Madagascar - Second Irrigation Rehabilitation Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No.12691-MAG STAFF APPRAISAL REPORT MADAGASCAR SECOND IRRIGATION REHABILITATION PROJECT JUNE 20, 1994 Agriculture and Environment Division South-Central and Indian Ocean Department African Region This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. CURRENCY EQUIVALENTS Currency Umt = Malagasy Franc (FMG) US$1 00 FMG 3,(X) (Mas 1994) SDRI (X) -- FMG 4,230 (Ma> 1994) WEIGHTS AND MEASURES Metric Systen ABBREVIATIONS ADR Rural Developient Agent (Agent de développement rural) AfDB African Development Bank AIR Rural Insitutions Support (Appui aux insitutions rurales) APD Deiailed Feasibility Study (Avant projet détaillé) CA Rural Sociologist (Conseiller animateur> CCCE Caisse Centrale pour Coopération Econonique (nw CuD) CFD Caisse française de développement (forinerly CCCE) CH Irriganon Schtene Advisor (Conseiller hydraulique) CIRGR Rural Works Sub-Prefecture Area (Circonscription du génie rural) CIRVA Agricultural Extensioti Zone (Circonscripton de vulgarisation agricole) CP FAO/World Bank Cooperation Program DA Directorate for Agriculture (Direction de l'agriculture> DGR Rural Works Directorate (Direction du Génie Rural) DGRH Human Resource Directorate (Direction de la Gestion des ressources humaines) EAO In Depit Orientaion Siudy (Etude approfondie d'orientation) EDF Furopean Development Fund >Fonds européen de développeieni) ERR Fcononic Rate of Return ET Iield Teaî (Equipe de terrain) FAO Food and Agriculture Organzation FIFAMANOR Dairy and Agricultural Extension Project (Projet d'Assistance Norvégienne au Développeiment de l'Agriculture sur les Hauts- Plateaux) FMG Malagas> Franc (Franc Malga,hei FOFIFA Naîtonal Appied Research Center for Rural Developtent (Centre national de recherche appliquée pour le développement rural) GDP Gross Domestic Product (Produit national brut) GPl Large Irrigated Schemes (Grands péritéres irrigués) GR Rural wîorks (Gétîe rural) ICB Internattonal Competitve Bidding ( Appel d'offre internaional) IDA Internattonal Developient Associatîo (Association pour le développement international) KfW Kreditanstalt fur Wiederaulbau KOBAMA Minoterie d'Etat MEADR Ministry of Agriculture and Rural Developneni (Ministère d'Lia( à l'Agriculture et au Développenent Rural) MPI Micro Irrigation Schenes (imicro périmètres irrigués) O&M Operations and Maintenance (Gestion et entretien) ODOC Land Registration Operation (Opératon dornaniale concertée) OGL Open General Licence ONG Non-governiertal Orgatzation (Organisuie noi-gt,uverrneîiiemetal) PAF Emvironiental Acion Plan fPlati d'action pour l'environnement) PAM \World Food Prograni Programtrne aimentaire mondiale) PF FamilI ]rrigated (Périmtètre famial) PIP Public Investîtment Progran (Prograr d'intvestissuement publtei PM E Smtall and Medjuni Scale Enterprises (Petite et ioyenne entreprises) PNUD Programme des Nations Umes pour le développement ilîied Naîtuons Developieni Programti) PPF Project Preparation Facility (Avance pour préparatîon de projet) PP[ Snall Scale Irrigation (Petits périmètres irrigués) PNVA Nattonal Agrieultural E'xteitsion Project (Programme national de vulgarisauon agricole SAC Sectoral Adjustmreni Credit SAF Fnancial and Adîitîstruise Service (Service administratif et fnanier) SAMANGOKY Bas Mangoky Developmtîent Coipany (Société pour l'aménagement et la ise en valeur de la vallée du Bas Mangoky) SDA (Schétna directeur d'aménagement) SDR Special Drawing Rights (Droits de tirage spécial) SMITIS Directorate for Programmttttmig and Fîmances/Methods and Processmng Service SOMALAC State Society of Lac Alaotra (Société d'Eiai du Lac Alaotra> SOE Statenient t Expendîture UNDP United Naotins Developtîtent Progratt (Programme des Natîons Uies pour le développenent) WUA Water Users' Associations (Associations des IUsagers d'Eau> GOVERNMENT FISCAL YEAR January 1 - December 31 FOR OFFICIAL USE ONLY MADAGASCAR SECOND IRRIGATION REHABILITATION PROJECT STAFF APPRAISAL REPORT CONTENTS CREDIT AND PROJECT SUMMARY ................................. i I. BACKGRO U ND ...................................... 1 A . Project Background ............................... 1 B. Economic Background .............................. 1 G eneral . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Economic Performance ........................... 2 Structural Adjustment Policies and Results ............... 2 C . The Agricultural Sector ............................. 3 Agriculture in the Economy ........................ 3 Recent Agricultural Developments ........................ 4 Potential for Agricultural Growth .................... 4 Constraints to Agricultural Development ................ 4 Government Strategy for Agriculture ...................... 4 D . The Irrigation Sub-Sector............................. 5 Irrigation Development Strategy .................... . 6 E. Administrative Organization ............................... 7 F. Bank Group Involvement ............................ 7 IDA Lending Strategy for Madagascar .................... 7 Recent Lending Experience ........................ 8 Lessons from Previous IDA Experience .................... 9 II. THE FIRST IRRIGATION REHABILITATION PROJECT ............. 10 A . Project Implementation ............................. 10 O bjectives. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 Im plem entation................................. 10 B. Lessons from the First Project ......................... 11 Inventory Study ........ ..... .. ....... ......... 11 Feasibility Studies . . ... . . . .. . .. . . . . .. .. .. .... . . . 11 Definition of Responsibilities ...................... . 11 C o sts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1 Environmental Impact ........................... 12 W ater Users' Associations ........................ . 12 Agricultural Development ......................... 12 Institutional Structure............................. 13 III. THE PROJECT....................................... 14 This report is based on the findings of the appraisal mission undertaken in March 1993 by a team consisting of Messrs. C. Trapman (Agricultural Economist, Mission leader), Adel Bichara (Irrigation Engineer), John Buursink (Environment Planner, consultant), Enda J. Byrne (local participation/NGO specialist, consultant), Solange Elfassy (agricultural extension and community development specialist, consultant), Bruno Ribon (Financial Analyst, consultant) and Jean-Claude Rocaboy (Institutional Specialist, consultant). Peer review was carried out by Messrs G. Donovan and J. ter Vrugt. The Director, the Operations Adviser and the Division Chief of the responsible department and division are F. Aguirre-Sacasa, S. Capoluongo and N.O. Tcheyan, respectively. This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. A. Project Rationale ...................................... 14 B. Project Objectives and Design . . . . . . . . . . . . . . . . . . . . . . . . . 14 Objectives........................................ 14 Main Features ..................................... 15 The Project ....................................... 15 The Challenge ..................................... 16 C. Summary Project Description................ . . . . . . . .. 16 D. Detailed Project Description..................... . . . .. 17 Cyclone Damage Repair and Corrective Rehabilitation Works . 17 Environmental Protection................. . . . . . . .. 18 Rehabilitation of New Irrigation Schemes (US$8.2 million) . . . . 18 Support to the MEADR..................... . . . .. 19 Training............................. . . . . . .. 21 Monitoring and Evaluation/Audit............. . . . . . .. 21 Operations and Maintenance by Water Users Associations ..... .21 E. Project Costs and Financing Costs ...................... 21 Contingency Allowances .......................... 22 Incremental Recurrent Costs ....................... 22 Project Sustainability ............................ 23 F. Financing ...................................... 23 IDA Credit .................................. 23 C o-Financing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 Government's Share ................................ 23 Beneficiaries' Share ................................. 24 IV. PROJECT IMPLEMENTATION ............................... 25 A. Status of Project Preparation . . . . . . . . . . . . . . . . . . . . . . . . . 25 B. Project Implementation ................................. 25 Project Implementation Timetable . . . . . . . . . . . . . . . . . . . . 25 Directorate of Rural W orks ........................ 25 Directorate for Agriculture ........................ 27 Directorate for Programming and Finances/Methods and Processing Service.......................... 28 Project Coordination ............................ 28 Financial and Administrative Service . . . . . . . . . . . . . . . . . . 29 Beneficiaries ...................................... 29 C. Project Monitoring and Evaluation . . . . . . . . . . . . . . . . . . . . . . 30 D. Procurement and Disbursements . . . . . . . . . . . . . . . . . . . . . . . 30 Civil Works....................................... 31 G oods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 1 Consulting Services........................ . . . . .. 31 Prior Review Procedures and Reporting.......... . . . . .. 31 E. Accounting, Auditing, Budgeting, Disbursement and Reporting Procedures .................................. 33 Accounts and Audits ............................ 33 Budgets . ................................... 34 Disbursement Procedures ......................... 34 Disbursement by IDA against Statements of Expenditure . . . . . . 34 Credit Disbursement....................... . . . .. 35 F. Project Impact on Government Budget......... . . . . . . . . .. 36 During Project Implementation............... . . . . . .. 36 G. Legal and Regulatory Framework ...................... 36 Land Tenure ................................. 36 Water Users' Associations ......................... 37 V. AGRICULTURAL DEVELOPMENT AND PRODUCTION IMPACT .... 38 Technological Production Improvements ................ . 38 Financial Impact of the Adoption of New Techniques ........ 39 VI. MARKETS AND PRICES ............................... 41 VII. BENEFITS AND RISKS .................................. 42 A. Financial and Economic Benefits ........................ 42 Institutional Development Benefits .................... . 42 Direct Quantifiable Benefits ....................... 42 Indirect Benefits ................................. 42 Economic Rate of Return .......................... 42 B. Environmental Impact ................................ 43 Soil Fertility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 Health Risks ................................... 44 Effects of Intensification of Agricultural Production . ........ 44 Environmental Actions ............................ 44 C . R isks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 VIII. AGREEMENTS REACHED WITH THE BORROWER, CONDITIONS AND RECOMMENDATIONS ................. .46 TABLES Table 3.1: Irrigation Rehabilitation Project Estimated Project Costs ............. 22 Table 3.2: Physical and Price Contingency Assumptions .................... 22 Table 3.3: Proposed Project Financing ............................... 24 Table 4.1: Procurement ......................................... 33 Table 4.2: Disbursement of IDA Credit ............................... 35 Table 4.3: Estimated IDA Disbursements .............................. 36 Table 5.1: Potential Technical Improvements in Agricultural Production ........... 38 Table 5.2: Increases in Production and Yields with and without Project ........... 39 Table 5.3: Cropping Intensity and Net Revenue Per Model Farm ............... 40 Table 7.1: Sensitivity Analysis of the ERR to Changes in Costs and Benefits ........ 43 ANNEXES Annex 1: Detailed Cost Tables...................................... 47 Annex 2: Declaration of Irrigation Policy............................... 55 Annex 3: Environmental Assessment .................................. 61 Annex 4: Rehabilitation of Medium-Sized Irrigation Schemes (PPIs) by Source of External Funding During the First Project (PPI-1) .................. 64 Annex 5: Calculation of Incremental Benefit/Costs Stream and Internal Economic Return ....................................... 65 Annex 6: Monitoring and Evaluation Performance Indicators ................... 66 Annex 7: Estimated Schedule of Disbursement .......................... 67 Annex 8: Documents Contained in the Project File ........................ 68  MADAGASCAR SECOND IRRIGATION REHABILITATION PROJECT CREDIT AND PROJECT SUMMARY Borrower Madagascar Beneficiary Ministry of Agriculture and Rural Development (MEADR) Credit Amount : US$21.2 million, equivalent to SDR 15 million Terms : Standard IDA terms Project Objectives The project would support the country development strategy on four and Directives fronts: (a) Promoting growth. It would contribute to increased agricultural output through better water management and intensified agricultural extension. (b) Poverty reduction. It would contribute to improved rural incomes and food security by improving production and productivity on irrigation schemes country-wide. (c) Improving environmental management. It would develop techniques for environmental assessment of irrigation rehabilitation and reinforce local capacity for such assessment. (d) Enhancing project implementation capacity. It would establish operational procedures for the transfer of management of the operation and maintenance of public irrigation schemes to water users and thus would serve to reduce the burden of public expenditure on the maintenance of such irrigation infrastructure. It would provide the means for more efficient management of the decentralized services of the Ministry of Agriculture in agricultural extension, rural works supervision and local planning and programming. The main policy reforms are: (i) The transfer of responsibility for maintenance and operations of public irrigation schemes to water users; (ii) the acceleration of the process of the decentralization of support services from the Ministry of Agriculture to provinces and districts; (iii) the activation and testing of a policy for encouraging local participation in the provision of public services in the irrigation sub-sector. The components would include: (i) the consolidation of rehabilitation and transfer of O&M responsibilities started on some 16,000 ha of irrigation schemes rehabilitated under the first phase project, including urgent post-cyclone rehabilitation; (ii) assessment and execution of the environmental protection works needed to ensure the sustainability of rehabilitated irrigation schemes; (iii) rehabilitation of a further 4,000 ha of new irrigated schemes; (iv) strengthening the Ministry of Agriculture's agricultural extension, rural works and programming and planning services, at the provincial, district and local levels to improve local technical supervision of rehabilitation and support for the establishment and functioning of water users associations; (v) training project staff and farmers with an emphasis on developing local capacity and talents to ensure sustainable activities for future phases of the rehabilitation program; and (vi) monitoring and evaluation of project impact. 