Pistes Rurales - Situation et Perspectives, Ministere du DAveloppement Rural/BSD, 18 Janvier 1988. B. PREPARATI6N DOCUMENTS AND WORKING PAPERS 1. Institutional Develolment 1.1 Organizational Structures 1.2 Training Program (MARA/DNFPR) 2. Investment Program 2.1 Feeder Roads (Preparation Document, SCET-AGRI) 2.2 Feeder Roads (Working Paper SEGMENT) 2.3 Bottomlands (SCET-AGRI) 2.4 Water Supply and Sanitation (RWSG/WA/Abidjan) 3. Studies and Pilot Proarams 3.1 Feeder Roads Master Plan (BCEOM) 3.2 Labor Intensive "Cantonnagew and 'TAcheronnage' components (SEGMENT) 3.3 Pilot Feeder Roads Program f 0) 4. Financial and Economic Analyses 4.1 Project Financial Procedures 4.2 Methodology for the Economic Analysis 5. Work ProRrams 5.1 Detailed Work Program for PY1  - {i - REPUBLIC OP GUINEA NATIONAL RURAL INFRASTRUCTURE PROJECT Credit and Proiect Summary Borrowers The Government of Guinea Beneficiaries: Ministry of Agriculture and Animal Resources (MARA), Rural Engineering Directorate (DNGR), and Rural Water Supply Agency (SHAPE) Credit Amountt SDR 31.3 million (US$40.0 million equivalent) Terms: Standard IDA terms, with 40 years maturity. Proiect Description: .he project would, over a five-year period, assist the Government in implementing policies in rural infrastructure by reinforcing institutions involved in the sub-sector; implementing a specific rural Infrastructure investment program; and formulating, through master plans and pilot programs, coherent frameworks to integrate rural infrastructure operations and appropriate methods to enlist grassroots participation. The project would include three components: (a) support to DNGR and SNAPE, including technical assistance, staff training, renovation of offices and provision ef equipment and supplies; (b) implementation of an investment program, including (i) rehabilitation and subsequent mwintenance of about 2,600 km of feeder roads; (ii) development of about 2,100 ha of small bottomlands; and (iii) improvement in r'ral water supply and sanitation (RWSS) through the drilling of 220 boreholes to be equipped with handpumps, and implementation of two small community piped water supply systems with sanitation components; and (c) preparation of studies and pilot programs including (i) a Feeder Roads Master Plan; (ii) a Pilot Feeder Roads Program to develop labor-based methods for road maintenance and rehabilitation, and to promote the emergence of small and medium scale contractors; (iii) a Pilot Bottomland Development Program to develop bottomlands with the assistance of a NGO; and (iv) studies and pilot actions to define a national strategy for RWSS development.  _ liii - Z of Estimated Proiect Costs a/ Local Foreinn Total Base Cost --__- (US$ million) ---- Institutional Development DNGR Support 2.3 7.5 9.8 14.4 SNAPE Support 0.4 1.2 1.5 2.2 Training 1.8 7.4 9.1 13.5 Investment aMd Maintenance Feeder Roads 10.2 25.0 35.2 51.8 Bottomlands 0.7 1.6 2.2 3.3 Water Supply 0.9 2.9 3.8 5.5 Studies and Pilot Programs Studies 0.2 2.0 2.3 3.3 Pilot Programs 0.8 3.3 4.1 6.0 Total Base Costs 17.3 50.7 68.0 100.0 Physical Contingencies 1.4 3.8 5.2 7.6 Price Contingencies 7.0 8.7 15.7 23.0 Total Project Costs 25.6 63.2 88.8 130.6 (Including Taxes) Taxes 6.3 ,6.3 Total Project Costs 19.3 63.2 82.5 (Net of Taxes) a/ Including three PPF advances totalling US$1.45 million. Financing Plan Local Foreign Total ---------(us$ million)-------- Government 7.9 5.6 13.5 IDA 7.3 32.7 40.0 KfW 1.8 5.0 6.8 USAID 7.5 19.5 27.0 FAC 0.1 0.4 0.5 Beneficiaries 1.0 - 1.0 Total 25.6 63.2 88.8 _~ - Estimated Disbursements from IDA Credit IDA Fiscal Year 1990 1991 1992 1993 1994 1995 1996 --(__----------_-(US$ million)-------------- Annual 3.2 8.8 10.4 8.4 5.2 3.6 0.4 Cumulative 3.2 12.0 22.4 30.8 36.0 39.6 40.0  - iv - Benefits and Risks Benefits. Agricultural productivity cannot increase significantly wbere access roads are impassable and ability to get to the markets is uncertain, and where minimus infrastructure at village level (water supply as a priority) is non-existent. The proposed project aims at resolving such problems sd would have the following benefiss. Institution buildings The project would make the present government policies operational, while strengthening the institutions, including private enterprises, responsible for rural infrastructure development. Rural roads: Benefits are those normally associated with rural roads projects, i.e., acceleration of economic grwvth, reduction of transportation costs, and expansion of commercial services through better communications. Bottomlands: The main benefits would be increased food production and farmer incomes, and land conservation. Water Supl-Y: Improvement of sanitary conditions and living standards would directly benefit the rural population, including women, children and the poor. Risks. The main risk of the project lies in the fragility of governaent institutions following the changes in structures carried out recently. The Government reorganization of ministries could result in a lack of commitment of those civil servants in charge of implementing these reforms. These risks are being addressed as part of the establishment, under the Government's overall administrative reform program, of adequate incentives for civil servants. Technical risks due to shortages of trained staff are minimized through training programs, technical assistance, and a minimal use of force account in carrying out the works. The project was designed to explicitly address the issue of maintenance of the rural infrastructure to ensure its sustainability. Concerning the RWSS component, boreholes would be built in villages that would make substantial financial contributions toward the purchase of handpumps. The risk is that villages may not be willing or able to make such substantial contributions, thereby undermining the establishment of a viable maintenance system. This risk is minimized by the sensitization program to be undertaken as part of the project. A final risk concerns the timely availability of counterpart funding. This risk is minimized, however, by the fact that the Government is taking drastic steps to improve public finance management under the SAL and that participation by the beneficiaries would be sought, thereby decreasing the Government's share of local funding.  - v- RUPIEIC OF uInEA RATIONAL RMRAL PROMC KUY FEATURES OF THE RURAL SECTOR Unit ~~Average U 1. PotuqIon 4unit *-76 i9B urs X of total 3 34.0 7 1Ca Rural Fawillee 011.0 770.0 2. Economic Over-view GP7R/ad of Total Population (cons. 191) I 310.0 804.0 Agric. COP/Head of Rural Dovelopeent t 140.0 165.0 Agricultgeal MP as X of Total x 48.0 40.0 S. Lend Ue GLnCdT-Xo Gq.hm 246.0 246.0 Perennial Crops '000 ha 181.0 211.0 Annual Crop. '040 ha 912.0 1,820.0 4. Are Foreted Dene- Foreat '000 ha 600.0 Wooded Savanna '000 ha 1,700.0 Bush SaVanna '000 ha 12,000.0 Shrub Savanna end F llow '000 ha C,160.0 Mangrove '000 he 200.0 S. Otpu p ATcel 74-75 AW. * 100 100.0 44:0 Fonlo *100.0 lSt.0 Caeeava 100.0 120.0 Coffee 100.0 61.0 F0. PI [zers '00Q ton - 1.0 7. Trade (Current Prices) To-UT Export (FS) OS 1a 0. 6S4.6 of whIcn: - Agricultural Produc'e X 5.4 8.4 a/ - Forest Products G O - Fish Product. ( )p Total Imports * * 267.0 461.7 TOZfO R eurrn* Budget 212.0 251 O O of which: Agriculture X - 1.0 Total Capital Budget I a 174.0 121.0 of which: Agriculture x 17.0 7.0 L sIUnly corfee reported.~ /Govrnment prohibits log and timbor xport. / Excluding extrnally financed project. / Not reported owing to lack of monitoring of Industriol catch.  UiPUBLIC OF GUINEA NATIONAL RURAL INFRASTRUCTURE PROJECT I. INTRODUCTION 1.01 The Government of Guinea has undertaken an ambitious program of economic reforms aimed at reversing the extremely poor economic management which characterized the country from independence to 1984. Under the ongoing Structural Adjustment (SAL) program, the Guinean economy is changing from a highly centralized, inefficient command economy to one where resource allocation decisions are based primarily on amrket signals. The SAL program is designed to lay the foundations of growth through enhancing efficiency in production and in public sector management in the context of a new role for the state in promoting development. The proposed National Rural Infrastruc- ture Project is consistent with the general thrust of this program. It would, in addition, contribute to lifting one of the major technical constraints to agricultural growth, namely the dilapidated state of rural infrastructure. 1.02 The SAL program, which entails a radical departure from the policies of the past and requires a systemic change, has been supported by twc structural adjustment creaits, a Special African Facility, and two annual arrangements under a structural adjustment facility. Its main elements are: (a) withdrawal of the state from directly productive sectors and restructuring of the public sector through a comprehensive administrative and public enterprise reform; (b) resource allocation policies consistent with the establishment of a flexible, liberal economy, through inter alia, realistic exchange rate, interest rate, price and trade policies; and (c) macroeconomic policies consistent with long-term fiscal and balance of paymen-. equilibria. 1.03 The reform program has already generate. -4bstantial improvement in economic performance emanating principally ft. - the private sector with significant increases in agriculture, construction, transport and trade activities. Weak institutions and the administration's limited capacity for economic and financial management, however, continue to hamper the full realization of Guinea's growth potential. During the second phase of the program, the Government aims to alleviate these constraints in order to deepen and consolidate the ongoing process of adjustment. This is supported by the Second Structural Adjustment Credit (SAL II, Cr. 1926, FY88) and the Economic Management Support Project (EM4SP II, Cr. 1963, FY89) as well as by the Private Sector Promotion Credit whi5ch is being appraised. These operations are designed to pursue the refozm of public sector management and at the same time strengthen the incentives environment for private sector activities. 1.04 In parallel to the SAL program, the Government is also aiming, under its Public Investment Program (PIP), at alleviating a major constraint to  growth constituted by the extreme dilapidation of the country's infrastructure inherited from the First Republic. Rehabilitation of infrastructure in the primary transport sector (trunk roads, ports and airports), and in the urban and industrial sectors is being addressed by Government in the context of various projects financed by IDA and other donors. Rural infrastructure rehabilitation and development are particularly critical to the growth of agriculture in Guinea, but have until recently been addressed only in the context of ad hoc operations and area-based projects. This fragmented approach results in duplication of resources and inefficiencies, and an analysis has determined that it is not sustainable. Government has therefore prepared a national rural infrastructure strategy and requested IDA's assistance to support its implementation. 1.05 The proposed project would assist the Government in implementing the SAL policies in the rural infrastructure sub-sector in particular, by (a) strengthening the capacity of the concerned institutions to plan and oversee the execution and maintenance of rural infrastructure; and (b) promoting private sector development through participation of small and medium scale contractors and grassroots organizations in building and maintaining rural infrastructure. At the same time, as an important element of the PIP, the project would substantially contribute to lifting the infrastructure constraint in rural areas by implementing a critical mass of feeder roads, bottomlands, and water supply and sanitation investments. In line with the Government's objective to withdraw from directly productive activities, this would be achieved mostly through reliance on work execution by private contractors. 1.06 The proposed project was prepared from October 1988 to April 1989 by the National Directorate of Rural Engineering (DNGR) and the Rural Water Supply Agency (SNAPE) of the Ministry of Agriculture and Animal Resources (MARA) with the help of consultants. An amount of US$400,0(;0 equivalent was provided by IDA to help in this preparation under a first advance of the Project Preparation Facility (PPF) approved in August 1988. A second PPF advance approved in December 1988 served to finance a Rural Roads Master Plan which was used as the basis of the project rural roads component. IDA preparation missions visited Guinea in March and October 1988. The project was pre-appraised in February 1989 and appraised in June 1989 by a mission including representatives from KfW, FAC, and USAID who have agreed to cofinance the project. II. THE RURAL INFRASTRUCTURE SUB-SECTOR A. Economic and Sector Settint 2.01 The Rural Sector. The rural sector is the mainstay of Guinea's economy, providing a livelihood to over 75Z of the population and generating about 302 of GNP. Agricultural production is predominantly based on smallholder farming and largely subsistence-oriented. Approximately 1.5 million ha Is cultivated annually by about 700,000 familie6. While this  represents about one quarter of total arable land, the practice of leaving land fallow implies that land capacity has nearly reached its limits, and that further growth will have to result from intensification of agriculture on the better soils rather than from extension of cultivated areas. 