11 Project Area: Countrywide Estimated Project Costs a\ (US$ million) Local Foreign Total A. Cyclone rehabilitation and corrective rehabilitation works B. Environmental assessment studies and protection works 1.4 2.7 4.1 C. Rehabilitation of additional schemes 0.6 0.9 1.5 D. Support for services of Ministry of Agriculture 2.8 5.4 8.2 E. Human resource development 2.5 2.7 5.2 F. Monitoring and Evaluation 0.4 0.3 0.7 G. Operations and maintenance by WUAs 0.1 0.1 0.2 1.3 0.0 1.3 Base Cost 9.1 12.1 21.2 Physical contingencies 0.9 1.2 2.1 Price contingencies 1.4 1.0 2.4 Total Project Costs 11.4 14.3 25.7 a/ Includes US$2.7 million of taxes and duties Proposed Project Financing US$ million Percent IDA 21.2 82.6 Beneficiaries 1.7 6.7 Government 2.8 10.7 Total 25.7 100.0 Estimated IDA Disbursements (US$ millions) FY95 FY96 FY97 FY98 FY99 FY00 FY01 6 ms 6 ms Annual 0.7 1.5 2.0 4.0 4.8 5.2 3.0 Cumulative 0.7 2.2 4.2 8.2 13.0 18.2 21.2 Percentage 3.0 10.0 20.0 39.0 61.0 86.0 100.0 Agriculture Sector Disbursement Profile 0.0 10.0 22.0 38.0 58.0 82.0 94.0 iii Project Benefits: The project will result in a better use of irrigation infrastructure and increased production and revenue for about 40,000 small scale farmers directly benefitting from the project. It will lead to reduced Government financial obligations to the irrigation sub-sector in the long term. One of the main benefits of successful implementation will be the strengthening of the agricultural extension and rural works services of the Ministry of Agriculture. The project will improve the capacity of the Ministry in environmental assessment of proposed irrigation rehabilitation projects. Environmental Impact: Improved drainage of stagnant waters would have a beneficial effect on the accumulation of salinity and mineral deposits in the soils and the elimination of breeding grounds for vectors of diseases such as malaria and bilharzia. In addition to selection criteria based on water users' commitment to taking over O&M responsibility, economic viability and proximity to road and market outlets, selection criteria of schemes for rehabilitation will include an assessment of environment risks in the catchment areas to minimize subsequent silting and high maintenance costs. Crop intensity recommendations involve low levels of insecticides and fertilizers. Risks: Project sustainability depends upon the water users effectively taking over responsibility for the rehabilitated schemes. Experience under the first project has not been decisive on this point. Part of the cause was the lack of sufficient involvement of water users in the process of defining and deciding rehabilitation proposals. This is being addressed under the project. In order to reduce the risk of insufficient commitment by water users, upfront contributions, in the form of cash or of labor for maintenance will be monitored at specific intervals before new investments in rehabilitation are confirmed. To redress the modest improvements in production and productivity registered under the first phase project, support will be provided to agricultural extension and research and development activities. Although the increased production generated by the project is relatively modest, the criteria for the selection of additional schemes for rehabilitation will continue to include satisfactory access to a road network, in order to avoid the risk of problems created by insufficient market outlets. Internal Rate of Return: The internal rate of return is estimated at 27 percent on benefit and cost streams projected over 10 years. Poverty Category - Program of Targeted Interventions: 34 percent of the population in Madagascar are estimated to be affected by food insecurity. The project will contribute to reducing food insecurity in rural areas and will increase rural incomes among smallholder rice producers, who are among the poorest of the rural population. Map: IBRD No25595 - Second Irrigation Rehabilitation Project  MADAGASCAR SECOND IRRIGATION REHABILITATION PROJECT STAFF APPRAISAL REPORT I. BACKGROUND A. Project Background 1.1 In 1985, the Government of Madagascar obtained funding from IDA, the French Government and the European Development Fund for a project which was to establish a systematic approach to the rehabilitation of medium- and small-sized irrigation schemes. The objective was to move away from a reliance on government management and funding, of the schemes known as the Petits P6rimatres Irriguds (PPIs), which had been characterized since the 1970s by poor maintenance, infrastructure deterioration and excessive public expenditure . The project was to be the first phase of a long term program which would gradually cover the country's publicly-funded medium- and small-scale irrigation schemes. It would strengthen the institutions responsible for selection, preparation and implementation of rehabilitation, encourage the participation of private consulting engineering companies and develop local maintenance responsibility and capacity. A major objective was to transfer responsibility for future rehabilitation from the government to the private sector, notably through the introduction of Water Users' Associations. 1.2 Seven years after start-up of the first phase project, whose physical implementation impact has been positive, the government invited the FAO Investment Center to prepare a second phase of the project, which would build on the experience of the first. The preparation mission for this second phase took place in October 1992 and was followed by an appraisal mission by the Bank in March/April 1993. Representatives of the European Development Fund and the Caisse Frangaise de D6veloppement also contributed to the work of the mission. At the same time, the Agricultural and Environment Division of the South-Central and Indian Ocean Department of the Africa Region of the Bank produced a discussion paper outlining a strategy for strengthening the contribution of the irrigation sub-sector to agricultural growth in Madagascar. Elements of this strategy were endorsed by Government in early 1993 and incorporated in the design of the second phase project, details of which are presented below. B. Economic Background General 1.3 Madagascar is the world's fourth largest island, 1,500 km long, with an area of 590,000 km2 (slightly larger than France). It is characterized by a rough topography, variations in soils and climate, some mineral endowment and unique flora and fauna. Rice is the country's main staple and the major exports are coffee, vanilla, cloves and shellfish. The population in 1993 is estimated at about 12.4 million inhabitants, growing at a rate of 2.8 per cent per year; its distribution is very uneven, with an average of 18.5 per km2, but exceeding 100 per km2 in the central highlands. Close to half of the urban population of 3 million lives in the capital, Antananarivo. 2 Economic Performance 1.4 With a per capita income of US$230 in 1993, Madagascar is one of the poorest countries in the world. The agriculturally dominated economy, which had grown modestly in the late 1980s, following a decade of decline, was disrupted by political developments since 1991. Output fell by almost 6 percent in 1991, recovering to only I and 2 percent respectively in 1992 and 1993, accompanied by annual inflation of 13 percent for the past two years. 1.5 The poor economic performance of the 1970s and early 1980s stemmed largely from inappropriate economic policies, with emphasis on public investment financed from external borrowing - mostly for economically non-viable projects, industrialization, nationalization of foreign-owned enterprises, extensive consumer subsidies and pervasive controls over private economic activity. Structural Adjustment Policies and Results 1.6 Policy adjustment between 1981 and 1986 aimed at financial stabilization and centered on demand management; during that period, the fiscal deficit fell from 15 percent to 4 percent of GDP through reduced expenditures, inflation dropped from 30 percent to 13 percent per annum, and imports were reduced by more than half in real terms. 1.7 Starting in 1985, the government's reform program was supported by four IDA sector adjustment credits covering industry, agriculture, trade, and the public sector. A number of important reforms were undertaken with the support of these credits. In industry, the government eliminated most price controls and export taxes, reduced controls on profit margins, and began to dismantle the heavy administrative regulations, which were a disincentive to private activity. In agriculture, it liberalized the pricing and domestic marketing of rice (Madagascar's main staple), eliminated most state monopolies, and abolished export taxes except for vanilla. In the external sector, the government substantially eased administrative export restrictions, greatly enhanced the country's international competitiveness through a series of devaluations of the Malagasy franc (Fmg), ended quantitative import restrictions, introduced an automatic allocation of foreign exchange for merchandise imports through an open general licensing (OGL) system, and initiated major tariff reform to reduce effective protection. An improved budgetary system was introduced, and the process for public investment programming was improved. Steps were taken to withdraw the state from productive activities, and to close down a number of long- standing projects which absorbed an inordinate amount of domestic financing. Financial institutions were partially privatized, and the sector was opened to two new privately-owned banks. Finally, a new domestic petroleum price mechanism was adopted; under this mechanism, prices are automatically adjusted based on international prices, the exchange rate, and the cost of transport and unloading. 1.8 Economic management suffered during the political transition starting in 1991. Budgetary discipline deteriorated; monetary management was lax; external arrears accumulated; the OGL system was replaced by rationing of foreign exchange in 1991; large-scale import restrictions were re-imposed in early 1994 as a way to address foreign exchange shortages; divestiture of public enterprises was slow, with only about 50 enterprises out of 170 divested and the largest, and those with the most acute financial and economic problems, remaining in the public sector; ill-advised borrowing schemes were explored; and reforms of the legal and regulatory framework to improve the business environment were discontinued. While Madagascar maintained military 3 expenditures at a relatively low level, expenditures on general administration and indirect transfers to public enterprises represented an important burden on the economy. Because of the shortcomings in economic management and the government's inability to address some of the major issues affecting the public sector, on September 30, 1993, the Association closed the FY88 public sector adjustment credit and canceled its third tranche. 1.9 In March 1994, the prime minister publicly stated his government's intention to resume discussions with the Bank and Fund with a view to reaching agreement on a reform program. During a visit to Washington in late March/early April, he agreed to take strong preliminary measures to show his commitment to reform, including immediate steps to correct the overvaluation of the Malagasy franc (Fmg), eliminate import restrictions and revise the 1994 budget and public investment program to reduce the deficit and improve resource allocation. These steps were all taken in early May, including the floating of the Fmg, and a joint Bank/IMF mission is in Madagascar to negotiate a PFP. C. The Agricultural Sector Agriculture in the Economy 1.10 Agriculture is the most important contributor to the national economy, contributing 33 percent to gross domestic product. The sector employs over 80 percent of the active population and accounts for over 60 percent of export earnings (during the early 1980s, agricultural exports represented 85 percent of total exports, and over 90 percent if processed agricultural products were added). The sector supplies the primary inputs to the food and textile industries which together account for 75 percent of the value added of the industrial sector. The economic stagnation which has affected other sectors of the economy more severely than agriculture, has led to an increase in agriculture's share of GDP from 31 percent in 1965 to 33 percent in 1993. 1.11 The agricultural sector grew at an average annual rate of 0.6 percent between 1970 and 1980, well below the estimated population growth rate of about 3 percent and stagnated between 1980 and 1983. While macro-economic stabilization policies concentrated on controlling inflation and public and current account deficits, the provision of services and infrastructure maintenance continued to deteriorate. Trade liberalization measures and elimination of export taxation measures were initiated in 1983 and completed in 1992 for most sub-sectors, with the exception of vanilla, cotton, sugar, tobacco and wheat. From 1984 to 1988, the agriculture sector grew on average at slightly more than 2 percent per year. After a surge of more than 5 percent in 1989, agricultural growth has again been sluggish because of the combined effects of poor weather and political change, reaching 2.4 percent in 1990 and 1.6 percent in 1991. 1.12 Although most tropical and temperate climate crops can be found in Madagascar, only about 2.3 million ha, i.e. about 6 percent of the country's arable land, are cultivated. Rice production is the most important agricultural activity, accounting for 41 percent of agricultural output (1992). Fishing is the second most important activity (20 percent of agricultural production). The traditional export crops (coffee, vanilla, cloves and pepper) together account for 7 percent of agricultural output. Root crops and maize contribute 11 percent. Vegetables and fruits contribute 10 percent, industrial crops 3 percent and livestock 5 percent. While agriculture and agro-industries are still contributing about 80 percent to total exports, there has been a shift in the relative importance of various products. The share of traditional exports has declined to about 40 percent in 1991, while the share of non-traditional products, such as shrimps and 4 lobsters, which now accounts for about 15 percent of total exports, has been rapidly growing by over 20 percent per year since 1986. Recent Agricultural Developments 1.13 The response from the agricultural sector to the adjustment measures throughout the 1980s has been mixed, varying across sub-sectors and regions. Fisheries and non-traditional exports, which benefitted from early trade liberalization and adjustment measures, significantly expanded after 1983; similarly, non-rice agricultural activities in the central highlands - where communications are more favorable (proximity to large urban markets, adequacy of physical infrastructure and agricultural services) - responded positively to improved price incentives (dairy products, potatoes, sweet potatoes, maize, barley, wheat). On the contrary, the traditional exports (coffee, vanilla, cloves, pepper) and cotton sub-sectors and their corresponding regions (East Coast, Northern region, West Coast) did not respond to the adjustment measures, partly because of the drop in world market prices (coffee, cloves) and partly due to the continued deterioration of communications and the absence of full trade liberalization (vanilla). The production of rice, after first responding to the liberalization of the rice trade in 1986, has also stagnated because of poor market infrastructure (roads, markets and storage). Potential for Agricultural Growth 1.14 Madagascar possesses a considerable potential for agricultural growth. It has a comparative advantage in agricultural production based on low labor costs and availability of land in different ecological zones. It can produce a variety of temperate and tropical crops for the domestic and export markets. It possesses an extensive irrigation infrastructure, largely underutilized and representing an important asset to enable intensification and diversification. Moreover, crop yields are uniformly low and the potential to increase yields on the basis of known technologies is important. Constraints to Agricultural Development 1.15 Sector adjustment reforms undertaken by Madagascar in the recent past such as the liberalization of domestic trade have improved prices and market demand for agricultural produce but have produced only a limited supply response so far. Sustained and spontaneous growth of the sector is constrained by the continuing lack of reliable transport and communication infrastructure linking farmers with a market, poor control and lack of management of water resources, inadequate protection of watershed areas and poor agricultural support services and transfer of technology. Public investment in the agricultural sector has concentrated too heavily over recent years on undertaking activities which could be carried out by farmers themselves and too little on providing basic services and creating an environment to enable farmers to fend better for themselves. Government Strategy for Agriculture 1.16 Government has begun to reformulate its strategy for agricultural and rural development. It recognizes that a rural development policy based on the intensification of the small farm sector 5 in the high potential areas of Madagascar would accelerate the development of the entire rural non-farm economy, increase the size of the domestic market and provide sizeable farm and off- farm employment opportunities and environmental benefits. The driving force for agricultural growth and development would be the increased integration of the agricultural sector into the market economy. Actions aimed at establishing the appropriate environment for market development would represent a powerful instrument for boosting agricultural growth and incomes. The strategy calls for: (a) the removal of the remaining macro-economic distortions; (b) provision of adequate rural infrastructure (transport and communications, irrigation, rural market infrastructure), and regulatory framework (business legislation, agricultural institutions, rural financial policies) in the rural areas with a view to promote the intensification of the small- farm sector and the rural non farm sector, and to facilitate inter-regional trade; (c) strengthening of the public sector capacity for coordination, policy and strategy formulation, administration and monitoring of implementation; (d) provision by the public sector of cost-effective market support services; (e) decentralization of decision-making and financial responsibility to local communities and private operators; (f) more effective cost recovery on publicly funded investments; and (g) development of viable private sector organizations (producers' associations, water users' associations, professional and inter-professional organizations etc) that can progressively take over support services (inputs and products marketing, certain areas of applied research, extension, credit, management of irrigation schemes etc). Ongoing discussions of the 1994-96 public expenditure program have concentrated on ways to reduce the number of investment projects and concentrate public expenditure on activities directly linked to the stated strategy. D. The Irrigation Sub-Sector 1.17 The agricultural sector is characterized by the relative importance of irrigation and by the long tradition of its farmers in irrigation. After Sudan, Madagascar has the second largest irrigated agriculture sector in sub-saharan Africa. More than one million hectares or about 50 percent of the total cultivated area of the country is under irrigation or potentially irrigable. It is estimated that the irrigation sub-sector contributes almost 15 percent of Madagascar's GDP, 70 percent of agricultural production and 88 percent of total rice production and employs directly or indirectly 85 percent of the active population. Irrigation schemes are located throughout the country except in the south and in the far north, where possibilities for irrigation are more limited. While almost all irrigation is based on capturing surface water and directing it by gravity to final users, there is a broad diversity of size, technology and organization of irrigated systems. They include the large irrigated schemes (GPIs - covering about 120,000 hectares) of between 3,000 and 30,000 hectares, with sophisticated water control and distribution networks, the medium-sized schemes (PPIs - covering about 170,000 hectares), often with sophisticated infrastructure , the micro-schemes (MPIs - covering about 500,000 hectares) and the individual family schemes (PFs - representing an estimated 300,000 hectares. About two thirds of the area under irrigation is entirely managed and maintained by small scale resident farmers, without any intervention on the part of the Government. The remainder is made up of schemes which have been developed by the Government before and since independence, under major public sector investment projects designed to increase agricultural production. The potential for new irrigation schemes is limited, given that about 80 percent of land potentially irrigable is already under irrigation. Possible expansion is confined to those large scale schemes or areas around them, situated in the broad alluvial plains of the west coast. 6 Irrigation Development Strategy 1.18 The Government's irrigation subsector development strategy (see declaration of sector development policy in Annex 2 is consistent with the overall sector development strategy. The Government maintains its commitment to providing adequate rural infrastructure by pursuing ways of rehabilitating the irrigation network under its charge (Grand P6rimtres Irrigu6s and Petits P6rim6tres Irrigu6s). Over recent years, almost 50 percent of the Ministry of Agriculture's budget has been used to maintain and support these publicly-managed schemes. The Government recognizes that a reduction in the level of investments in these schemes would release scarce funds to support other services, such as research, extension and rural roads. With this objective in view, and with the aim of decentralizing financial responsibility to local communities and achieving more effective cost recovery, the government has begun to withdraw gradually from the operation and maintenance of irrigated schemes. By encouraging and providing support for Water Users' Associations (WUAs), the government aims to transfer these functions to the farmers themselves. Local enterprises will be encouraged to develop capacity to undertake rehabilitation and maintenance work on behalf of local WUAs. The role of the services of the Ministry of Agriculture will be redefined to respond to this change of responsibilities. A considerable intensification of the dialogue between water users and the administration will be required to achieve the commitment necessary to ensure sustainable transfer of responsibility for operations and maintenance (O&M). The rationale for the proposed investments under the ongoing and the proposed irrigation rehabilitation projects is based upon supporting the government's efforts to this end. 1.19 Considerable public resources have been invested in irrigation, especially in the large scale and medium-sized schemes (GPIs and PPIs). These schemes were established by the national government (or by the colonial government before independence) and both management and maintenance are provided either directly, through the Rural Works Directorate (Direction du G6nie Rurale) or by way of parastatal companies (for the GPIs). Lack of maintenance of these schemes has led to the increasing need for heavy investments since the 1970s by the Government with funding from external donors. Currently investments in irrigation rehabilitation absorb about half of the investment budget of the Ministry of Agriculture (1993 PIP) and a high proportion of the Ministry's recurrent budget. In an attempt to secure the transfer of the financial and physical responsibility for the maintenance of these schemes, the government has embarked upon a program to encourage the creation of Water Users' Associations, to whom the responsibility for O&M will be transferred. It has also started to dissolve the parastatal companies responsible for the large scale irrigated schemes, although so far with little provision for structures to replace them (e.g SOMALAC). 1.20 The irrigation subsector benefits from considerable external financial and technical assistance. In addition to cofinancing the first PPI irrigation rehabilitation project, the Caisse Frangaise pour le D6veloppement (CFD) is funding rehabilitation work on the south-east of the Lac Alaotra scheme, as well as environment protection on the Imamba-Ivakaka to the west of Lac Alaotra. The CFD has undertaken a feasibility study of repair and rehabilitation work on the Morondava GPI, which suffered from cyclone damage in 1991. The European Development Fund (EDF) is funding irrigation rehabilitation in the south west around Toliara, as part of the first PPI rehabilitation project and is preparing a project, which will include irrigation rehabilitation in the north-east. In addition, the EDF is funding a program in support of rehabilitation of micro- irrigation schemes. The German Government with KfW is funding a rice development project in the north-west at Betsiboka, which is also relying on water users' associations to ensure long term sustainability. In addition to the current support it is providing to the Directorate for Rural 7 Works (DGR), the African Development Bank (ADB) has proposed to finance the rehabilitation of irrigation schemes in the south of Madagascar. 1.21 In spite of the significant levels of public expenditure, the performance of the irrigation sector remains weak and well below its potential. In particular, yields of agricultural production have stagnated (national yields of rice are below 2 tons/ha), there has been little attempt to diversify (rice remains the main crop) and the levels of cost recovery have remained low (virtually no contributions were received from beneficiaries of medium-scale irrigation schemes rehabilitated since 1985, until 1993, when some WUAs finally started to collect dues). 1.22 On the basis of this recent performance, the continued high level of public sector funding in support of irrigation is hard to justify. However, Government continues to expect irrigated agriculture to make a major contribution to meet the food deficit, which is growing as population expands faster than growth in agricultural production. Irrigation sector policy should seek therefore to define solutions which will (a) reduce the call on public sector funding for irrigation and (b) tackle those constraints which have been at the roots of past poor performance. These include (i) clarification of the responsibilities of the state and provision of resources to meet these responsibilities; (ii) clarification of the responsibilities of the water users for operations and maintenance of rehabilitated irrigation schemes; (iii) strengthening the technical advisory services available to farmers and WUAs; (iv) protection of the environment on and around irrigation schemes; (v) reinforcement of land tenure security; (vi) provision of the means to respond to emergency conditions threatening rural infrastructure; and (vii) coordination of institutions involved in the development of the sector including external donors. The proposed project tackles some of these critical constraints. E. Administrative Organization 1.23 The 1975 Constitution divided the political and administrative structure of the country into four separate levels of local government, called decentralized collectivities. They include 11,393 Fokontany (village), 1,252 Firaisina (group of Fokontany), 110 Fivondronana (department) and 6 Faritany (province). Each local government had a popular council, an executive committee (both elected), and an administrative committee (a technical body composed of internal departments and representatives of local departments of all ministries). The lack of an official text specifying areas of intervention for each level of governing body has resulted in dilution of responsibilities, confusion, and delays associated with approval procedures. All local governments are tied to central government through a system of hierarchical supervisory authorities. While it is not expected that the local administration will be associated directly with the execution of the project, the local Fokontany in the individual project areas will contribute to sensitizing the local farming population to the advantages and responsibilities of the water users' associations. F. Bank Group Involvement IDA Lending Strategy for Madagascar 1.24 The main objectives of IDA lending to Madagascar are to create sustained and equitable export-led growth, supporting poverty alleviation, improved environmental management and enhancing project implementation capacity. The strategy to achieve this objective concentrates 8 on areas where the Bank has a comparative advantage, avoiding direct involvement where either the private sector or other donors can more effectively take the lead. The priority for promoting private sector and export-led growth will be to encourage the establishment of sound macroeconomic management, including a market-determined exchange rate, prudent monetary and credit policies, fiscal restraint and domestic resource mobilization. The liberalization process in foreign exchange allocation, tariffs and pricing will be completed. Attention will be given to improving financial intermediation, simplifying the regulatory framework and eliminating public and private sector monopolies. Investment projects will address infrastructure constraints and major problems in the delivery of services in the social sectors. Support for poverty alleviation will continue through ongoing programs to enhance food security, to generate employment and to organize emergency actions for vulnerable groups. New infrastructure and social sector projects will be designed so as to maximize efforts towards poverty reduction. Improvement to environmental management will continue to be encouraged through the free-standing environment program which is being supported by the Bank and other donors. New projects will promote environmentally compatible development and population control. In order to improve project implementation capacity, the Bank will devote more resources to project supervision and undertake regular public expenditure reviews to ensure that IDA-financed projects fit with agreed sector strategies. Institution and capacity building will be important aspects of the program. Recent Lending Experience 1.25 With support from the Bank and other donors, Madagascar has embarked on a series of operations in agriculture, infrastructure and human resources, which have been designed to implement the above strategy. Specific actions have included (a) infrastructure improvement; (b) removal of remaining marketing constraints; (c) investment in research and extension; (d) irrigation rehabilitation; (e) education sector development; and (f) health sector development. However, the execution of many of these programs has been hampered by the weakness and bureaucracy of the central administration combined with an unwillingness to delegate control of resources outside of the capital. Funding has emphasized construction and rehabilitation rather than maintenance. Preference has been given to public enterprises to carry out project implementation rather than to encourage local private enterprises. There has been insufficient involvement of local communities in the definition of development priorities and in the execution of the programs designed to achieve these development objectives. 