2.02 The favorable climate and agro-ecological diversity of Guinea offer a wide range of possibilities fo- growth in the rural sector. Despite this potential, agricultural performance has been disappointing during the past three decades, resulting in low rural incomes (US$150 equivalent per year and per capita on average) and an alarming urban migration. The failure to develop the agriculture sector stemed largely from distorted nacroeconomic conditions, inadequate policies, ineffectual institutions and dilapidated infrastructure. The economic, policy-related and institutional constraints are being satisfactorily addressed by the Government under the SAL program. Development of the infrastructure sector will be a critical factor in realizing the benefits expected from these policies. 2.03 The Rural Infrastructure Sub-Sector. Rural infrastructure (mainly feeder roads, irrigation and drainage works and rural water supply and sanitation - RWSS) is relatively undeveloped in Guinea compared to neighboring countries (para 2.19) and in severe disrepair. This situation is the combined result of the historical past of the country, difficult terrain, and the inevitable time lag rejuired to implement the newly established government policies. 2.04 Lack of satisfactory rural infrastructure is a major constr.int to Guinea's development. Without infrastructure, agricultural production can not be stored and evacuated, inputs can not be transported, practically no commercial activity can take place, and the sector production potential and comparative advantage remain undeveloped, with farmers unable to shift away from subsistence production. 2.05 Investments in infrastructure represent about 402 of the total capital expenditures programmed under the PIP for 1989-91. The lion's share goes to the public works and transport uub-sectors, which account for 30S of total PIP disbursements. The proposed project would represent about 62 of investments in these sub-sectors. The above situation vividly demonstrates the priority given by the Government to continue rehabilitating the country's basic infrastructure, particularly rural infrastructure. B. Government Strateav and Policies 2.06 Strategy. the Government's strategic objective in the rural sector is to improve the standards of living of the rural population by increasing rural incomes and ensuring that the population's basic needs (food security, health and education) are met and essential services are available. The Government's strategy to meet this objective, supported by IDA, is to (a) implement the macroeconomic reforms agreed in the SAL and in sector adjustment programs, in particular to create a functioning market-oriented economy; (b) provide agricultural services to the farming community. especially applied research, extension, and input supply; (c) improve rural  - 4 - infrastructure; and (d) develop the capacity of the Guinean public and private institutions to provide services to the rural sector (mainly in planning, strategy formulation, investment selection, implementation and management of investment programs in government departments, and private services in all sectors). 2.07 The development of agricultural research and extension, including related institution building and staff training, is being addressed under the National Research and Extensio, Project (PNRVA, Cr. 1955, 1988). The proposed National Rural Infrastructure Project addresses the rural infrastructure sub-sector. 2.08 Policies. The main aspects of government policies applicable to the rural infrastructure sub-sector are consistent with the general policy orientation described above. They are summarized below: (a) the Government will progressively withdraw from services which could be provided efficiently by the private sector. Consequently, in the context of the project, the amount of force account work performed by government services (especially DNGR and SNAPE) would be kept to a minimwm; (b) beneficiaries will participate in the selection of rural investments and contribute financially and in kind to the maintenance of infrastructures, and when possible, to their construction. This policy aims at ensuring the sustainability of investments by generating commitment of beneficiaries and recovering most of the costs; (c) the Government will integrate externally financed projects into national programs, and through this integration, will progressively reduce the number of projects being implemented in the agriculture sector. This policy aims at reducing duplication of resources and at ensuring the sustainability of actions undertaken in the context of individual projects. It does not exclude the financing of projects t.. :geted to specific regions by some financing agencies; and (d) foreign experts will emphasize the training of national staff in order to progressively reduce the need for such experts. This policy aims at the sustainability of institution building actions implemented under various projects. The project general objective is to assist the Government in implementing these policies.  -5- C. Detailed Features of the Rural Infrastructure Sub-Sector 1. Institutions 2.09 In the context of the proposed project, the term rural infrastrucLure' applies to investments made outside the main urban centers under the jurisdiction of the Rural Engineering Directorate (DNGR) (i.e., rural roads, irrigation and drainage works) and of the Rural Water Supply Agency (SNAPE) both within the Hinistry of Agriculture and Animal Resources (MARA). National roads (i.e., trunk roads) are under the responsibility of the Ministry of Transport and Public Works (MTTP), as described in para 2.18. Responsibility for rural investments in the health and education sectors is vested in the ministries of Health and of Education respectively, but these ministries might delegate some works in the future to DNGR on an ad hoc basis. Only infrastructure development under the responsibility of MARA is included in the proposed project. Other infrastructures, such as national roads in rural areas, rural schools, and health facilities are developed outside the project by relevant public institutions. In some cases IDA provides projects in support of such infrastructure. 2.10 Administrative Structure and Institutional Reform. The country is divided into four administrative regions which correspond to the four natural regions of Guinee Maritime, Moyenne Guinee, Haute Guinee and Guin4e Foresti6re, and 33 spr4fectures". The latter are in turn divided into sous- prefecturess and these into districts which form the basic administrative units of the country. 2.11 An administrative reform has been implemented in the context of the SAL program to resolve the problem of staff redundancy at lower levels, combined with a shortage of adequately trained staff at both management and technical levels. As part of this reform, most civil servants have been formally tested, organizational charts for the main ministries have been prepared (including a quantification of authorized positions), and the organization charts of central ministerial structures have been completed. Todate about 30,000 staff have been laid off or retired from the civil service. The final objective is 45,000. 2.12 Similarly, recent decrees have defined the organization and the assignments of the territorial administration ("administration d6concentr4e") i.e., services of the State geographically located in the four regions and the prefectures and sub-prefectures (Annexes 1 and 2). The legal texts reinforce the administrative responsibilities entrusted to the four Resident Ministers, who are government members residing in the regions and considered the direct representatives of the President of the Republic. They also give the prefects an expanded role in the coordination and control of technical services regrouped in six 'directions pr4fectorales'. At the same time, however, the texts lay the foundation for a future decentralized administration which will give more weight to scollectivites d4centralisees" (decentralized collectivities) which will have freely elected leaders at  village and regional levels. This is particularly significant and important for the proposed project which aims at mobilizing grassroots participation. 2.13 The main institutions directly involved in the rural infrastructure sub-sector are the followings (a) the Rural Engineering Directorate (DNGR) is part of MARA and would be one of the two major bodies implementing the project. DNGR is in charge of policy formulation in the rural infrastructure sub-sector, and of planning and managing the implementation of national rural infrastructure programs. The structure of MARU is shown in Annex 3. The organization of DNGR, including its central and regional services, is shown in Annex 4; (b) the Rural Water SuDDlY AQencY (SNAPE) is a public agency in charge of implementing rural water supply (RWS) programs, mostly through private contractors, and follows the general government policies summarized above. SNAPE is a relatively efficient agency and has benefited from the support of various donors. As a result of the institutional reform, SNAPE was recently incorporated into MARA, but continues to operate with a large degree of autonomy. The structure of SNAPE is summarized in Annex 5. Under its proposed new statutes, it weruld become an "Organisms PersonnalisA', i.e., a state agency having legal personality with financial and administrative autonomy (para 4.03); and (c) the Ministry of Transport and Public Works (MTTP) is in charge of all activities concerning the national road network. Reorganization of MTTP and an investment program In the transport sector are underway in the context of the Transport Sector Project and the Fourth Highway Project Cpara 2.48). Arrangements to coordinate the activities of MTTP and DNGR are described in para 2.18. 2.14 Construction Industry. In the past, large scale construction works were carried out almost exclusively by foreign contractors, because domestic construction contractors were extremely small and disorganized. However, as a result of this exclusion, domestic contractors had no opportunity to emerge. Maintenance wores were done by force account, in particular for road works. Since the establishment of the new economic framework by the Government, some medium scale contractors have appeared, coming mainly from neighboring countries. These firms have expanded their operations in Guinea's building sector and have started competing with large foreign firms for road rehabi'itation and maintenance, and other construction work. There are small Guinean contractors for the construction of relatively small buildings, an activity which is thriving in the main towns, but the workmanship is usually substandard. 2.15 To promote the construction industry and help create an intermediate class of small scale contractors, the following actions are proposeds (a) international contractors would be provided with information on Guinean contractors and would be encouraged to subcontract certain  -7- portions of the works (e.g., drainage systems, small bridges, culverts, etc.) to small national private contractores (b) some items of work (e.g., small road rehabilitation lots of 30 km or less) would be specified in such a way that they would appeal to small local private entrepreneurs; (c) the preference clause of 7.5X provided for local contracting under Bank procedures would apply; and td) some sections of roads to be rehabilitated would be used to train Guinean contractors (para 3.45). 2.16 As the business climate in the country improves, there will be good potential for the establishment of a small and medium scale construction industry. Under the SAL program, the Goverfment aims at supporting small industries, including construction industries, by (a) revising and easing regulations applicable to them which have restrained development of the industry in the past; (b) improving the efficiency of support institutions; and (c) improving the availability of credit to local contractors. The Pilot Feeder Roads Program, to be undertaken as part of the proposed project (paras 3.36 - 3.46), would supplement these efforts. 2. Feeder Roads 2.17 Terminoloiv. Road Classification and Administration. No consistent terminology concerning the various kinds of roads is used in Guinea because no official system of road classification exists in the country. However, such a system is under preparation in the context of the Transport Sector Project. In the present report, the following terms are useds (a) the National Road Network of about 4,000 km (including 1,300 km of paved roads, connecting the capital, Conakry, to most prefectures and to the main border points and sea ports); (b) the Regional Road Network of about 3,000 km, linking isolated remaining prefectures to the national network, prefectures to sous- prefectures, and prefectures to secondary border points; and (c) the Rural Road Network of about 11,500 km, consisting of all other roads. This network includes roads connecting the subprefectures to villages and areas where agricultural production or other activities take place. Most of these roads are impassable during part of the year. Roads under items (b) and (c) above are subsequently referred to as 'feederw roads. 