1.26 Experience with lending to the irrigation sub-sector in Madagascar has been disappointing. The first IDA-financed operation was designed to improve irrigation and drainage of some 12,000 ha in the Lac Alaotra area. The project was carried out between 1970 and 1976. The post evaluation impact of the project was initially considered satisfactory, based on satisfactory completion of physical works, but a later review concluded that the project had failed to meet its fundamental goals, given the lack of sustainable impact on production or farmer incomes. Poor maintenance, ineffectiveness of the extension services and environmental problems resulting from excessive drainage were cited as reasons for poor performance. The project had a negative rate of return. This project was followed by the Morondava irrigation and rural development project which was implemented between 1974 and 1982. There was considerable divergence between the Bank and the Government of Madagascar on the project objectives and components. Benefits were low and the project's rate of return was zero or negative. A later phase of irrigation development on Lac Alaotra followed between 1984 and 1991, with cofinancing from CFD and covering 35,000 ha managed by the former parastatal SOMALAC. This project was rated satisfactory, having successfully achieved its physical objectives , with 9 paddy yields and production increasing significantly. However, sustainability was not assured because of uncertainty regarding maintenance arrangements. A fourth project in support of irrigation rehabilitation, the Mangoky agricultural development project, was implemented between 1979 and 1986. As with the preceding projects, implementation was the responsibility of a parastatal organization, whose inability to execute works or coordinate activities hampered the project from the start. Crop yields and project benefits were considerably lower than anticipated and once again the project terminated with a negative rate of return. Over-sophistication in the technical and managerial design of the project were responsible for its poor performance. The ex-post impact evaluation recommended, among other things, the need for private sector involvement and close collaboration with water users and water user associations. It was only with the implementation of the first Irrigation rehabilitation project (detailed analysis is presented below), which started in 1985, that a more successful formula for irrigation development and rehabilitation was developed. Lessons from Previous IDA Experience 1.27 Experience with irrigation rehabilitation in both Madagascar and elsewhere has demonstrated the importance of satisfactory cost recovery systems to ensure long term sustainability of irrigation schemes. In turn, unsatisfactory cost recovery has been linked to factors such as poor government commitment, unreliable water supply from inefficient irrigation authorities and unprofitable farming resulting from discouraging price policies. 1.28 The design of the proposed project has taken account of these lessons and builds on past experience. Considerable involvement of water users is envisaged at an early stage and throughout the rehabilitation process. Maintenance and rehabilitation will be ensured by charges raised by Water Users' Associations on their members. There are no parastatal agencies involved and attempts will be made to make the water users themselves responsible for reliable water supply. Local regional services of the Ministry of Agriculture will be strengthened to support local farmers' organizations and provide professional agricultural extension services for a variety of crop and livestock enterprises. The development of local enterprises to ensure longer term maintenance of the irrigation networks will be encouraged by sub-contracting rehabilitation and maintenance work locally and by developing a mechanism for regular consultation between project beneficiaries and representatives of the local administration. 10 II. THE FIRST IRRIGATION REHABILITATION PROJECT A. Project Implementation 2.1 Objectives. The first Irrigation Rehabilitation project (Cr. 1589-MAG) was designed to fund the rehabilitation of about 16 medium-sized irrigation networks, irrigating a total area of about 17,000 ha and of the large-scale SAMANGOKY scheme. During project execution, estimated to last for five years, but extended to eight years, a systematic approach for scheme rehabilitation was to be developed, which could be progressively applied to the remainder of the country's publicly-managed schemes. 2.2 Implementation. Due to political and social problems encountered in late 1989, the SAMANGOKY component of the project was dropped. Progress on the remaining three components of the project, namely (a) the inventory of schemes to be rehabilitated and their preselection; (b) the socio-economic and technical studies for rehabilitation and (c) physical rehabilitation work has been as follows: (a) a total of 312 irrigated schemes representing a total of 146,000 ha have been inventoried. This inventory led to the identification of some 140 schemes, representing a total of 66,000 ha, which could potentially be rehabilitated based on criteria which included the motivation of future beneficiaries, the appropriateness of rehabilitation in the light of the local regional development strategy, road access, the existence of development support services such as extension and credit and an internal rate of return above 14 percent. The impact of the local environment was not among these criteria, but has since become a major focus for the second phase project. (b) socio-economic and technical studies were carried out by local consulting firms and included a local physical and socio-economic survey of the scheme, a prefeasibility and feasibility study of the proposed rehabilitation, preparation of tender documents for the works, a management handbook of the scheme and an agricultural development plan; (c) As of December 31, 1993, the first project had completed the rehabilitation of 13,000 ha (76 percent of appraisal estimates) on 21 schemes and work is advanced on the rehabilitation of a further 4,200 ha on 9 schemes. However, the number of schemes on which supplementary "corrective" rehabilitation work has been required has lessened the impact of these physical realizations. 84 Water Users' Associations have been created on a total of 39 irrigated schemes, comprising more than 29,000 farmers on 17,200 ha of irrigated farmland. Agricultural yields and production has increased on certain schemes by as much as 25 percent, largely due to improved water management but with little extension input, which could have significantly increased these improvements. The estimated rate ERR of the project is 18 percent, compared with 24 percent at appraisal, the difference being due in part to the increased unit costs of rehabilitation. The transfer of the rehabilitated schemes has yet to be fully effective, since the levels of financial contributions to maintenance and operations of the schemes by WUA members, the critical test for sustainability, are still low. 2.3 Cofinancing was provided by the Caisse Franqaise pour le D6veloppement (CFD, formerly CCCE) and the European Development Fund (EDF). Each cofinancier concentrated its support upon specific geographic areas, in such a way that the CFD funded the rehabilitation of the schemes in Antananarivo and Fianarantsoa provinces, the EDF the schemes in Toliara and 11 Mahajanga province and IDA the schemes in Fianarantsoa and Mahajanga. Differences of opinion as to the time and detail required prior to rehabilitation meant that work on CFD and EDF schemes started earlier than IDA-funded schemes. However, implementation on schemes funded by these two donors slowed after a review of the first five years' experience. There is now more of a consensus on the pace at which sustainable rehabilitation can be implemented, although differences persist as to the intensity of technical assistance support required to achieve sustainable results. B. Lessons from the First Project 2.4 Inventory Study. The inventory study of the schemes requiring rehabilitation permitted the preparation of a Master Plan for rehabilitation, but the study had two shortcomings. Firstly, the real motivation of the beneficiaries was not assessed in depth at inception and the consequent remedies for encouraging the water user associations to take responsibility for the maintenance of their schemes were applied without due regard to the individual characteristics and prerogatives of the different groups. Secondly, the environmental aspects of the rehabilitation proposals were underestimated, leading in several cases to important difficulties in the maintenance of the schemes due to excessive sedimentation. 2.5 Feasibility Studies. The processing and approval of the technical feasibility studies for rehabilitation were handled centrally by the project Coordination Unit and the Ministry of Agriculture technical services. The local services of the Directorate for Rural Works (DGR) and the beneficiaries were insufficiently involved and consequently felt dissociated from the process and the project. This led to insufficient supervision by the local DGR services, which in turn resulted, in some cases, in (i) technical design faults; (ii) an overestimation of available water resources; (iii) a lack of appreciation of traditional irrigation systems; and (iv) over-sophisticated design. 2.6 Definition of Responsibilities. The experience of the first project has shown that the responsibilities of the parties involved in the project must be clearly defined, understood and accepted by all from an early stage. This is particularly true for (i) the local technical and extension services of the DGR and of the Directorate for Agriculture (DA); (ii) the share of responsibilities for O&M between the government and the Water Users' Associations; (iii) the responsibilities of the design and supervisory consultants. 2.7 Costs. Unit costs of rehabilitation of the schemes increased disproportionately with other economic indicators during the project period. This was due in part to (i) an underestimation of the difficulties involved; (ii) excessive stacking of mobilization and demobilization costs by entrepreneurs; (ii) contract completion and payment delays; (iii) problems of communication to and from the interior; and (iv) the limited number of bidders and thus lack of competition for individual works. Two further potential sources for excessive increases in costs are related to the inability of Government to meet its 35 percent contribution to civil works costs, allowance for which was gradually factored into bids by all contractors for new works. Additionally, operational costs of the project coordinating unit were included in the overheads of supervision contracts, as a way for making up for the lack of government contribution to these costs. For the next phase of the project, there is a need for a closer control of quality in the feasibility studies and the definition of standard designs to bring down unit costs. More discipline in the supervision and monitoring of works contracts will help to keep actual costs within the set limits of the contract bids. Regular and effective payment for contracts completed on schedule must be 12 assured and no further dependence by the project coordinating unit on suppliers' contracts will be tolerated. 2.8 Environmental Impact. The underestimation of the environmental impact of the project has had two contradictory effects. On the one hand, the degradation of the catchment areas above the schemes reduced the quantity of water available to the scheme headworks and increased the risks of flooding and silting of the irrigation works. On the other hand, the environmental impact of the project resulting from the rehabilitation of the schemes appears to have been negligible or at best, positive. An assessment of the environmental impact of new and existing schemes will help to determine the justification for further investments. 2.9 Water Users' Associations (WUAs). Establishment of WUAs under the first project was initially guided only by the proposals in the project appraisal report and the rather ambivalent 1985 law which defined the operational structure for WUAs and the obligations of their members. These included participation in maintenance of schemes; respect of rules and policing; annual payment of maintenance fees; and participation by users in the investment costs of rehabilitation. As problems arose, project management modified procedures and obtained legislative action for the creation of WUAs, which facilitated adaptation to changing circumstances. The main modifications have been concerned with (a) acquiring greater participation by users, as rehabilitation took place; (b) relaxing the rigid operating principles of the WUAs; (c) adapting the procedures to facilitate the adoption of operations and maintenance and policing of the rehabilitated schemes; and (d) reinforcing support to the beneficiaries and the WUAs. In spite of these improvements, the level of financial contributions to O&M does not yet reflect a sufficient level of commitment. Reasons include (i) selection of the schemes for rehabilitation on the basis of criteria, which do not take full account of complex local desiderata and social forces; (ii) insufficient preparation of the WUAs and inadequate consultation with potential members about their future tasks; (iii) start-up of work on full scale rehabilitation before establishing local commitment; and (iv) a lack of legislation defining clearly the respective responsibilities of users and of government, once rehabilitation has been completed. 2.10 Legislation passed in July 1990, made specific reference to the definition of responsibility for the operations and maintenance and for the policing of irrigated schemes under a section entitled "transfer of infrastructure management" 1/. The detailed regulations of this law have been drafted, discussed with representatives of WUAs and formally adopted in March 1994. Staff from the DGR continue to provide certain maintenance services; some local officials meet the costs of rehabilitation work without requesting counterpart funding from the beneficiaries; others are reluctant to take measures against delinquent WUAs where dues are not paid regularly. On many schemes, the WUAs are unaware of their responsibilities. The detailed regulations, which are being finalized, will clarify procedures and policies to avoid future inconsistencies. 2.11 Agricultural Development. Different agricultural extension approaches were used on rehabilitated schemes, depending upon the resources available to the local Directorate for Agriculture (DA) or to project entities in the areas benefitting from the project. The agricultural plans drawn up for the rehabilitated schemes only went into force once the rehabilitation works had been completed. Thus many farmers located on the upstream areas of schemes, who had access to existing water, received no immediate extension benefits until the whole scheme had I/ Section V, Article 16 of law no 90-016 of July 20, 1990. Article 17 specified the Minister responsible for the Public Domain as responsible for fixing the procedures for this transfer. 13 been completed. Technical messages concentrated on increased rice production. Although a monitoring and evaluation system was not effective until 1991, preliminary results estimate an improvement in rice yields from about 1.5 tons/ha to 2.4 tons/ha (national average 2 tons/ha). These results were due to the return to production of fields previously in fallow; to an improvement in water control; to an improvement of productivity in the rice fields; and to intensification of production resulting from double cropping in some areas. Considerably more impressive agricultural results could be obtained by an earlier introduction of better organized agricultural extension services and the strengthening of links with agricultural research services. 2.12 Institutional Structure. In addition to two technical Directorates (DGR and DA), project implementation involved a central coordination unit and two operational directorates of the Ministry of Agriculture (Programming and Finance, Human Resources Management). A central Coordinating Committee was intended to bring together the four main Directorates which were to participate in the project and the Ministry of Public Works. The Committee only met three or four times in the first two years and then ceased altogether. Due to the relative inactivity of this Committee, the national coordinator took almost the entire responsibility for the project, with less than full involvement of the other national directors. During project execution, insufficient attention was given to ensuring consultation among the different parties (rural works, agricultural development and farmer organizations) involved in the process of rehabilitation. For the future, the role of coordination should be better integrated within the existing structure of the Ministry of Agriculture. There will be a continued need for central coordination given that several separate directorates are responsible for project implementation. 2.13 Project coordinators were nominated at the provincial level but insufficient attention was given to defining their roles vis-A-vis the existing provincial directorates and they encountered difficulties similar to those of the central Coordination Unit. On the rehabilitated irrigation schemes, "Conseillers-Animateurs" (CA) were posted to supplement the work of the less qualified rural development agents (ADRs) and irrigation scheme chiefs (Chefs de R&seaux). The CAs made a sound contribution to facilitating the creation of the WUAs. Their role could be improved, however, if they are ensured greater stability in their assignments (they currently have short term contracts), regular budgetary support for their work and clearer responsibilities as part of the local team designated to ensure maximum contact and collaboration with WUAs at all stages of the process. 14 III. THE PROJECT A. Project Rationale 3.1 The second phase of the irrigation rehabilitation project would be a further tranche of a national program, which began in 1985, to rehabilitate and to transfer operations and maintenance responsibility for public irrigation schemes to water users. The program remains a high priority for the government since, if successful, it will mean the release of scarce public funds currently tied up in irrigation maintenance, which will be available for alternative development priorities. About 50 percent of the 1993 agriculture sector PIP was allocated to expenditures related to the maintenance or development of public irrigation schemes. The 1994 public investment program envisages a significant decrease in the proportion of such expenditures (to 34 percent) in support of public irrigation schemes. The project will contribute to this tendency to rationalize the public expenditure program for the agricultural sector. It will also ensure the more equitable deployment of public expenditure, which has over recent years benefitted farmers who cultivate land on public irrigation schemes. In these respects, the project will contribute to the realization of the Government's irrigation policy, as announced in Annex 2. The project is in line with the Bank's country assistance program in support of accelerated growth, poverty reduction and socially, financially and environmentally sustainable policies and programs. Lessons drawn from the project will contribute to the search for a sustainable solution for the operation and maintenance of the large scale irrigation schemes whose rehabilitation has been funded by several donors over the past three decades but whose sustainability remains elusive. 3.2 Rehabilitation of irrigation schemes can lead to potential increases in productivity and intensification of production for smallholders, when accompanied by professional extension support. As such the project responds to the objective of improved food security, supported jointly by the government and the Bank. More progress can be made in this area by adopting the system of extension management and training currently being introduced under the national agricultural extension project (PNVA, Cr. 2150-MAG). B. Project Objectives and Design 3.3 Objectives. The project would support the country development strategy on four fronts: (a) Promoting growth. It would contribute to increased agricultural output through better water management and intensified agricultural extension. (b) Poverty reduction - it would contribute to improved rural incomes and food security by improving production and productivity on irrigation schemes in 5 of Madagascar's 6 rural provinces; (c) improving environmental management - it would develop techniques for environmental assessment of irrigation rehabilitation and reinforce local capacity for environmental assessment; (d) enhancing project implementation capacity - it would establish operational procedures for the transfer of management and financial responsibility for the operations and maintenance of public irrigation schemes to water users and thus would serve to reduce the burden of public expenditure on the maintenance of such irrigation infrastructure. The project would build on experience gained under the first Irrigation Rehabilitation project. It would strengthen the effectiveness of decentralized public services and the control and supervision of public works at regional and local levels. Through the decentralization and strengthening of the Ministry of Agriculture's regional activities, the project would contribute to improving coordination of activities of local projects and programs, currently operating independently. It would contribute to improve levels of income of the rural population 15 using accessible and proven technology and would serve to promote private sector initiatives, local consultants and engineering services. It would also establish criteria for the selection of schemes to be rehabilitated taking account of environmental implications. 3.4 Another project objective will be to contribute to a change in the approach towards community development. This change should result from realigning the present roles of the state and of the local communities and be reflected in the attitudes and the behavior of government agents and of members of local communities. The members of the Water Users' Associations would become the real managers and operators of their irrigation schemes, while the state, through its agents, would become a supplier of services. To achieve this ambitious objective and to change the perception of the role of state and farmers, techniques will be developed which will ensure closer collaboration between local beneficiaries of project activities and representatives of the local and central administration. 3.5 In the longer term, the responsibility for operation and maintenance of the irrigation schemes which will have been rehabilitated under the first and second projects will revert to the water users on these schemes. In order to ensure successful transfer, the Government needs to invest funds to render the schemes operational. In many cases schemes have fallen into such disrepair that it would be unrealistic to expect the water users to assume financial and operational responsibility without such investments. Financing under this project is designed to meet this need. Following rehabilitation, although the transfer of the schemes' operations and maintenance is assumed by the water users, the schemes still remain public assets. In its role as proprietor of this asset the State will still retain certain responsibilities for maintenance and reconstruction of complex headworks and primary canal systems. As the responsibility for each scheme is transferred to water users, the detailed breakdown of responsibilities between the State and farmers will be defined in a manual, to be prepared during the detailed phase of definition of the rehabilitation works and the costing of their maintenance and operation. 3.6 Since the objective of transferring the responsibility for rehabilitated infrastructure goes beyond the medium-sized irrigation schemes (PPIs), covered by the first project, the lessons and the experience drawn from the project will provide guidelines for future efforts on some large- scale irrigation schemes, which have benefitted from recent rehabilitation works and whose sustainable upkeep and maintenance represents a major liability for public funding. 3.7 Main Features. The project will support activities (a) to transfer operations and maintenance (O&M) to water users on rehabilitated schemes from the first project and ensure full farmer commitment to the process; (b) to strengthen local agricultural advisory and technical services in support of the newly rehabilitated schemes; (c) to evaluate the environmental risks and measures to protect rehabilitation; and (d) to start on further rehabilitation using a more gradual approach which ensures user commitment as the process of rehabilitation goes forward. Responsibilities of the state and of water users will be defined and promulgated as part of new legal regulations. Means will be made available to enable the government to strengthen regional and local services, as part of a broader effort to decentralize activities of the administration. Resources are also included to support training, environmental studies and monitoring and evaluation activities. 3.8 The Project. The new project will begin by ensuring the successful transfer of operations and management of the 30 irrigation schemes rehabilitated under the first project. There are still certain works which are required to ensure the proper functioning of these schemes. In addition, emergency funding would be provided to repair some of the damage caused 16 to these schemes by cyclones Daisy and Geralda in early 1994. These extra works represent a valid testing ground for the introduction of more intensive farmer participation. At the same time, the vulnerability of the schemes to environmental degradation will be assessed. Once commitment by water users on the existing rehabilitated schemes has been established, then rehabilitation work on a further 4,000 ha of new irrigation schemes may proceed, on the evidence of proven commitment by beneficiaries (e.g. cash contributions to O&M or works carried out by the WUAs), which will be evaluated and reviewed as the process goes forward. New physical rehabilitation will be modest during this project (4,000 ha compared with 17,000 ha under the first project). This is due to the important emphasis, which will be placed on institutional strengthening and the adoption of techniques, including participative training, which will ensure long term sustainability of investments already made under the first project and the definition of an approach which, if successful, can be applied on a broader scale during subsequent projects. 3.9 The project will fund actions in the previous provinces of Mahajanga, Toliara, Antananarivo, Fianarantsoa, where rehabilitation works have already been carried out under the first phase project and also on the large scale irrigation schemes in the Lac Alaotra area. New rehabilitation will follow an indicative program traced out by the PPI Coordination Unit and based on the survey of irrigation schemes qualifying for rehabilitation, under the first project. Identification of the additional schemes which will benefit from the project is not possible at this stage, since the choice will depend upon evidence of commitment by the WUAs. 3.10 The Challenge. In spite of the problems encountered during the first phase, water users have never rejected the principle of assuming operating, maintenance and policing responsibilities of the irrigation schemes by their associations. The challenge of the proposed second phase project is to convert this willingness into concrete commitment and into the mobilization of funds by members of the WUAs. Proven results on this initial sample of irrigated perimeters (30/40) are essential to promote, as a next step, a larger scale transfer of public irrigation schemes to beneficiaries. C. Summary Project Description 3.11 The project would comprise six main components, which would be implemented progressively over a five-year period: (a) Cyclone damage repairs and corrective rehabilitation works, as a means to ensure the transfer of operations and maintenance to water users on schemes rehabilitated under the first project, subject to full farmer commitment to the process; (b) Environmental protection, to include an environmental assessment of existing rehabilitated schemes, definition of guidelines for environmental protection works on new schemes, local capacity building in environment assessment, implementation of protection works and the preparation of guidelines for natural disaster compensation. (c) Rehabilitation of new irrigation schemes - comprising programming in collaboration with beneficiaries, local staff and design consultants, preparation of feasibility studies and works execution. 17 (d) Support for the services of the Ministry of Agriculture and Rural Development to ensure central project coordination and more effective field service operations in nine decentralized Directorates, where the project will be active; (e) Training of staff and farmers with an emphasis on developing local capacity and talents to ensure sustainable project activities for future phases of the rehabilitation program; (f) Monitoring and evaluation of project impact and audit of accounts. 3.12 Total project costs, including the provision for price and physical contingencies, is estimated at FMG 89.5 million (US$25.7 million) of which FMG 49.7 million (US$14.3 million) would be in foreign exchange. The Ministry of Agriculture, by way of its technical directorates responsible for Rural Works and Agriculture, with the support of the directorates of programming and finance would be responsible for project execution. A national coordination unit, already established under the Secretary General to coordinate activities of the first project, would centralize monitoring and evaluation of project activities, would coordinate informally the activities of the project's executive directorates and would ensure consistency between project activities and broader policy options of the Ministry such as (a) the public expenditure program for the irrigation sub-sector and (b) the decentralization of activities of the Ministry of Agriculture. D. Detailed Project Description 3.13 Cyclone Damage Repair and Corrective Rehabilitation Works (US$4.1 million): The first two years of the project will concentrate on getting a commitment, as demonstrated by contributions in cash and in kind from members of the WUAs on schemes which have been rehabilitated under the first project. Through the intensification of relations between these associations and field staff (G6nie Rural, Agriculture staff and the Rural Development Agents - ADRs), support will be provided for the following: (a) implementation of the new regulations on the transfer of irrigation schemes, through local workshops during which the specific implications for individual schemes would be defined and agreed; (b) emergency repairs of damage caused by cyclones Daisy and Geralda, to be carried out during the first twelve months of the project; (c) implementation of corrective works (as identified by the diagnostic study to be carried out using PPF funding) and environment mitigation works (as identified by the environmental impact study, referred to below - para. 3.14) would be carried out to ensure satisfactory operation of the newly rehabilitated schemes, either directly by WUAs under the supervision of DGR staff or by consulting engineers and local construction companies under the supervision of regional and local Rural Works (GR) staff, in both cases in conjunction with water users; (d) intensification of agricultural extension measures in support of agricultural operations on the newly rehabilitated schemes, using an extension management approach adapted to 18 local requirements. The potential role of associations for women producers on the rehabilitated irrigated schemes will be given special attention. 