2.18 The MTTP is theoretically responsible for the national and regional networks, but de facto delegates responsibility for the latter network to MARA. MARA (through DNGR) is therefore responsible for the planning,  construction, rehabilitation and maintenance of all feeder roads, i.e., regional and rural roads. The rationale for DNGR's constructing and maintaining regional roads in Guinea at present is the linkage with agriculture. Since the interior of Guinea is so much behind in terms of economic and infrastructure development, the regional roads are still mostly used for agricultural traffic. Therefore, policies and investments concerning the regional network largely affect agriculture. As the economic fabric develops in the interior, the situation is bound to evolve, with the regional roads gradually playing a larger role in support of broad-based economic activities. As a result, MTTP should gradually be given responsibility for these roads. To this end, it should undertake a commensurate expansion of its operational capacity which is presently insufficient. Assurances were obtained at negotiations that the present delineation of responsibilities, wherein MTTP is responsible for the national and regional networks and MARA for the rural roads network, be made official with the proviso that MTTP would delegate its responsibility for the regional network to MARA over the medium-term. In the meantime, HTTP and MAU would coordinate the planning and execution of their respective work programs. In particular, MTTP would be associated to the programming and supervision of activities concerning the regional network carried out under the proposed project. Assurances to this effect were obtained at negotiations. 2.19 The Road Network. Guinea has a road network presently 'passable" (in the sense that vehicles are able to use them) totalling about 12,000 km. They are classified as indicated in para 2.17 (i.e., only 5,000 km of feeder roads can be considered "passable' at present, among the total of about 14,500 km of feeder roads identified under the Master Plan prepared as part of project preparation (para 3.35)). In terms of road density, Guinea is one of the more poorly served countries in Africa, with 4.3 km of roads per 100 km2 and 1.8 km per 1,000 inhabitants, compared with 11.9 km in Ghana, 13.6 km in Cameroon, and 14.8 km in COte d'Ivoire. 2.20 The extremely poor condition of the network is a major constraint to the movement of produce along the coast and from the hinterland to Conakry; in some instances these flows have stopped altogether and large areas of the country have become completely isolated. These conditions jeopardize government efforts to develop a productive market-oriented agriculture as part of the newly liberalized economy. 2.21 Rehabilitation of the national network has been initiated under the IDA-financed Third Highway Project; it is being pursued under the IDA- financed Transport Sector and Fourth Highway Projects (para 2.48) and under several other projects, particularly in Guinee Forestiere, with financing provided by the Saudi and Kuwaiti development funds. Various ongoing projects provide for the rehabilitation and construction of feeder roads in specific locations. Such projects includes the IDA-financed Gueckedou Agricultural Development Project (PAG) and the FRG-financed Kissidougou Integrated Rural Development Project (DERIK) in Guinde Forestiere; a CCCEIEEC-financed cotton development project and the IFAD/AfDB-financed Siguiri Rice Project (ORS) in Haute-Guinee; the IFAD-financed Area Development Project and the UNCDF/UNDP-financed Fouta-Djallon Integrated  Development Project in Moyenne Guinee; and various small operations scattered across the country. The works undertaken in the context of these specific projects are unlikely to be maintained and kept operational after the end of these projects. Therefore, in accordance with government policy to integrate all projects into national plans (para 2.08(c)), DNGR would progressively assume or formalize the delegation of responsibility for the maintenance of these roads. 2.22 Traffic and Road Transport. Due to inadequate data, it is difficult to determine the size of Guinea's over-aged and heterogeneous vehicle fleet. The existing fleet is adequate to meet present transport demand, but when economic activity picks up, a considerable increase in transport capacity will be required. At present there is little transport on most of the regional and rural road networks, because most existing roads are impassable during parts of the year, and transport costs are prohibitive due to the extremely poor conditions of the roads. Project preparation studies estimate that, once the roads become passable and are properly maintained, traffic will increase and transport cost will decrease dramatically. 2.23 Execution of Road Maintenance. Maintenance of -ural roads is severely neglected due to lack of public funds. However, more often than not, some critical works, such as the repair of bridges and the filling up of washed- away road sections, have been carried out through informal arrangements between DNGR and local communities, DNGR supplying civil works equipment and the local communities providing fuel (and controlling its use). There are also instances of bridges built with little or no government assistance. Although such arrangements are relatively rare dnd have not yet been sufficiently encouraged by DNGR, they demonstrate vividly the interest of some communities in improving their roads. 2.24 With the improvements expected under the project, the need for capabilities in road maintenance would further increase. Although DNGR would retain some limited force account capability on a temporary basis to carry out emergency repairs, all major maintenance works would be contracted out. Periodic maintenance would be carried out by local medium scale contractors, whereas routine maintenance would be entrusted to laymen and small contractors (Itacherons') (para 3.26). 'Cantonnage" schemes, in which maintenance is undertaken by local paid labor responsible for specific sections of road, would be tested under the project (para 3.43). 2.25 Funding of Road Maintenance. Funding for road maintenance is presently inadequate. This results in most roads--feeder roads especially-- being left unmaintained and deteriorating, with attendant negative consequences on agricultural production. The issue has to do mostly with the slowness of disbursement of funds from the Government's general budget. IDA recently indicated to the Government that the issue could be best addressed by means of a budgetary line item for road maintenance, subject to an accelerated disbursement procedure (provided for within the framework of existing government procedures), whereby funds would be made available each quarter and in advance to a special account at the Central Bank, for expected expenditures (para 5.09).  - 10 - 2.26 Road Planning. Under the Third Highway Project, a small Transport Planning Unit was created in late 1985 within the former Secretariat of State for Transport, to start the data base required for a medium-term transport plan. As part of the Fourth Highway Project, an indicative road network development plan was established. This plan includes investments required to build, upgrade and improve the paved, gravel and earth roads of both the national road network and the regional road network. It served as the basis for a donors meeting on the transport sector which took place in Conakry in April 1988. 2.27 The planning process for feeder roads undertaken by MAP.IDNGR has evolved as followst (a) in the context of preparation studies for projects in the agriculture sector in 1986 and 1987, consultants established a rural roads program of 1,500 km to be rehabilitated over five years and submitted to donor agencies, including IDA (SCET-AGRI report, Project File No. 1); (b) following a donors meeting on the agriculture sector in October 1987, DNGR carried out further studies, mostly synthesizing existing information, and MARA commissioned a study to prepare a Master Plan for the development of feeder roads; (c) the Master Plan was initially entrusted to the MARA's Central Planning Unit (Bureau de Stratdgie et de Developpement - BSD). However, following curtailment and reorientation of BSD activities in October 1988, it was decided that this plan would be undertaken as an integral part of the project preparation activities, under a second PPF advance of US$300,000; and (d) the Master Plan was completed in April 1989, and is being made operational under the project. It is based on data for 14,500 km of roads and provides for the ranking of a sublet of 8,000 km of these roads. It is the framework for the project investment program. 3. Bottomlands 2.28 General Characteristics and Taxonomy. Bottomlands are natural depressions which result in woodlands which themselves are characterized by a relatively shallow valley with a flat floor. Wet conditions throughout, or at certain times of the year, are essential features, and these conditions can be caused by flooding or a high groundwater table. Bottomlands are among the potentially more productive lands in Guinea when properly utilized. The main favorable factors aret (a) relatively flat topography which permits a wide range of cultivation methods including animal traction; (b) soils of moderate fertility on which a number of crops can be grown; (c) water availability (surface and ground water) allowing longer cropping seasons; and (d) good potential for intensifying agriculture in this ecological zone.  - 11 - 2.29 The bottomlands of Guinea can be broadly divided into two categories based on sizes small bottomlands (less than 50 ha) of which there are about 60,000 ha, with about 2,000 ha under some form of cultivations and relativoly large plains (ranging from 50 to 1,000 ha) which total about 100,000 ha and are undeveloped for the most part. In both ecosystems, water depth varies considerably based on a number of factors, such as size of watershed, duration and intensity of rainfall, natural drainage, and other sources of water. Bottomlands are fcund in all four ecological zones of the country. A large number of crops (annuals and perennials) are grown on these bottomlands, and livestock graze on them particularly in the dry season when vegetation is scarce elsewhere. The main crops grown are rice, vegetables, bananas, maize, groundnuts, sweet potatoes and cassava. 2.30 Rationale. Objective and Scope for Bottomland Develogment. The development of small bottomlands is of particular interest because they are well suited for simple water management techniques and their small size makes them manageable by local communities. Many of these communities have on their own initiative been exploiting bottomlands for many years. However, with the exception of recent externally funded projects, very little research and extension have been available in support of improving bottomlavd cultivation techniques. Also, the utilization of bottomlands has been viewed largely in relation to increasing rice production. The development of the large plains is not a priority because of the technical and managerial complexity inherent to this kind of development, and the present lack of implementation capacity in the village communities. In addition, the technical and economic feasibility of developing large plains varies and should be assessed on an individual basis. The proposed project would therefore focus on the development of small bottomlands, giving priority to those in which exploitation has already started. 2.31 To fully exploit small bottomlands, a corbination of agronomic and infrastructure improvements are necessary. Agronomic improvements would include water management techniques, improved land preparation methods (including animal traction), improved varieties suited to existing conditions, improved seed, timely planting activities to facilitate multiple cropping, and selective use of fertilizers and agro-chemicals for crop protection. Typical infrastructure improvements include reshaping water drainage courses, in-field bonding to permit retention of sufficient water for rice cultivation, and building small control structures to limit draw- down of the water table at the beginning of the dry season. 2.32 The *Bottomland Comilex'. Bottomland components are included in several ongoing projects. The IDA-financed Gueckedou Agricultural Development Project in particular has successfully improved cultivation techniques in bottomlands and has introduced the concept of developing the *bottomland complex" which includes adjacent slopes, in order that farmers and their f' milies concentrate agricultural activities around the bottomland. This approuch encourages the replacement of the slash-and-burn itinerant form of cultivation with a stable improved system which can be sustained by the farmer. It therefore reduces the ecological damage that is common in the  - 12 - slash-and-burn system presently practiced in the forest zone. These techniques are directly replicable, and the Government has indicated its intention to adopt this approach for all projects with bottomland components in forest zones. 