3.14 Environmental Protection (US$1.5 million): The project will contribute to remedy or to avoid adverse environmental effects of rehabilitation on past and future rehabilitated schemes. Environmental assessments will be conducted prior to all future rehabilitation. Procedures, parameters and criteria to be included in this assessment will be based on the experience of the 30 irrigated schemes rehabilitated under the first project. Since the details of the environmental impact of rehabilitation on these schemes is only partially known, the project will fund a study of the impact of environmental degradation on the operation of these schemes. The study would be carried out in three phases (a) identification of environmental problems on each scheme; (b) definition of proposed solutions; and (c) implementation of recommendations. Phases (a) and (b) will be carried out by specialized consultants in conjunction with staff of the special Environment Unit to be established in the DGR. Implementation of the recommendations of the study would be carried out jointly by local water users, consulting engineers and construction companies depending upon the complexity of works involved and supervision would be the responsibility of the Rural Works Directorate staff. Decisions on the complexity of environment mitigation measures to be undertaken would be guided by consideration of economic viability, in line with criteria being developed under the National Environment Program (Cr. 2125-MAG). 3.15 To make the fullest use of this exercise, the project will support the preparation of an environmental assessment guide for irrigation schemes. The guide will make use of the conclusions of the study mentioned in para. 3.14. It will also include a synthesis of the approaches and techniques of watershed management currently known and applied in Madagascar's four main ecological regions. This guide will contribute to the assessment of environment mitigation measures to be included in the design of future work on the rehabilitation of new irrigated schemes. For all new rehabilitation, an environment assessment will be included as part of the detailed feasibility study, to be carried out in close collaboration with beneficiaries. 3.16 As part of an effort to strengthen the national capacity for environment assessment, the project would provide resources to train local consultants in the assessment of environment mitigation by organizing seminars based on the experience gained under the project. This should improve the capacity of local companies and contribute to more effective rehabilitation design and implementation in the future. Such training will be carried out in collaboration with similar efforts being launched under the Environment Action Program (Cr 2125-MAG). 3.17 The detrimental effects of cyclones and other major natural hazards before, during and after rehabilitation of selected irrigated schemes can be significant. Responsibility for making good the damages caused by such natural hazards is not clearly defined. In most cases, the extent of the damage goes beyond the financial capacity of the water users' own resources, particularly during the early years following rehabilitation. The principle of compensation from public funds in such cases is acknowledged by the government, which has a limited annual appropriation in the national budget for such liabilities. However, the procedures and criteria for allocating such funds do not automatically compensate damage on irrigation schemes. The project will fund a study of the current legislation covering disaster compensation which will propose clearer guidelines for compensating damages to irrigation works during rehabilitation. 3.18 Rehabilitation of New Irrigation Schemes (US$8.2 million): Provision is included for the rehabilitation of a further 4,000 ha of additional schemes. Preliminary identification of the schemes has been made in the inventory study prepared under the first project but final selection 19 will depend upon (a) the level of firm commitment identified among the water users; (b) the economic viability of the schemes (currently 14 percent minimum) and (c) the environmental assessment of the schemes. Implementation of this component will be carried out in phases, each phase designed to reaffirm continued commitment by the beneficiaries. This will require a more gradual process of rehabilitation than under the first project but will help to guarantee longer term sustainability. The provision for new rehabilitation of 4,000 ha is only indicative and, if the earlier phases of the project are successful, it is expected that this component could be considerably expanded, using funding from alternative sources, such as the CFD or the EDF, who should therefore be closely associated during supervision of the project. 3.19 The first phase will include in-depth orientation studies (6tude approfondie d'orientation - EAO), during which water users, with support from technical staff, will define an irrigation master plan (sch6ma directeur d'am6nagement - SDA), providing details of such works as can be undertaken immediately by the users. The SDA will also define those works which will be undertaken during subsequent stages, subcontracted to outside enterprises. During this phase, the willingness of the water users to start immediately on simple rehabilitation and the quality of the work carried out will provide a good indication of the commitment to longer term O&M. There would be regular contact between the field team and the local population on the irrigation schemes to define the responsibilities of all parties once rehabilitation begins. Support and assistance in the creation of Water Users' associations would be provided. Up to six months should be allowed to establish the level of commitment of the water users. 3.20 During the second phase, which would proceed once a sufficient level of commitment has been confirmed by the local field team, the agricultural development plan and the individual phases of rehabilitation would be determined. On the basis of the agricultural development plan accepted by the water users, the detailed rehabilitation proposal (avant projet d6taill6 - APD) and tender documents would be prepared for those segments of work which cannot be carried out by the farmers themselves. 3.21 The third phase, (the works phase), would follow, based on the in-depth orientation study (EAO), which should have identified those types of works which could be carried out under the supervision of the DGR, with direct input from the beneficiaries and those more complex works, for which tender documents and works supervision would be handled by consulting engineers. In each case, the work would be planned and executed in such a way that local commitment and contributions from the beneficiary community can be assessed, as the works progress and interrupted in the event of flagging commitment. 3.22 Support to the MEADR (US$5.2 million): The project would provide support to the Directorates of Rural Works, Agriculture, Programming and Finance, Lands (Domaines) and Water and Forests to enable the MEADR to provide more professional and effective technical support services for irrigation scheme operation and maintenance and transfer of agricultural technology. Emphasis would be (i) on ensuring the organization of farmer associations to take over responsibility for irrigation O&M; (ii) on strengthening the transfer of both agronomic and irrigation management technology; (iii) on reinforcing local coordination between the different local services; and (iv) on strengthening the planning, monitoring and evaluation of public resource mobilization and use. The project will support these services in nine zones, as part of support for a more generalized decentralization of the Ministry of Agriculture and Rural Development's services envisaged under the new constitution. The zones have been selected in five out of six of Madagascar's former faritany (provinces), on the basis of the following criteria: density of irrigation schemes already rehabilitated; consolidation of existing professional 20 agricultural structures already under way; the presence of the pilot national agricultural extension program; the existence of satisfactory local infrastructure and the existence of large-scale irrigated schemes either being rehabilitated or consolidated. The zones which have been identified to benefit from project support under this component are the following: Miarinarivo, Fianarantsoa, Mahajanga, Toamasina, Toliara, Antsirabe, Ambositra, Manakara, Ambatondrazaka/ Aparafaravolo. By providing resources in support of the move to more decentralized activities, the project would emphasize improved coordination among ongoing programs and projects in each of the regions where it is active. 3.23 The national directorate of agriculture would reorganize and strengthen the agricultural extension services in each of the nine zones covered by the project, over a three year period. Using a system of agricultural extension adapted to local needs, such as that adopted under the pilot national agricultural extension project, the DA would introduce systematic in-service training of extension staff in the new zones and would ensure improved links between extension and research through adaptive research and on-farm trials, with the active participation of farmers. Extension services would be offered to farmers both on and off the rehabilitated irrigation schemes. Through the initial orientation workshops and follow-up monitoring workshops, all local staff working for projects, with a vocation for improving agricultural extension and research should be associated, to ensure coordination and limit duplication, irrespective of whether projects are supported by governmental or non-governmental resources. Coordination with the National Agricultural Research project (Cr. 2042-MAG) would be important to ensure a steady flow of technical recommendations to the extension staff. Funding for such coordination could be provided out of the present project or by the National Agricultural Research project, as defined in an annual agreement to be drawn up between those responsible for the two projects each year. Special support would be provided to ensure training for the creation of women's groups and associations on the rehabilitated schemes. To this end, the annual work program will define resources to be mobilized in support of the activities of public or non-governmental institutions specialized in the promotion of women's activities. 3.24 The geographical limits of these proposed zones during the transition phase, while national decentralization decisions are being confirmed, will be based on the limits of existing fivondronana (departments). In each case the proposed area corresponds to the area currently covered by existing CIRVA or CIRGR. For each of the local directorates, the project would fund investment and operational expenses including equipment and vehicles, the rehabilitation of local offices and accommodation, field allowances and training of staff. The levels of field allowances paid under the project should be consistent with a single policy to be applied for all projects under the control of the Ministry of Agriculture. Assurances were received during negotiations that the Ministry would fix a common policy for the application of field allowances for all projects under its authority before the end of 1994 (para 8.2 - v). The project will also support actions to increase the autonomy of decision-making at the decentralized level and the delegation of financial responsibility for certain current activities. 3.25 Resources will also be made available to ensure coordination of project activities, with emphasis placed on giving direct responsibility for implementation to the Ministry's Directorates concerned. The national project coordination unit would continue to represent the main contact for external donors to the project and would continue to act as Secretary to the project coordinating committee, to which outside donors to the irrigation sub-sector would be invited for key programming and budgeting meetings. The coordination unit would centralize the results of project monitoring and evaluation activities, carried out by the respective Directorates responsible for implementation and would commission the annual auditing of project accounts. 21 3.26 Training (US$0.7 million): Training of field staff, local technical and administrative staff and national level technical and coordination staff will continue to be a major component of the second phase project. Special emphasis will be placed on regular and continuous in- country training programs for field staff, undertaken by local facilitators. Training in financial management, including computerization, and monitoring and evaluation will also be supported for local, regional and central level staff. Overseas training will only be considered on a case-by- case basis, in the framework of a detailed training program to be included in the annual work program submitted for agreement of all cofinanciers. The onus will be on project management to demonstrate why such training cannot be undertaken locally and to indicate the direct benefits in terms of career advancement for the candidates retained. 3.27 Monitoring and Evaluation/Audit (US$0.2 million): The project would provide for continuous monitoring and evaluation of project impact including increases in production and yields, adoption rates of extension messages and performance by water users' associations both in collection of dues for O&M and in implementation of operation and maintenance of the rehabilitated schemes (ref para 4.20). 3.28 Operations and Maintenance by Water Users Associations (US$1.3 million): Once rehabilitated schemes have been formally transferred to WUAs, maintenance and operations will be carried out by the Associations. Funding under this component will come solely from the WUAs. Collection of contributions from members of Water Users' Associations will be a critical parameter to be monitored in project execution. To the extent that such funds would represent the seed money for the future long term sustainability of the rehabilitated irrigation schemes, monitoring of the mobilization of funds by WUAs has been included as a specific project component. Contributions to the WUAs by members will be monitored closely during project implementation. E. Project Costs and Financing Costs 3.29 The project would be implemented over a five year period from the date of project effectiveness. Total project costs, including price and physical contingencies, are estimated at FMG 89.5 million (US$25.7 million) of which FMG 49.7 million (56 percent) would be in foreign exchange and FMG 10.1 million (11 percent) taxes. Base costs are estimated at US$21.2 million equivalent and contingencies amount to US$4.5 million equivalent (17.5 percent of total costs). Project costs by category of expenditure are summarized in Table 3.1 below and detailed cost tables are provided in Annex 1. Per hectare costs of rehabilitation, studies and works supervision are based on actual average costs of the most recent contracts awarded under the first project. 