2.33 Land Tenure. Land tenure plays an important role in the utilization of land, and the Government policy has in the past left problems which could affect the exploitation of some bottomlands. Legally the land belongs to the state. However, villagers have the right of use, such right being attributed to the first occupants of the land, who can transfer these rights for parts of their property to others without losing their own priority rights. Almost all land hae been attributed to families and, therefore, there is hardly any free land available. Land distribution between families of the same lineage is the responsibility of the head of the larger family group, without committing a definitive division of the attributed family land. Arrangements for the use of a plot of land by a non-family member can only be made with the approval of the head of the family. Generally, these arrangements exclude the establishment of perennial crops and are of short duration for the cultivation of annual crops. During the previous regime, disruptions in this system occurred through appropriation by the Government of agricultural lands (including bottomlands) for such projects as the state farms for resettling young agriculturists (FAPAs). Since the dissolution of these projects, land, which in some cases had been allocated to government workers, is a continued source of dispute because traditional rights have not been restored. 2.34 MARA has commissioned the drafting of a new land law which is to be based on the recognition of customary rights and should lead to tCe evolution of private ownership. However, the Government has taken no final decision about the principles on which land rights are to be based, nor has it decided which land the state will retain as its property. To address this issue, the TDA-financed Forestry and Fisheries Management Project (FY9lL) signed in November 1989, includes a pilot land tenure operation which aims at strengthening MARA's capacity to identify and record rural land rights and prepare an inventory and registry for selected areas, particularly in the forest zone which requires protection from increased encroachment. 2.35 In the case of bottomlands, problems of land tenure are most likely to be found in relation to those in or near towns or large villages, where pressure to exploit is greatest. At present, the prefects resolve these issues, and in some cases, have set up commissions to re-allocate land in bottomlands previously held by the Government. The project would concentrate on bottomlands already in use, but would ensure, through a system of consultation with village heads and farmers associated with the cultivation of the bottomlands, that no disputes exist over land tenure ilghts where project intervention is expected. This approach would be further aided by the creation of farmers associations as proposed in para 2.36 below, the basis of which would be the mutual exploitation of bottomlands by members. 2.36 Farmers Associations. Farmers cooperatives in Guinea date back to colonial times, particularly in Guinde Maritime where they exported as much  - 13 - as 100,000 tons of fruit. Production and export s drupped dramatically in the post-colonial period and have remained negligible since then. The late government-imposed substantial changes in the cooperative movement in keeping with its policy of collectivization gave rise to the creation of the PAPAs (para 2.33). Many of the old farmers cooperatives continue to exist although only in name, since commercial activities have long ceased. A new approach would be required to restart cooperatives in these areas. Elsewhere in Guinea, various forms of traditional work groups exist. In the Forest Region, these groups consist of 7 to 10 farmers of approximately the same age, who carry out essential agricultural and other activities on each other's farm on a rotational basis. In the Gudckedou Agricultural Development Project, these traditional groups are being used as the base for the formation of pre-cooperative groups to faciiitate the diffusion of extension messages, for credit and inputs distribution and management, and for developing bottomlands. In Coyah, under the direction of technical assistAnts furnished by a French NGO, the Association Frangaise des Volontaires du Progres (AFVP), a four-hectare bottomiand is being cultivated by farmers who have formed an association for the production and marketing of their produce (rice and vegetables). They have established a revolving fund for purchasing inputs and decide collectively on crops to be grown. 2.37 The Government's policy is geared towar( encouraging the free establishment of farmers cooperatives. A National Cooperative Support Agency (SNATC) has been established, which would be responsible for providing guidance to new and old cooperatives whose activities are expected to include procurement and distribution of inputs, marketing, credit and the mobilization of members savings. However, its statutes id to be revised as they do not provide for sufficient operating autonomy. nding has been provided under the Guinea National Research and Extension Project (Credit 1955-GUI, FY89) to assist SNATC in developing its activities. In the development of bottomlands, cooperatives can be a vehicle for carrying out improved water control measures, for which coordination is important. Other benefits would be those related to inputs, marketing, etc. The project would, therefore, actively seek SNATC assistance in promoting cooperatives among bottomland farmers. 4. Rural Water Supply and Sanitation (RWSS) 2.38 Well endowed with annual rainfall that varies from 1,200 to 4,000 mm, Guinea has aDundant surface water which is often perennial. There are also many springs in mountainous areas (Fouta Djallon and to a lesser extent Guinee Forestiare). The water supply problem in terms of quantity is only acute in the northernmost part of Moyenne Guinee and Haute Guinde. The problem of RWS in Guinde Forestibre is not one of quantity but quality of water. 2.39 Three-fourths of the rural population still relies on water from traditional wells, ponds and streams which is often contaminated. At the end of the dry season about 302 of the rural inhabitants typically has to travel long distances to get water, and per capita consumption drops sharply (as low as 5 liters per capita per day--lcd). Moreover, rura.l sanitation practically  - 14 - does not exist. The Ministry of Health estimates that about 50? of all morbidity and mortality rates (which for infants and children are among the highest in Africa) is associated with waterborne disease and lack of sanitation. The Government has therefore decided to provide for inclusion of sanitation componeuts as integral parts or complements to RWS projects. 2.40 To meet its RWS requirements, Guinea has established a program in three phases: (a) phase 1 (1979-1995) will provide 10 lcd to 55? of the rural population by 1995, by creating 6,500 water points for villages with population over 300; (b) phase 2 (1996-2000) is designed to provide the same amount of water supply (10 lcd) to smaller rural settlements by creating an additional 5,700 water points by the year 2000: and (c) phase 3 will aim at supplying 20 lcd to the rural poprlation by constructing an additional 10,000 vater points. As of end-1988, 2,300 rural water points had been created under Phase 1 (Annex 6) at a cost of US$20 million (half of these are boreholes equipped with handpumps, one-third is wells partly equipped with handpumps, and the remainder 17 is spring catchments). This has permitted to serve about 1.25 million rural dwellers, i.e., 23Z of the total rural population of 5.4 million. Annex 7 gives a list of RWSS projects included in the Public Investment Program fox 1989-90. With respectively 47Z and 40? of their Phase 1 objectives achieved, Hoyenne Guinee and Haute Guinee are the two better served regions. 2.41 Guinee Forestiere is the region that has least benefited from Phase 1 of the program: 145 water points have been constructed in o;.e prefecture, representing only a fraction (2.22) of the national objective. About 60? of the remaining villages, i.e., 1,100 villages, are ln the project area (Kissidougou and Gudck4dou prefectures), with a population of about 100,000 each. There are several markets of regional importance in these two areas, particularly in the towns of Kissidougou and Gueck0dou snd the villages of Yend0-Millimou and Nongoa. Several businesses in Conakry have branches in Kissidougou and Gu0ck0dou. 2.42 Information on the sanitary conditions of the population in the area, even though partial, shows the omnipresence of bilharzia and hookworm disease, and a high incidence of tuberculosis and leprosy. In general, the sanitary conditions of the villages in Guin6e Forestibre are much more precarious than in the other regions of the country. Even the larger villages, where the need for sanitary means of excreta disposal is great, have only small traditional latrines built by individual households. 2.43 More than 902 of the villages is situated on hill tops or plateaus. The water sources located at lower levels are rarely farther than 500 m from the villages and exceptionally more than 1 km. Drawing water requires little effort because the water level even in wells is close to the ground surface. Carrying water, however, is a strenuous task for women, who have to walk on steep and slippery footpaths. 2.44 Villagers obtain their water chiefly from non-protected surface water (backwaters, ponds and springs), but sometimes traditional wells are also used, particularly in Kissidougou. Because most water points are perennial  - - 3 - and have a reduced yield during the dry season, the waiting time is more than one hour. The water consumed in the area is often contaminated. Thus, the problem of water supply in most of Guinee Foresti&re and in the project area is of quality and not of quantity. In this regard, even though the potential for wells and to a certain extent for springs is good, priority should be given to the drilling of boreholes. These boreholes should be located in or near villages, so as to ensure that they will be used and maintained. Consistent with the Government's objective to limit expansion of force account vork, the drilling of the boreholes under the pr)ject would be done by private contractors. Similarly, the private sector should be encouraged to undertake the construction of wells and development of springs. To this effect, the feasibility of promoting and supporting small and medium scale contractors would be studied under the project (para 3.51). 2.45 Because the area has not yet benefited from any RWS project, villagers are not accustomed to using lined wells, boreholes or pumps. Therefore, the advantages of clean water may not be immediately evident to them. In the prefecture of Kissidougou, however, there are indications that the villagers' attitude toward improved water supply may be changing. In the framework of the DERIK project (para 2.21), for example, they have accepted to pay significant amounts (up to GF 72,000 per well) for the digging of a few open wells. D. IDA's Sector -_rateay and ExDerience 2.46 IDA's assistance ItAn Guinea's rural sector has emphasized raising smallholders' incomes while creating durable institutions in both the public and private sectors, basically supporting the Government's strategy outlined in para 2.06. The improvement of rural infrastructure plays a major role in implementing this strategy. At the macroeconomic level. IDA has supported economic and financia'l reforms aimed, inter alia. at pzoviding incentives to agricultural production, through SAL I (IDA Cr. 1659-GUI and the Special Facility for Sub-Saharan Africa, Cr. Alil-GUI of February 1986) and SAL II (IDA Cr. 1926-GUI of June 1988). 2.47 At the sector level, IDA has financed the following projects, which include substantial infrastructure components: (a) the Rice DeveloDment Proiect (Cr. 952-GUI, US$10.4 million, September 1979) promoted rice development in Guinee Forestiere; (b) the Livestock Development Proiect (Cr. 1063-GUI, US$21.1 million, September 1980) was cancelled in September 1983 due to the Government's failure to take adequate policy measures; (c) the Gu4ckedou Asricultural DeveloDment Proiect (Cr. 1635-GUI, US$6.7 million, November 1985) was built upon the success of the previous project and pioneered techniques in bottomland development (para 2.31). It has progressed up to the point where, consistent with the Government's policy to set up programs of national scope in each major sub-sector, the management of the various components is being  - 16 - integrated into the framework provided by corresponding national projects, including the proposed project; (d) an %aricultura' Services Pro4,ect (Cr. 1636-GUI, US$7.5 million, November 1985) was designed to improve NARA's capability in planning, policy formulation, and project analysis, through the establishment of a plannirng office (BSD). The project has had mixed results. It has recently been scaled down and reoriented; (e) the Livestock Sector Rehabilitation Pro1ect (Cr. 1725-GUI, US$9.8 million, July 1986) includes most of the components of the previous livestock project, including rural infrastructure related to livestock development. After a slow start-up, the project is now progressing satisfactorily; and (f) the National Research and Extension Proiect (Cr. 1955-GUI, US$18.4 million, Sentember 1988) is establishing a nation-wide research and extension service. 