22 Table 3.1: Irrigation Rehabilitation Project Estimated Project Costs a/ (US$ million) Local Foreign Total A. Cyclone rehabilitation and corrective 1.4 2.7 4.1 rehabilitation works B. Environmental assessment and protection 0.6 0.9 1.5 C. Rehabilitation of additional schemes 2.8 5.4 8.2 D. Support for services of Ministry of Agriculture 2.5 2.7 5.2 E. Training 0.4 0.3 0.7 F. Monitoring and Evaluation 0.1 0.1 0.2 G. Operations and maintenance by WUAs 1.3 0.0 1.3 Base Cost 9.1 12.1 21.2 Physical Contingencies 0.9 1.2 2.1 Price Contingencies 1.4 1.0 2.4 Total Project Costs 11.4 14.3 25.7 a/ Includes US$2.7 million of taxes and duties 3.30 Contingency Allowances. Assumptions on physical and price contingencies used in the calculation of project costs are presented in Table 3.2 below: Table 3.2: Physical and Price Contingency Assumptions PY I PY 2 PY 3 PY 4 PY 5 Physical contingencies for rehabilitation, 10 10 10 10 10 equipment, vehicles and materials (%) Local inflation rates (%) 10 9 8 8 8 Foreign inflation rates (%) 2.0 2.5 2.7 2.5 2.6 (MUV Index May 1994) Exchange rates for FMG/US$ at mid point 3000 3300 3630 3630 3630 of project years 3.31 Incremental Recurrent Costs. Incremental recurrent costs average US$517,000 per annum (see Tables 7, 9, and 10 of Annex 1), including contingencies and cover the supervision costs of the Rural Works Directorate, the costs of intensified agricultural extension activities of the Department of Agriculture and the costs of project coordination. This represents the equivalent of US$13 per farm family working on the irrigated schemes. Approximately 50 percent of this cost is attributable to intensified extension efforts. Taking account of the benefits 23 of improved extension which will also accrue to those farmers who live in the areas surrounding the irrigation schemes, the cost per family would be roughly equivalent to the per family costs of agricultural services provided under the pilot agricultural extension service (approximately US$5 per farm family). 3.32 Project Sustainability. Project execution places a heavy emphasis on the establishment of commitment by local water users' associations to take over responsibility for costs of operation and maintenance of the rehabilitated irrigation schemes. To the extent that this transfer of responsibility is effectively carried out, the conditions for longer term sustainability of the rehabilitation investments will have been established. The project will introduce, through its participative approach to encourage farmers' involvement in local decision-making, awareness of the need for joint responsibility for maintenance and upkeep. Through more professional and regular farm extension services, which will require a continuing financial commitment from the Government of approximately $6 per farm family per annum (see para 3.31), the project will also contribute to reestablish a degree of confidence between the administration and the farming community, which is currently lacking and which should, if successful, ensure longer term sustainability of project efforts. F. Financing 3.33 IDA Credit. The IDA Credit of US$21 million would finance an estimated 83 percent of total project costs. IDA would disburse 100 percent of the foreign exchange costs or 85% of the local costs, net of taxes, of the incremental funds involved in the transfer of operations and maintenance of schemes rehabilitated under the first project, of the environmental protection measures, of the new rehabilitation works, of support services for the MEADR and of training, monitoring and evaluation. The Government of Madagascar would pass on all funds to the MEADR. PPF funds would be mobilized to meet the costs of the repair of urgent cyclone damage, participation workshops and start-up of the environmental diagnostic study. 3.34 Co-Financing. Project cofinancing arrangements had not been finalized as of the date of appraisal, due to the suspension of disbursements by the CFD and FED. The project has been designed in such a way that, once the intentions of cofinanciers in support of activities covered by the project, have become firm, they can then be included in the overall program of work under the national irrigation coordination unit. Should there be real progress in achieving successful transfer of rehabilitated schemes to beneficiaries during the early years of the project, then the area for future rehabilitation (presently fixed tentatively at 4,000 ha) could be increased significantly, for which cofinancing from alternative sources would be required. 3.35 Government's Share. Government's participation in the project costs is estimated at US$2.8 million and would cover 24 percent of the local currency costs of all project activities. This figure represents almost 11 percent of total project costs, including taxes and duties. The Government contribution would be equivalent to approximately 5 percent of funding from domestic sources in the 1993 agricultural public investment program. The figure is considered high enough to ensure Government commitment to the project but is also sensitive to the extreme scarcity of budget resources, which is not expected to improve during the project period. If successful, the project should lead to reduced public expenditure in the irrigation sector over time. Close supervision of the timely release of Government's share of project funding will be required and performance will be monitored at the time of the annual budget review by cofinanciers. Agreement was reached at negotiations that the Government shall carry out, 24 no later than October 31 each year, in consultation with all concerned donors, (a) an annual review of the Public Expenditure Program for the agricultural sector (8.2 - viii); and (b) prepare a proposed work program, budget and financing plan, including training proposals for the project for the forthcoming year and details of procurement procedures to be followed (para 8.2 - i). 3.36 Beneficiaries' Share. Contributions from members of Water Users' Associations towards the operation and maintenance of the rehabilitated schemes is estimated at FMG 50,000 (US$17) per hectare per annum. Such contributions are expected to average approximately US$340,000 per annum during the project period for a total of 20,000 ha rehabilitated under this and the first project. Total contributions from WUAs are expected to represent US$1.7 million over the project period. Table 3.3: Proposed Project Financing US$ million Percent IDA 21.2 82.6 Beneficiaries 1.7 6.7 Government 2.8 10.7 Total 25.7 100.0 25 IV. PROJECT IMPLEMENTATION A. Status of Project Preparation 4.1 The proposed project is a follow-up to the ongoing first irrigation project, cofinanced by the Caisse frangaise pour le d6veloppement (CFD), the European Development Fund (EDF) and IDA (Cr. 1589-MAG). An initial project preparation report was prepared by the FAO/CP after a mission to Madagascar in October 1992. Following the production of a major internal evaluation of the first project activities between 1985 to 1992 2/ in March 1993, the shape of the follow-up project was considerably modified by the appraisal mission which visited Madagascar in March/April 1993. In order to ensure that the conclusions of the appraisal mission were in line with the objectives and intentions of the Malagasy Ministry of Agriculture and the Directorates who were to be responsible for project implementation, a post-appraisal mission to Madagascar was carried out in August 1993, followed by a visit in November 1993 from a delegation from the MEADR to review the draft appraisal report, before circulation within the Bank. In this way, closer participation of Malagasy authorities in the definition of the final project description has been assured. B. Project Implementation 4.2 Project Implementation Timetable. The project will be implemented over five years starting in mid-1994, with disbursements extending over seven years. The environmental impact study and the diagnostic study of the schemes rehabilitated under the first project should be carried out at the beginning of the project period. These studies could be launched before project effectiveness with support from a Project Preparation Facility (PPF). Activities during the first two years of the project will concentrate on confirming commitment of water users associations to operations and maintenance of the newly rehabilitated schemes. Rehabilitation of additional schemes will start only when clear commitment has been established and be carried out over the last three years of the project. 4.3 The project will be implemented by the Ministry of Agriculture and Rural Development. Overall responsibility for project activities would reside with the Secretary General of the MEADR, as senior civil servant in that Ministry. The Secretary General would chair a project coordinating committee, comprising the Directors of the main Directorates responsible for project implementation, namely Rural Works (G6nie Rural), Agriculture, Programming and Finances, Domaines and Water and Forests. The secretariat to this committee would be assured by the national coordinator for irrigation rehabilitation, a Coordination Unit established under the first project, whose functions would be marginally changed to provide greater delegation to the central and regional staff of the MEADR Directorates to carry out their respective responsibilities for project implementation. These would be as follows: 4.4 Directorate of Rural Works (DGR) would be responsible for engineering design of civil works for irrigation schemes, selected for rehabilitation. During the first project, the DGR has 2/ Rapport d'Evaluation 1985 1 1992 by the Cellule Suivi-Evaluation, Coordination Nationale du Projet de R6habilitation des PPI, Mars 1993 26 acquired considerable experience in the process of selection of consulting engineers, in the definition of supervisory responsibilities and in the definition of appropriate timetables to carry out rehabilitation work. Full use of this experience and of the recommendations included in the evaluation report 3/ prepared by the PPI coordination unit should be made during implementation of the second phase of the project. 4.5 The DGR will be responsible for ensuring the execution of the irrigation rehabilitation funded under the project, including the environment protection measures. Once the rehabilitation works have been carried out, the DGR will ensure that a formal transfer agreement defining the details of the transfer of the rehabilitated schemes to water users. In the transfer agreement, enunciating the O&M activities for each of the schemes, the DGR will define the respective responsibilities between the water users (WUAs) and the state. In some cases, where there are relatively simple headworks and water control installations, the continuing responsibilities of the state will be limited to technical advice and extension. In the case of larger schemes, with more complex water management systems, it may be necessary for the state (through the DGR) to maintain direct responsibility for the O&M of parts of the rehabilitated scheme. In such cases, details of the financial contributions from the state and from beneficiaries should be included in the text of the transfer agreement between the DGR and the WUAs. 4.6 The central staff of the DGR would (a) prepare the detailed tender documents for new scheme rehabilitation; (b) launch tenders for civil engineering and for surveillance contracts by consultant engineers; (c) provide technical support to staff in the decentralized offices; (d) supervise project implementation at the regional and local levels. Three services would be responsible in the central DGR for project activities and would be made responsible for the technical supervision of the rehabilitation studies: (i) a "network" service responsible for checking the technical specifications, such as the baseline parameters, water flows, technical designs, the stability of the works proposed; (ii) a second "environment" unit will check the viability of the schemes from an environmental perspective. This unit will be responsible for launching specific environment impact and mitigation studies for any proposed extensions; (iii) a third technical monitoring and evaluation unit would check on the respect of contractual obligations. Short term specialized consultants will be mobilized to help the DGR to ensure the effectiveness of these units. 4.7 At the decentralized level, the G6nie Rural bureau chief would be responsible for studies and works control and supervision on the ground and for water user participation during the rehabilitation phase. He would be supported by an engineer at the level of each Circonscription where the project is operational and on the ground by an irrigation schemes advisor (Conseiller Hydraulique). The decentralized services would be backed up by short term external technical assistance, which would provide support in the areas of scheme design and technical supervision of the rehabilitation works, as well as on environmental aspects. Irrigation scheme advisors (CH) would be selected for each new and existing scheme, and would participate as members of the field team (ET) to establish close relations with the WUAs on each scheme to be rehabilitated or which have been rehabilitated. The field team would spend several months on the irrigation schemes with the water users to appreciate in detail the needs and priorities of the latter. They would thus be closely associated in the design of the rehabilitation works and made aware early on of the implications for future 0 & M. The CH would contribute to the in-depth orientation 2/ Rapport d'Evaluation 1985-1992 - Coordination Nationale du Projet de R6habilitation des Petits P6rimtres Irrigu6s, Cellule Suivi-Evaluation, Mars 1993 27 studies (EAO), and evaluate the quality of work by local communities, as the first step to ensuring longer term commitment. As part of the field team (ET), the CH would assist in the calculation and definition of the costs of operations and maintenance of the schemes after rehabilitation and would participate in meetings with WUAs to ensure that members fully understand the implications of these costs and charges. 4.8 Depending upon the complexity of the works to be carried out (to be evaluated by the central DGR in conjunction with project supervision and advisory staff at the time of the preparation of the annual work program), the DGR would either use its own engineers or subcontract to local or international engineering consultants for design and technical supervision. The central DGR services would coordinate procedures for the selection of consultants to undertake in-depth orientation studies (EAO), followed by the preparation of the tender documents for major works. The DGR (central and local staff) would supervise the work of such consultants and of the construction companies commissioned after tender to carry out new rehabilitation works. The DGR local staff would supervise the work of WUAs or local enterprises selected to carry out light rehabilitation works. The local staff of the DGR would work with WUAs to define the respective share of responsibilities for operation and maintenance of newly rehabilitated schemes, between the water users and the public authorities, to reflect the new regulations to this effect. The Ministerial decree providing for these regulations was signed on March 16, 1994. 4.9 The project would ensure increased local responsibility of regional and sub-regional echelons of the DGR, by delegating direct responsibility to the local offices for routine operational and financial activities. Further delegation of responsibilities for technical control and supervision will be increased, as local management competence is identified and confirmed. 4.10 The DGR would take back the functions the Rural Institutions Support Services (AIR), which has been attached to the PPI Coordination unit since 1990. The main activities of the central AIR Service will continue to provide training and support to the local rural institutions support services, who in turn provide back-up to the Conseillers Animateurs (CA). The CA, as member of the field team (6quipe de terrain), would remain responsible for the creation, facilitation, monitoring and support of the WUAs. The staff of the AIR would continue to work closely with the Directorate of Agriculture's Regular Training Service (Division de Formation Continue) and would act as Subject Matter Specialists (Techniciens sp6cialis6s) for the regular training of agricultural extension staff on all questions concerning support to and development of farmers' associations. 