2.48 Other infrastruc,.ure projects with direct impact on the rural sector include investments in the transport sector, for which IDA has been miong the principal sources of external financing. These projects include: (a) the First Hiphwav Proiect (Cr. 596-GUI, PY75, US$14 million). The project was satisfactorily completed in 1980; (b) the Second Hiphwav Proiect (Cr. 853-GUI and EEC-33 GUI, FY79, US$17 million). The project institution-building objectives were not achieved, but after strong remedial action, the project was completed in 1984 and yielded a satisfactory rate of return; (c) the Third Highway Prolect (Cr. 1457-GUI, May 1984, US$28 million); this ongoing project was prepared during the previous regime and designed for a different institutional and economic environment. A major feature was the establishment of an autonomous Roads Project Office (OPR). At the onset of the Second Republic, the project was redesigned. to reorient OPR's structure toward that of a commercial contracting enterprise. OPR is now being privatized as part of the Transport Sector Project; (d) the TransDort Sector Prolect (Cr.1815-GUI, June 1987, US$55 million) provides a policy framework to ensure efficiency and effectiveness of public and private institutions operating in the sector; and (e) the Fourth Hiahway Proiect (Cr. 1915-GUI, June 1988, US$55 million). Rehabilitation programs included in the Third and Fourth Highway Projects focus on national roads which form the backbone of the feeder roads network to be rehabilitated under the proposed project. 2.49 The project would be IDA's first involvement in the RVSS sub-sector in Guinea. Previous projects in the water supply and sanitation sector are  - 17 - the Conakry WaterlSanitation Project (Credit 870-GUI. FY80) and the recently approved Second Water Supply Project. Through the latter, the Bank would continue to support a major reorganization of the urban water supply sub- sector that would involve partial privatization of operations in line with the Government's policy to increase participation of the private sector in the provision of services. 2.50 Finally, IDA, under a US$250,000 SPPF, i' assuming leadership of a group of donors (the main ones are USAID, -ne Canadian Agency for International Development, and EEC) in assisting the Government to design an Environmental Action Plan (EAP). As part of this process, an Environmental Advisory Group has been created at the Planning Ministry (MPCI). The Guinean official who would be nominated at DNGR to deal with the environmental impact of feeder roads as part of project implementation (para 5.12) would liaise closely with this advisory group in making the required environmental assessments. III. THE PROJECT A. Proiect Obiectives 3.01 The project general objective would be to strengthen the Government's ability to implement policies in the rural infrastructure sub-sector, as described in para 2.08. Specific objectives would be to: (a) reinf rce the capacity of institutions involved in the sub-sector; (b) implement a specific rural infrastructure investment program; and (c) formulate master plans for further development of rural infrastructure operations, and implement pilot actions to test new methods of construction and maintenance of rural infrastr"cture, particularly actions involving grassroots participation. B. Summary Description 3.02 Proiect Comnonents. To meet the above objectives, the project would include the following componentss 1. Institutional Development (a) support to the Rural Engineering Directorate (DNGR); (b) support to the Rural Water Supply Agency (SNAPE); and Cc) staff training.  - 18 - 2. Investment and Maintenance Proaram Ca) rehabilitation and subsequent maintenance of about 2,600 km of feeder roads; (b) development of about 2,100 ha of small bottomlands; and (c) provision of water to about 140 villages through the drilling of 220 boreholes to be equipped with handpumps, and implementation of two small community piped water supply systems with sanitation components. 3. Studies and Pilot Programs (a) preparation of a Feeder Roads Master Plan to serve as blue-print to implement further feeder roads investments; (b) implementation of a Pilot Feeder Roads Program to develop labor-based methods for maintenance and rehabilitation of feeder roads, and to promote the emergence of small and medium scale contractors; (c) implementation of a Pilot NGO-Supported Bottomland Development Program; and (d) studies and pilot actions to define a national strategy for RWSS development. 3.03 Proiect Scone. The project activities would take place in all four natural regions, but would support in priority zones covered by specific agricultural development activities in 16 of the 33 prefectures of the country, including, but not limited to, the areas covered by the IDA-financed National Research and Extension project. 3.04 Rationale for IDA Involvement. Rural infrastructure rehabilitation requires support of a scope and nature that IDA is in a position to provide. The proposed project would complement other IDA-financed projects (National Research and Extension, Transport Sector, and Fourth Highway) in stimulating agricultural production and income. The project RSS component would benefit from the UNDP/World Bank Water and Sanitation Program through the Abidjan- based Regional Water and Sanitatiou Group. Through the project, IDA would play a catalytic role in enabling the Government to coordinate the activities of the many donors involved in the rural infrastructure sector. In particular, the project would assist the Government in integrating all externally financed projects into a national program for rural infrastructure development. IDA has a significant comparative advantage in providing this kind of support given its involvement in structural adjustment, anx agriculture and infrastructure development in Guinea. Moreover, its leadership role would ensure that adequate attention is paid to critical aspects such as institutional development and cost recovery issues.  - 19 _ C. Detailed Features 1. Institutional Development (a) Support to the Rural Engineering Directorate (DNGR) 3.05 General. This component would provide support to DNGR and its decentralized services as presently established following their recent restructuring under th-- Government's administrative reform program. This would include techni al assistance, staff training, construction and renovation of offices and facilities, and provision of miscellaneous equipment. The component would also provide for similar support to the four Regional Bureaus of Rural Engineering (BTGRs) to be created under the project (para 4.01). 3.06 Technical Assistance. Technical assistance would consist of resident and visiting experts. All internationally recruited resident experts would be placed in advisory positions, and these experts would give priority to the training of national staff, in line with government policies (para 2.08). They would assist DNGR in all its activities, including rural engineering. administration, accounting, finance, procurement and contract management. The emphasis would be placed on the implementation of a program of periodic visits of two to three weeks by visiting experts, on average every three months. At the end of each visit, a three-month work program for each key DNGR staff would be agreed with the Director General of DNGR and the resident experts. This approach to technical assistance might, in the short-tenm, result in relatively slow progress of DNGR's work; however, it would place the burden of implementation responsibility entirely on national staff and ensure the sustainability of investments in technical assistance. 3.07 The list of the key national staff with whom the expatriate staff would work directly according to the above three-month rolling work programs is given in Annex 8. The list of resident technical assistants and their terms of reference are presented in Annex 9. The experts financed under the proposed project would be the following: (i) Resident ExDerts 3.08 The project would provide a total of 18 man-years of expatriate technical assistance to DNGR's central and territorial services. Two resident experts would be recruited as technical advisors for DNGR's headquarters in Conakry. The first expert would, for five years, assist the Division Chief for Rural Roads and Constructions in all administrative and technical aspects of the preparation of studies, the establishment of contracts and work supervision concerning rural roads and constructions with specific emphasis on operations connected to the project road investment program. The second expert would be posted for three years at the DNGR's cabinet to assist in the programming and execution of the project training program (paras 3.17 - 3.20). Three other resident expatriate staff, posted up-country, for a total of 10 man-years, would assist in the implementation  - 20 - and subsequent operation of the regional BTGRs to be created under the project. (ii) Visiting Experts 3.09 Short-term consultancies would be required under the project to assist DNGR in completing specialized punctual tasks (socio-economic surveys, soils analysis, computing assistance, financial audits, environmental assessment, etc.). Also, in view of the likely difficulty by IDA and other co-financiers to supervise the project given its complexity, e special provision of a total of five man-months of short-term consultancies (one annually) has been made to specifically strengthen DNGR's capacity to contribute to supervision missions. The short-term consultancy requirements have been estimated at a total of 67 man-months. 3.10 Renovation of Offices and Facilities. DNGR's office space requirements are estimated at about 900 m2 for its central staff at headquarters in Conakry who number about 165 persons. These staff are presently housed in Conakry downtown area, in a sub-standard building dating back from the colonial era, whose total area is only about 300 m2. It has therefore been decided that they would be relocated to a 2 ha site belonging to MARA in Matoto, Conakry's burgeoning industrial area. Although this site is somewhat far from MARA's headquarters located downtown about 20 km away, it offers the advantage of being close to Conakry's exit and the main Conakry - Hamou road axis. This location permits easy links with the interior and would therefore greatly facilitate DNGR's required field work under the project. Also, it is much closer than the downtown area from the Sonfonia training center where the bulk of the class training envisaged under the project would be carried out. The several buildings of a total ground area of abcut 300 m2, which already exist on the site, would be renovated under the project. DNGR's regional offices (300 m2), workshops (600 m2), and prefectural buildings (1,600 m2) would also be renovated. 3.11 Provision of Equipment. Miscellaneous office and technical equipment, office furniture and vehicles for DNGR's headquarters and the BTGRs would be provided under vhe project, including communications equipment to facilitate communication both with HAM's headquarters downtown and operational sites and administrative offices up-country. At prefectural level, one vehicle and small basic equipment would be provided. (b) Support to the Rural Water SupPlY Agency (SNAPE) 3.12 General. This component would provide support to SNAPE headquarters, its existing decentralized services, and the new regional and sub-regional offices to be established under the project. Support would include technical assistance, staff training, renovation of buildings, provision of office furniture and facilities, and provision of miscellaneous equipment. 3.13 Technical Assistance. Technical assistance would consist of resident experts to supplement the assistance already available to SNAPE. Inter- nationally recruited experts would (i) assist SNAPE in strengthening its  - 21 - extension and maintenance program in RWSS with a focus on community management of water points and small piped systems, and on cost recovery; (ii) improve the management and organization of the drilling work on force account; and (iii) build up SNAPE's capacity to supervise boreholes construction undertaken by contractors. A Community Development Specialist and a Drilling Expert would be recruited for three and two years respectively. These experts would also provide on-the-job training to national staff. Their terms of reference are given in Annex 9. 3.14 The later phases of the sensitization program under the RWSS investment components and the follow-up activities after boreholes construction would be the responsibility of specialized SNAPE staff. These staff would be assisted for a total of 12 man-months by short-term consultants specialized in community development. The consultants would make periodic visits of four weeks, at intervals of two to three months during the first two years, and c,f about six months during the remaining project period. During their visits, they would train national staff, monitor their work and assist them in establishing their work programs. At the end of each consultant's visit, the work programs would be revised and agreed with the Director General of SNAPE. 3.15 Renovation of Offices and Facilities. SNAPE's headquarters staff under its new organic structure is estimated at 60, requiring approximately 300 m2 of office space. SNAPE's staff, like DNGR's, are presently located in sub-standard premises in Conakry downtown area close to NARA's headquarters. These premises would be renovated under the project. Up- country, a uew SNAPE sub-regional office would be created in Kissidougou to serve laitially as a project office and as a temporary regional office pending the establishment of a regional office in N'zer4korO. Then, it would become a sub-regional office because of the remote position of N'szrekore. At the same time, a SNAPE regional office would be established in N'zdrdkor6 which is the regional capital. Provisions for the renovation/construction of 200 m2 have been made for SNAPE's up-country offides. 3.16 Provision of Eauilment. Miscellaneous equipment, including office equipment and furniture, would be provided under the project to SNAPE's headquarters, the regional office in Labd, and the sub-regional and regional offices in Guinee Forestiere. (c) Staff Trainjng 3.17 A major thrust of the project would be to provide the training facilities required to support the strengthening of the Administration and the development of the small and medium scale contractors (SMCs). In this regard, in addition to the training received through the visiting expert program and the on-the-job training and coaching provided by the resident experts, specific training activities would be carried out. The target group for this training would be (a) the Administration's staff, i.e., not only DNGR's staff, but also the staff of the other technical ministries concerned, in particular MTTP; and (b) private sector contractors, including the staff of SMCs and the two Pilot Public Works Companies (SPTPs) to be created under  - 22 - i l the project (para 3.28), and the employees of local cemmity organizations ("collectivit4s locales'). The training activities would be undertaken through a variety of a sans, including in-house training, local seminars, and external training sessions. The training arrangements envisaged are given in Annex 10. At negotiations. the training program for PY1, prepared by consultants under the third PPF advance, was agreed with the Government. Agreement was also obtained that by September 30 of every calendar year, DNGR and SNAPE would submit an action plan for staff training as part of the annual work program for the following year (para 3.73). 3.18 One special feature of training activities carried out under the project would be the establishment of a Mobile School Unit (UME) in Mamou. It would be equipped as a medium scale contractor, i.e., it would have the same equipment as the SPTPs and would perform a similar type of work (para 4.04). However, its work methods would be such as to enable appropriate staff training with the support of specialized expatriate technical assistance. In this regard, two expatriate technical assistants would be recruited and posted at the UME: a training engineer for three years, and a training mechanic for five years. 3.19 The project would sub-contract its training activities to training centers in the country, the bulk of these activities to the Center for the Study and Improvement of Agricultural Mechanization (CEPERMAG) of Sonfonia (on the outskirts of Conakry). This center would be strengthened in buildings and housing and in public works equipment. It would dispense technical training in the classroom and would supplement the field training given at the UME, with particular emphasis on subject matters concerning: (i) preparation of technical reports and bids; (ii) verification of quantities for payment; (iii) establishment of progress reports; and (iv) collection of statistics and establishment of data bases on the national rural highway network. An internationally recruited expert, posted for three years at DNGR's cabinet as training coordinator, would assist in the design and implementation of the project training program. 3.20 Under the RUSS component, SNAPE staff would receive on-the-job training specifically as they participate in: (i) the preparation and superv3sion of the RWSS investment programs in the Kissidougou and Gu4ckddou prefectures; (ii) the nationwide study on Willingness To Pay (WTP) for water; (iii) the strengthening of SNAPE's extension and maintenance program; and (iv) the preparation of the RUSS program for Guinee Forestibre. 2. Investment and Maintenance Proaram (a) Feeder Roads 3.21 The total feeder roads rehabilitation program would cover about 2,600 km of road links which were selected at the appraisal stage. Road selection was carried out on the basis of a multi-criteria analysis applied to about 4,500 km of roads previously identified at the preparation stage in the 16 prefectures covered by the project. The implementation program is described in pars 4.08. The technical specifications and the estimated scope of  _ 23 - rehabilitation, together with the implementation schedule, are presented in Annex 11. 3.22 The roads would be brought up to minimum standards based on projected traffic volumes and topography, and existing alignment would be retained to the extent possible. As part of the road work, bridges and culverts would be constructed or rehabilitated to make roads passable at all '.es. In all cases, special attention would be given to drainage which is responsible for the bulk of disorders in Guinea where terrain is often hilly and rainfall occurs in violent storms. Safety would also be on the priority list as Guinean drivers have lost familiarity with good roads allowing high speed, and their vehicles are for the most part dilapidated. In the cost of road rehabilitation, provision has therefore been made for basic road signalling devices that would help prevent accidents. 3.23 Below are the general characteristics of the four specified categories of road standards: (i) Standard A. This standard, adopted for roads connecting prefectures to sub-prefectures, would permit traffic at an average speed of 40 km/h. The roadway would be 4.5 m wide and graveled, with a 1 m shoulder on each side, resulting in total road width of 6.5 m; (ii) Standard Ba. This standard is for roads connecting sub- prefectures to villages, villages to other villages, and villages to the national or regional network. Its technical standards are the same as Standard A, but with a 3.5 m wide roadway and a 0.5 m shoulder on each side (i.e., 4.5 m of total road width); (iii) Standard Bf. Same as Standard Ba. However, because the present traffic is too low to justify total improvement, the road surface would not be upgraded. In the future, these roads would be improved to bring them to Standard Ba; and (iv) Standard C. Exclusively adopted for 'cul-de-sac' links between villages and production areas. Standard C roads would only be used by animal-driven carts or by the occasional bush pick-up. These roads have a 3 m non-graveled surface which would not be improved under the project. They would get just enough maintenance to stay passable. 3.24 Four different types of maintenance work have been identified based on existing road conditions: (i) Tyne I - Rehabilitation. This work consists of re-opening a road to traffic. Also included is the rehabilitation of bridges, culverts and other drainage facilities, if needed. This work is similar to new construction except that it follows the existing profile and the alignment of the natural terrain as  - 24 - closely as possible, to minimize the need for additional earthwork; (ii) Type II - Periodic Maintenance. This type of work starts immediately following road rehabilitation and is performed twice yearly thereafter. It consists of one pass of motorgrader to resurface the roadway. The first year of maintenance would be the responsibility of the contractor in charge of rehabilitation. Approximately every five or six years, depending on traffic, new graveling is necessary. This would not be covered by the proposed project; (iii) Type III - Routine Maintenance. This consists of permanent maintenance, including the pruning of trees and clearing of vegetation to allow the sun to reach the roadway, maintenance of the drainage system (water-outlets cleaning, energy dissipators rehabilitation), the cutting of grass on shoulders and in ditches, and the reporting of problem areas to the responsible authorities for immediate action; and (iv) Type IV - Emeriency Snot Maintenance. Based on information received from the field, damages to the roadway (due to landslides. potholes, erosion of graveled surfaces, etc.), to the drainage system and to bridges and culverts, should be repaired through quick emergency action. This is necessary to avoid in- depth structural damages to the road and sometimes to re-open the road to traffic. 3.25 The total project rehabilitation program of approximately 2,600 km in total (including the roads rehabilitated under the Pilot Feeder Roads Program, para 3.38) would be the following, per standard (details, including the breakdown per year, are in Annex 11): (a) Standard At about 697 km; (b) Standard Bat about 1,067 km; (c) Standard Bfs about 586 km; and (d) Standard C: about 250 km. The project maintenance program, per category of maintenance work, is shown in Annex 11. 3.26 The rehabilitation of Category AIB roads would be contracted out to private entrepreneurs (large in size and foreign assisted), except for (a) about 161 km in the Coyah and JUubreka prefectures set aside for SMCs under the Pilot Feeder Roads Propram (para 3.38): (b) the 225 km under the community initiatives support program (para 4.08) that would be executed partly by the two SPTPs set up under the project (para 3.28) and partly by SMCs; and (c) 20 km under the Pilot Feeder Roads Program that would be carried out by using labor intensive methods. Work locations of similar characteristics have been regrouped based on preparation studies to make contracts sufficiently attractive to bidders. The rehabilitation of Category C roads would be performed by local foremen ("tacherons") with major labor input from village communities.  - 25 - 3.27 Periodic maintenance would be executed mainly under multi-year, concession-type contracts awarded to local SMCs for certain zones in each region. Part of it would also be executed urnder labor-based meth.ods with the assistance of local communities. Emergency maintenance would be entrusted to the SPTPs andlor contracted out to existing local private entrepreneurs. Routine maintenance would be performed entirely by labor intensive methods under Otacheronnagel and *cantonnage' systems to be experimented as part of the Pilot Feeder Roads Program. Under these systems, the petty contractors (Otfcheronsw), communities and NGOs would engage Olengthmen' responsible for the maintenance of certain sections of road or specific tasks, depending upon the type of work to be undertaken. In order to attract labor, and, at the same time, avoid a negative impact on agricultural and other productive activities, each lengtbman would generally be given tasks which would require only part time work on the road (say about 12 days of work per month), leaving sufficient time for other activities. On average, it is estimAted that routine maintenance will require one laborer for each 1.5-2.0 km of road. The petty contractors, communities or NGOs would be paid at an agreed rate on a monthly basis for works correctly and timely carried out; they would also receive food rations from the World Foo4 Program (WPP). 3.28 As part of the feeder roads component, two private Pilot Public Works Companies (SPTPs) would be created to provide the operational capacity required to undertake the emergency road maintenance in areas where such capacity (either public or private) does not exist. These SPTPs would be based in N'zerekore to cover the prefectures of N'z4r4kor4, Beyla, Macenta, Lola and KRrouane, and in Gaoual to cover the prefectures of Gaoual, Koundara, and Telimele. They would be equipped partly with heavy equipment presently belonging to MARA, as the inventory conducted under the third PPF advance has determined that some equipment was available and could therefore be allocated to the SPTPs after rehabilitation. The remainder of the heavy equipment would be provided on a rente' basis by the renting company to be set up by the Government (para 3.41) (see list of equipment required and breakdown between rehabilitated, purchased (new) and rented equipment in Annex 11). The SPTPs would be established as private enterprises, but would have to follow a pre-determined work plan ("contrat plan"' spelling out their obligations vis-&-vis DNGR (para 4.05). In order to proceed with the required reallocation of equipment to the SPTPs, assurances were obtained at negotiations that Government will prepare a plan of use of MARA's public works equipment based on the inventory undertaken as part of the third PPF advance. (b) Bottomlands 3.29 DNGR would be responsible for bottomland drainage and infrastructure development under the project's bottomlands component. Since bottomland development requires the support of the extension and research services to assist farmers in the bottomlands, a Coordinating Committee would be established for coordinating the implementation of this component. The Committee would comprise the Directors of Rural Engineering, Extension Services, and Agronomic Research. It would oversee the planning, budgeting  - 26 - and implementation of the bottomland development program of the project and would provide direction aLd guidance to the regions in the preparation of their medium- and long-ter. plans for the development of their bottoulands. The Committee would further review these plans and would assitt the Government in development planning for the exploitation of these bottomdands at the national level. Ihe Committee would meet at least twice a year and would be responsible for approving work programs and budgets related to this component. It would also review project reports and studies relating to these activities. Assurances were obtained from the Government at negotiations about the establishment and functioning of the Coordinating co-mittee. 3.30 A Multi-Disciplinary Team would be established in each region at BTOR level which would be responsible for the planning and studying necessary for developing the programs to be implemented under the project. Each team would consist of the region's rural engineer, the agronomist, and a sociologist and a soil scientist where possible. Each team would identify bottomlands presently in use by farmers, examine the sites, determine the farming systems, discuss their utilization with the farmers in order to establish their needs, assess the land tenure status, and make recommendations for improving productivity. These recommendations would be first based on the implementation of simple technological changes which the farmer could readily adopt without any major disruption of his farming goals. 3.31 The proposed project's program and levels of bottomland development (Types I - TV) are presented in Annex 12. Each team would propose a detailed annual program for itb region, which, when approved by the Coordinating Committee and IDA, would be implemented by the relevant BTGR and the Extension Services. In addition to monitoring the implementation of these programs, each team, with the help of a national inventory of bottomlands to be carried out by consultants, and based on criteria set out by the Coordinating Committee, would develop a medium- to long-term plan for bottomland development in the region. The four regional plans would form the basis for a national bottomland development program. These task are expected to be completed by PY5, although the team could be disbanded earlier should its work be completed before the fifth year (e.g., for lack of available bottomlands). Assurances were obtained cat negotiations that DNGR will establish Multi-Discinlinary Teams under the above conditions. 3.32 The Multi-Disciplinary Teams would determine priorities and types based on their assessment of the farming systems and the role of the bottomland in such systems, farmers' own needs, availability of labor, ease with which groups can be formed for carrying out the necessary works, physical characteristics of the bottomland, need for and availability of suitable inputs, farmers' ability to purchase required inputs, and marketing prospects for the proposed crops. Water management structures (where necessary) would be designed by the Hydro-Agricultural Division of DNGR. Earthworks would be carried out manually by the farmers themselves. However, construction of heavier structures such as small reservoirs for Type IV, would be contracted out to the SPTPs or to small contractors where practicable. The varietal and agronomic recommendations would be the  - 27 - responsibility of the Extension Services. However, the Agronomic Research Department vould embark on a program of research specific to the needs of the bottomland development program. Each Multi-Disciplinary Team would monitor emperience in implementation, which would be reflected in the annual work programs for the following years. (c) Rural Water SUDDlY and Sanitation (RUSS) 3.33 The RUSS investment program would consist of two partss the drilling of 220 productive boreholes and the construction of two village piped water systems. The first part would serve, by means of individual boreholes equipped with handpumps, about 140 villages .in the prefectures of Kissidougou and Gueckedou, having a total forecasted population of about 65,000 by the year 1995. These prefectures were selected because (i) they are located in Guinee Forestiere which has lagged behind in RWSS development compared with other regions; and (ii) they have not benefited todate of any RWSS programs, whereas such programs are needed to complement the development activities underway as part of the DERIK and PAG projects covering these prefectures. The investment program would comprise the drilling of the boreholes at an average depth of 50 m, the building of superstructures, the supply and installation of handpumps for boreholes, the setting up of a network of shops for spare parts, and the training for handpump caretakers, including provision of maintenance services, training manuals, and repair tool kits. The project would make provision for the consulting services required to determine borehole sites and supervise construction works (including drilling and testing boreholes, building superstructures, and installing and testing pumps), and for the sensitization of the target population on the maintenance of the water points and on the linkage of water and sanitation with health. 3.34 The second part, consisting of the construction of two small piped water systems would serve the villages of Yend4-Millimou in the prefecture of Kissidougou, and Nongoa in the prefecture of Gueckddou whose populations would be about 3,000 and 4,500 respectively by 1995. These villages were selected because they are markets of regional importance and are representative of the socio-ethnic fabric and physical environment prevailing in the two prefectures where they are located. Also, their accessibility would greatly facilitate carrying out the monitoring and evaluation activities required to assess the impact of the program from the SNAPE's sub- regional office in Kissidougou. The sevan boreholes necessary would be drilled as part of the productive boreholes program. The two systems would include small distribution networks equipped with water tanks, public standpipes, and solar or other suitable pumps. The project would make provision for the consulting services required for the preparation of detailed designs, tender documents and assistance to SNAPE in (i) bid evaluation and contract preparation; (ii) development and application of community management; (iii) construction supervision; and (iv) sensitization of the population in the two villages on the benefits of water and sanitation for health.  - 28 - 3. Studies and Pilot Proarams (a) Feeder Roads Master Plan 3.35 As part of project preparation (second PPF advance of US$300,000, para 3.55) government consultants have prepared a Master Plan for the rehabilitation of feeaer roads. Under this Master Plan, about 14,500 km of roads over the entire territory of Guinea were identified and surveyed, providing an estimate of rehabilitation costs and of the benefits that would accrue if these roads were rehabilitated. Subsequently, the road links were classified on the basis of a multi-criteria analysis whereby the cost of rehabilitation was weighted against the benefits. The methodology used involved a multi-stage decision sequence with classification at prefecture level prior to integration at regional and national level. Such a methodology has proved useful in that it has permitted to establish first an interface with local authorities and beneficiaries, prior to moving to a higher decision-making level. The Master Plan has served to calibrate the projects road investment program prepared concomitantly and in close liaison with other government consultants. Under the Plan, a comprehensive computerized data base is now available. As part of the work program for the third PPF advance, this data base has been further processed to make it more user-friendly, using a package compatible with a personal computer; also, Guinean nationals have been trained to use it. The Master P'an would be the principal planning instrument made available to DNGR under the project. With the assistance of short-term consultants, it would be further developed and thereafter continuously updated, as additional data becomes available and further experience and knowledge are gained as a result of project implementation. (b) Pilot Feeder Roads Program 3.36 Obiectives. The long-term objective of the Pilot Feeder Roads Program is to test labor-based methods for road rehabilitation and maintenance relying, to the maximum extent consistent with cost-efficiency, on local resources using domestic contractors, NGOs and communities for implementation in order to promote sustainable feeder roads development. The pilot program's immediate objectives would be to: (i) create a capacity within DNGR and a number of domestic private contracting firms to effectively apply low cost methods of feeder roads rehabilitation and maintenance, using locally available resources; and (ii) develop a capacity at the commity level and with NGOs to plan and implement rehabilitation and maintenance of feeder roads. 3.37 Summarv Description. To achieve the above objectives, the program would include the following actions: (i) assistance to domestic contractors for the rehabilitation, using mainly labor-based methods, of about 161 km of Category A/B roads in the prefectures of Coyah and Dubreka; (ii) experimental rehabilitation of 20 km of Category C roads and 20 km of Category A/B roads; (iii) minor repair and subsequent maintenance on a pilot basis of about 207 km of existing feeder roads (Category A/B) in the prefectures of Dabola and  - 29 - Tdlim6ld; and (iv) technical assistance to support DNGR in implementing the proposed programs and to train goverument staff, domestic contractors, coununities and NGOs in the implementation of labor-based feeder roads rehabilitatic,n and maintenance. 3.38 Rehabilitation of Casegory AIB Roads. It is envisaged under the pilot program to provide support to SMCs for the rehabilitation, over PY1 to PY4, of about 161 km of roads in the prefectures of Coyah and Dubreka. The roads would be upgraded to standards A, Ba and Bf, as described in Annex 11. Standards A and Ba roads would be provided with a lateritic surface layer. Average hauling distances would generally not exceed 2 km. The works would be carried out using mostly labor-based methods supported by ight equipment such as pedestrian rollers for compaction and tractor-trailer combinations for transport of selected materials. 3.39 Present official labor wages in the modern construction sector (GF 47,000/month or about US$80/month for unskilled labor), recently agreed between the labor unions and the contracting industry, represent a doubling of the wages which had been valid for several years. Wages in the informal sectors (including agriculture), however, are substantially lower. On these premises, and with the expected substantive local inflation (para 3.53), it must be expected that the level of local wages would enhance the competitiveness of domestic contractors using labor-based methods vis-&-vis international contractors using capital-intensive methods. In addition, equipment-based methods are only cost-effective when equipment is actually used intensively, which would not be the case for the scattered and seasonal works undertaken as part of the proposed program. Unskilled labor is available in the project area. 3.40 SMCs would be engaged to carry out improvement work on the 161 km of roads mentioned above. At least five well established domestic contractors have expressed a strong desire to enter the civil engineering market, &n area in which they have little experience to date. In contrast, they have substantial experience in building construction. In the present context, these SMCs cannot compete with international contractors for large feeder roads contracts because conditions are such that only firms with a large holding of heavy equipment can compete, and equipment for rent is not available on a regular basis. Furthermore, foreign contractors engage domestic firms as sub-contractors only to a limited extent. 3.41 For these reasons, to develop domestic SMCs, it is necessary initially to train them and subsequently provide them with contract opportunities within their technical and financial capability. For domestic contractors, the main conatraint would be capital outlays for equipment acquisition with the associated risk that future contracts may not suffice to amortize and renew the equipment fleet. Therefore, the strategy for developing the domestic contracting industry would be based on (i) using labor-based construction methods--this will over time lead to lower construction costs and foreign exchange costs, and better employment opportunities; (ii) encourag'.ng large contractors to sub-contract to SMCs; and (iii) relying on the serices of a renting company to be created by the Government. It was  _ 30 - agreed at negotiaticns that establishment of such a company on terms and conditions acceptable to IDA will be a condition of disbursement against road I maintenance expenditures. 3.42 It is unlikely that the domestic contractors would have enough suitable equipment for carrying out the works. Therefore, the project would initially procure most of the essential equipment required. After the first year of training, contractors would be given the opportunity to acquire the equipment throu h the renting company to be established. Arrangements should oe made whereby DNGR would deduct, from the contractors' monthly statements, equipment amortization and interest charges which would be paid directly to the renting company. 3.43 Rehab'litation of Cateacro C Roads. Experimental rehabilitation on 20 km of Category C roads using labor-based methods would be carried out under the proposed pilot program. The results would be used in rehabilitating about 230 km of Category C roads envisaged under the project (para 4.08), to permit vehicular traffic between villages and agricultural areas (mostly bottomlands). The rehabilitation works would be contracted out to petty contractors ('tAcherons'), NGOs or village communities. Of the total cost (including labor costs), about 15Z would be borne by the communities benefitirg from road rehabilitation. Subsequent road maintenance would be carried out by the communities. 3.44 Maintenance of Feeder Roads. Under the pilot program, permanent maintenance methods would be demonstrated, and government staff, contractors, NGOs and communities would be trained in implementing maintenance works. Two feeder roads rehabilitated under earlier projects in the prefectures of Dabola (72 km) and Telimele (135 ki) would be used as demonstration and training grounds for the introduction of permanent maintenance methods and procedures. The two feeder roads are 1resently in need of limited spot rehabilitation (over about 20 km) which would be carried out prior to permanent maintenance. 3.45 Technical Assistance and Training. Technical assistance would be provided to (i) assist DNGR in implementing rehabilitation and maintenance works and in setting up contracting procedures to carry out these works; and (ii) train DNGR staff, domestic contractors, communities and NGOs. In total, it is envisaged that the project would provide 63 man-months of expatriate technical assistance. This assistance, posted at DNGR's Rural Roads and Constructions division, would be recruited (i) partly internationally: a Chief Technical Advisor for 39 months, a Training Advisor for 15 months, a Tools Specialist for one month, and a Civil Engineer for ten months; and (ii) partly locally: a Road Construction Engineer and a Road Maintenance Engineer each for 51 months. This technical assistance would be provided through the International Labor Organization (ILO) acting as executing agency for the pilot program. 3.46 Implementation. For the pilot road rehabilitation program in the prefectures of Coyah and Dubreka, four to five domestic contractors would initially be selected for training during the first construction season  - 31 - (October 1990 - June 1991). The contractors would make available technicians, foremen and some skilled workers to be trained partly in the classroom, but mainly on site with the UME and as part of the work program for the first year. This program would consist of constructing about 15 km of feeder roads which sbould include all main work activities which may be encountered in the subsequent years. Of the trained contractors, three would be awarded contracts for the second construction season, using unit rates established in line with experience gained during the first year. In the third construction season, domestic contractors would be awarded contracts based on competitive oidding. Experimental rehabilitation over 10 km of Category C roads would be initiated in early 1990 by a NGO (the Association de Developpement de Poredaka - ADP) that has been identified in the prefecture of Mamou, with project and NtGOcommunity financing. The feeder roads maintenance program would start in PY1 over approximately 80 km equally distributed becween the two prefectures concerned (Dabola and N liLMl6) with the aim of covering the entire two stretches (207 km in total) as of PY2. (c) Pilot NW -SuDported Bottomland Development Program 3.47 The national NGO movement in Guinea is recent and few local NGOs have been involved independently in projects of any magnitude. However, several international NGOs are already active in rural development operations and would be interested in providing support to pilot operations carried out under the project, if they were allowed to preserve their own individual approach. One area in which they are already involved is bottomland development. 3.48 The need to organize and coordinate farmer activities for optimal bottomland utilization is imperative, and, for the most part, Guinea has little experience in this area. Some recent interventions by NG0s, particularly in Guin4e Maritime, indicate that these organizations are well placed to assist farmers in this respect. Already, by working directly with groups of farmers, they have gained their confidence, and it appears that the superior organizational and administrative skills that they bring to their projects (albeit on a very small scale) are more readily accepted by farmers because of the personal attention that they receive from NGOs. However, it is recognized that the full exploitation of bottomlands requires a wide range of considerations (technical and socio-economic), and that neither MAM nor the NG0s have sufficient experience to cover the gamut of requirements of this component. Also, there is as yet no formula for the intervention of NG0s in this field which would guarantee success of any such development program. There is a need, thcrefore, to determine what expertise can be best provided by the NGO when it intervenes, and its relationship to government agencies operating in the same field. 3.49 The Government, mindful of the complexity of the task of introducing the proposed new and higher technical standards combined with the organizing of farmers to manage their own affairs on a group basis, wishes to encourage NGOs to provide assistance in realizing these objectives. However, such assistance should complement the activities of its designated agencies and should be well structured. NGOs generally contribute human and material  - 32 - resources of a limited nature, and these could supplement the institutional arrangements that HARA is already making (via the proposed project and the National Research and Extension Project), for fulfilling its role as the principal engine of development. 3.50 In order to test NGO participatior., a sum of US$600,000 has been allocated under the project to finance a pilot NGO bottomland development program under the direction of DNGR. This experiment would help determine a model for NGO intervention. The NGO to be selected, which should already be active in this field, would be requested to prepare and discuss with MARA a work program specifying quantified outputs and a schedule of payments to be made by MARA. These payments would be free of usual contractual/ administrative checking and only tied to the respective work programs. Further, there would be an ongoing monitoring of its actions in order that the most effective levels of intervention by NGOs of similar competence could be determined as quickly as possible. Details of this pilot program would be prepared in PY1 by the Government, with the assistance of AFVP, a NGO which is already doing work on bottomlands for DNGR (para 2.36). This would be the basis for an implementation agreement between DNGR and the NG0 to be selected, for which IDA's prior approval would be necessary (para 3.61). (d) RWSS Studies and Pilot ProRrams 3.51 These stu4.es and pilot programs includes (i) Rural Water Supply and Sanitation Program for Guinee Forestiere. A RWSS master plan for the prefectures of Kissidougou and Gu6ckedou was prepared under the first PPF advance. Under the project, this Master Plan would be extended to the whole of Guin4e Forestiere and would include sanitation. The study would establish a master plan and program that will serve as a basis for the preparation and implementation of R'SS investment projects. The program would include detailed proposals for three projects: handpump-equipped boreholes, springs and handpump-equipped wells, and detailed studies for small water supply systems, with sanitation components, for the large or densely populated villages. The study would also include the feasibility of promoting small and medium scale private contractors in the RWSS sub-sector (para 2.44); (ii) Nationwide Study on WillinQness To Pay for Water. The assessment of the willingness to pay for water, undertaken in the framework of the third PPF in certain villages within the project area of the RWSS investment component, will be pursued and extended on a national basis. The purpose of the nationwide study would be to provide information for the definition of a national policy with regard to the financial participation of the beneficiaries in the purchase of handpumps. The study would be made in selected  - 33 - villages in each of the four natural regions, including the project area of the RWSS investment program; (iii) Strentthening of SNAPE's RWSS Extension and Maintenance Program. The objective of this study is to establish, implement and follow- up a sensitization and maintenance program. A community-based management system would be established. The participation of the population in the maintenance of water points and systems would be secured and SNAPE's extension program would be reinforced. The study would include the development, testing and application of sensitization and training materials, guidelines, and procedures to be used all over the country; and (iv) Pilot Sanitation ProAram. The program wculd be prepared and implemented to develop, test and apply methodologies for low-cost sanitation and solid waste disposal in large or densely populated villages. The testing and application would be carried out in the villages of Yende-Millimou and Nongoa. D. Pro4ect Cost and Financing 1. Cost Estimates 3.52 The total cost of the proposed project over the 1990-94 period, including all incremental capital and operating expenditures, is estimated at GF 54.6 billion (US$88.8 million), of which GF 38.8 billion (US$63.2 million) or 712 in foreign exchange and about GF 4.0 billion (US$6.3 million) or 7S in taxes and duties. 3.53 Base costs have been estimated at June 1989 prices using the prevailing rate of exchange of GF 590 to the U.S. dollar prevailing at that date. These costs have been readjusted using the rate of exchange of GF 615 to the dollar prevailing at negotiations in November 1989. Physical contingencies of varying amounts have been allowed on investment and operating cost items, except for vehicles, equipment, lorg-term technical assistance, PPF-financed activities, and staff salaries. The average physical contingency amounts to 7.6Z of base costs, with individual rates ranging from O to 102. Price contingencies have been applied to base costs over an estimated implementation period of five years (1990-94). Price contingencies on foreign exchange costs based on the Bank's projections of international price increases have been included at an average of 7.22 for 1990 and 1991, and 4.4Z from 1992 to 1994. Price contingencies on local costs are based on annual price increases of 20Z for 1990, 16Z for 1991, 11.5? for 1992, 7.22 for 1993, and 5.12 for 1994, to reflect current inflation trends in Guinea. Total price contingencies amount to 232 of base costs.  - 34 - 3.54 Project costs are summarized below and details are given in Annex 13. l..act Cast Summry A. . i 1. iON SUPOIrT 1,890.9 4.624.4 0.05. 2.4 7.6 9.6 7.9 14.4 2. S8APE SWPORT 221.8 714.4 91.7 0.4 1.2 1.A 76. 2.2 8. TRADN. 1AL 4-32.7 524-4 _1 .4 9 .-l.1 8.1S- 1 - Sub-TOal 2,674.0 9,901.5 12,675.4 4.5 16.0 20.4 78.7 80.1 13. TN A 1. iDE RUNS 6o099.8 15,88.8 21.68.1 10.2 28.0 88.2 71.8 51.6 2. 93W S 807.0 69".6 1,362.6 0.7 1.6 2.2 70.9 8.8 a. vAM AATOt SUlPPLT. M.1 180-4 2 '185 --0.0 78.1 -u Sub-Total 7,004.9 18,831.8 25,368.2 11.7 29.8 41.2 72.4 60.6 C. sTIDIOS a PILOt iWEm 1 SU16 14.0 1.258.7 1.890.7 0.2 2.0 2.8 90.4 8.8 2: PILOT ACIsl 4 s..L2L01 .M8 2..e -Ai JPAZ ..J, Sub-Totel 617.9 a,274.8 8,8s2.2 1.0 5.8 8.8 84.1 9.8 IMTAL bASELDNE COMZS 10,2W.7 81,807.1 41.800.8 17.3 80.7 18.0 74.6 100.0 Phy scal Continncies 044.8 2,826.6 8,170.7 1.4 8.8 8.2 78.4 7.6 Pr ic Catlngaa. 4,617.7 .L020J 9iJ28Z7 _._ .Z7 Il7. .mJ .1. TOTAL FROJ62TS CSTSr 15,7^8.7 88,s44.8 54,608.2 28.6 48.2 68.6 71.1 110.6 (Including taxes) .. _- _ _ ._ 3.55 Prolect Prenaration Facility. Two PPF advances of US$400,000 and US$300,000 were awarded to the Government in August and November 1988, respectively. The first PPF advance was used to finance project preparation studies and the second one to finance preparation of the Rural Roads Master Plan. A third PPF advance of US$750,000 was awarded in July 1980, inter alia, to (a) prepare technical documents for the first five lots of rural roads; (b) complete an inventory of heavy equipment belonging to MAMA and to MTTP's Roads Project Office (OPR); (c) evaluate staff and set up the 1990 training program; (d) train 12 engineers abroad; (e) strengthen DNGR's programming capability; (f) assess the willingness to pay for water in certain villages within the area of the RWSS investment component; (g) strengthen SNAPE's extension and maintenance program; and (h) complete the legal texts required for the establishment of project structures. 2. Financing 3.56 Project costs would be borne by the co-financiers in the following amounts and proportionst  - 35 - e;nane;aan Plan ila m - .ra_, Catm, Z~~am main m to D huf.g nj ma*hu 'Ii) ns 2t,.la rA