4.11 Support for the DGR under the project would include vehicles and equipment, training, field allowances and operational costs of vehicles to permit effective supervision of EAOs and APDs by consulting engineers and rehabilitation works by civil works companies. Funds would also be available for specialized consultant services and short term technical assistance to supervise the work of the DGR rural engineers. DGR staff would benefit from the intensive training programs offered under the project. 4.12 Directorate for Agriculture (DA) would be responsible for the provision of agricultural extension services to farmers on the rehabilitated schemes. The main objective of support to the DA under the project would be the provision of a pluridisciplinary professional agricultural extension service to all farmers on or around the irrigated schemes. Agricultural extension workshops would be organized in those areas, where the project is currently operational to bring together all those services or projects which have a vocation for agricultural extension. Through 28 these workshops, measures to rationalize multiple extension services in each zone of the project would be defined. In the first years of the project, field services would be strengthened in those areas where rehabilitation has already been completed and the principles of an agricultural extension system adapted to local conditions along the lines similar to those being introduced in areas covered by the PNVA would be introduced. The provision of improved extension services will not be limited only to those farmers on the rehabilitated irrigation schemes. It would be available to farmers on and around the newly rehabilitated schemes to ensure access to improved technology by the whole farming population over time. In particular, the Directorate would organize continuous training of field staff, in conjunction with subject matter specialists (techniciens sp6cialis6s) from FOFIFA and other research/development institutions, such as crop- or livestock-specific project entities or organizations. As such, support to the DA under the project will represent a second phase of the ongoing pilot agricultural extension project (Cr. 2150-MAG). 4.13 The project will fund vehicles and equipment, training, field allowances and operational costs of vehicles necessary to permit the DA to provide professional field extension services on a regular basis to farm families in the project areas. Funding will also be provided to permit the local extension services to organize local research and development trials and to carry out seed multiplication in conjunction with local farmers and to encourage the establishment of input distribution outlets, especially in the vicinity of the irrigated schemes, where demand for new technology should develop rapidly. 4.14 Directorate for Programming and Finances/Methods and Processing Service (SMTIS). For project monitoring and evaluation activities, the staff of the Programming and Finance's Directorate SMTIS would collaborate with the project coordinators at the local level and staff of the Directorates of Rural Works and of Agriculture in providing the enumerators to carry out surveys on project performance. Collection of data at the field level would be undertaken by staff specifically engaged for the purpose (as opposed to extension staff under the first project). Data would be processed on microcomputers to be supplied to each of the zones and technical support for this exercise would be provided by short term consultants during the first three years of the project. Consolidation of the data from the monitoring and evaluation service would be the responsibility of the national coordination unit of the project. 4.15 Project Coordination. Central coordination of overall project activities will be ensured by a streamlined coordination unit, located in the Ministry of Agriculture. The coordinator will fill the following functions: (a) secretariat to the project Coordinating Committee, set up under the first project. Agreement to the maintenance of an irrigation sector coordinating committee and the appointment of a coordinator was obtained from the Ministry of Agriculture during negotiations (para 8.2 - vi). The Coordination Committee should meet no less than twice a year and would invite representatives of external donors to attend key work programming and budget sessions; (b) facilitator of exchanges between the different Directorates responsible for project implementation; (c) responsible for the coordination of the preparation of annual budgets by each of the directorates participating in the project and organizing review and agreement by cofinanciers; (d) local supervisor of the project on behalf of the Secretary General of the MEADR to ensure that the project resources managed by the technical Directorates are used for the purposes for which they were agreed. In this role, the coordinator will ensure the preparation of annual work plans and their submission to the project Coordinating Committee, as well as half yearly progress reports on project implementation; (e) responsible for the consolidation of project accounts and the submission of disbursement requests to IDA and other cofinanciers; (f) responsible for centralization of monitoring and evaluation of project activities; (g) contact with 29 the external donors to ensure compliance with terms and conditions of agreements in support of the project; (h) responsible for the preparation of the subsequent phases of the project, work on which could begin as early as the mid-term review of the project at the end of 1996 (see para. 4.22), if the project is making positive progress. During the first year of the project, the coordination unit will ensure the smooth transfer of the following functions: (i) responsibility for monitoring and evaluation of project implementation to units established for this purpose within the Directorates for Rural Works and Agriculture, supported by enumerators from the Programming and Finance's Directorate SMTIS (para 4.14); (ii) tendering and the selection of consulting engineers and supervisory services to the Directorate for Rural Works; (iii) reintegration of rural institutions support services (AIR) to the Directorate for Rural Works. During and after this transfer, the coordination unit will provide advisory services to the implementation Directorates for the tasks which the latter will henceforth assume. 4.16 At the decentralized level, coordination of project activities will be ensured by local project coordinators, specifically nominated to fill similar supportive functions to those of the national coordinator and the central Directorates. To the extent possible, these coordinators would be selected from available local staff within the region or otherwise reassigned from services elsewhere. A local Coordination Committee bringing together representatives of the Directorates of the MEADR responsible for project implementation, the Directorate for Water and Forests (responsible for environmental protection, local ONGs associated with the project or with environmental protection and the local representative of the Ministry of Public Works to ensure that the latter is associated given its responsibility for rural road maintenance, will meet regularly. 4.17 At the local level, a local coordination committee will bring together the representatives of the WUAs benefitting or to benefit from the project and the local field team, comprising the agricultural extension officer, the local DGR representative (Conseiller Hydraulique) and the Conseiller Animateur. 4.18 The coordination of the preparation of training programs for project personnel at central, regional and field levels will be ensured by the national coordination unit in close collaboration with the Directorate responsible for human resource management (DGRH). The training program would be presented each year together with the annual work program and should reflect training priorities as expressed by each of the units responsible for project implementation, including the DGRH. 4.19 Financial and Administrative Service (SAF). A service responsible for administration and finances would be created at the national and decentralized levels. It will be responsible initially for financial management of the project resources, notably in the establishment and internal auditing of accounts. This service would ensure the computerization of accounts for the project and in due course for the Ministry's regional operations. At the local level, the SAF would provide support and training to WUAs in the setting up of appropriate financial management and accounting systems. It would also work in conjunction with DGR personnel to calculate operating costs of the rehabilitated schemes and the estimated contributions by water users to ensure sustained O&M. Short term technical assistance to the SAF would be funded by the project to support these functions. 4.20 Beneficiaries. Since the success of the project depends to such an extent upon the willing participation of the farmers and water users in taking over O&M responsibility, the role of the field teams (ET) assigned to each irrigation scheme and who are in regular touch with the 30 beneficiaries is critical. A clear mechanism, such as a regular meeting between field teams, representatives of WUAs and local coordinators, should be established to permit feed-back from the project beneficiaries to those responsible for project implementation. Such modifications as might be required in project design and implementation should be reviewed and adopted in consultation with the central coordination unit and donors as the project proceeds, maintaining the principle of flexibility adopted under the first project. C. Project Monitoring and Evaluation 4.21 Project monitoring and evaluation would be centralized by the national project coordination office. The monitoring and evaluation services of the Directorates of Rural Works and of Agriculture will undertake monitoring of the activities of their respective directorates. The local staff of the Directorate of Programming and Finance's Methods and Processing Service (SMTIS) would provide enumerators for local surveys in the project area to monitor progress project implementation progress, including (a) the growth in number and membership of WUAs; (b) progress on the collection of membership dues and contributions from members to O&M charges; (c) rate of adoption of agricultural extension messages; (d) increases in agricultural production, cropping intensity and yields; (e) incremental household income; and (f) evaluation of the impact of training and workshops. Local monitoring and evaluation should be carried out in close collaboration with the three members of the field teams and in association with members of the Water Users' Associations, to encourage the adoption of improvements in project implementation as identified by beneficiaries. The project monitoring and evaluation service will develop indicators by which the impact of the project on (i) food security; (ii) poverty alleviation; and (iii) capacity building can be measured. Progress on such monitoring will be reviewed as part of the annual work program review. Reports including the results of these surveys would be incorporated in the annual report prepared by the project Coordinator for the project Coordinating Committee and for distribution to donors. 4.22 Thirty months after project effectiveness, a mid-term review of the project would be carried out to evaluate progress and make recommendations for project orientation and activities for the second half of the project period. During the mid-term review, a special effort will be made to associate project beneficiaries, particularly members of the WUAs to draw on their experience and contribution to defining the direction which the second half of the project should follow. Agreement to scheduling the mid-term review not later than October 31, 1996 was reached at negotiations (para 8.2 - vii). At project completion, the project coordination unit would (i) prepare a completion report, which would present an analysis of project implementation experience and (ii) make proposals for a new cycle of investments in irrigation rehabilitation. D. Procurement and Disbursements 4.23 All goods and services funded by the IDA Credit would be procured in accordance with Bank guidelines. Bank Standard Bidding and Contract documents will be used for all ICB and consultant contracts. For fixed price contracts, any extension of bid validity would be accompanied by an automatic adjustment of price. Domestic civil works contractors would be allowed a preferential margin of 7.5 percent over the lowest evaluated bid, to encourage the participation of local contractors. Under ICB procurement of goods, local manufacturers would be allowed a preference of 15 percent or the import duty, whichever is lower. 31 4.24 Civil Works. Contracts for rehabilitation of additional schemes and environmental protection works would be grouped to the extent feasible and would be subject to international competitive bidding. Civil works, such as the rehabilitation of local buildings and offices, estimated to cost the equivalent of US$100,000 or less per contract, up to an aggregate amount not to exceed US$1,000,000 may be procured under contracts awarded on the basis of local competitive bidding, in accordance with procedures satisfactory to IDA. 4.25 Local competitive bidding for civil works and for the recruitment of local consultants for the control and surveillance of civil works would be limited to a list of qualified companies, which have already demonstrated satisfactory experience for such work preferably in Madagascar. The process would be open to all firms, local or foreign, and inclusion of firms in the list of approved firms, based on the required standards of qualification would be on a continuous basis. Thus the list of qualified firms who would be invited to bid shall be continuously updated (on a semi-annual basis) to generate adequate competition. Updating of the list would be subject to agreement of IDA, during the course of project supervision. Experience with contract implementation has already provided the National Coordination Unit with a list of approximately six local companies, which are considered reliable. 4.26 Contracts for civil works for the urgent repair of cyclone damage, which will commence with funding under the project preparation facility, are not expected to exceed US$75,000 per scheme. Contracts for such works, for a total not exceeding US$750,000 will be awarded either on the basis of comparative bids from three contractors or on the basis of sole source contracts. These sole source contracts would be limited to those cases where contractors have already been selected on the basis of competitive bidding under the first project and are either still completing works on the schemes or are bound by the maintenance guarantee of the contracts. Such sole source contracts would retain the same price and contract conditions. 4.27 Goods. Contracts for the purchase of vehicles, motor bicycles and bicycles would be subject to international competitive bidding (ICB) and would be grouped so as to obtain maximum economies of scale. Furniture, equipment and supplies (US$0.46 million) would be purchased using local competitive bidding procedures for contracts of less than US$100,000. Contracts estimated to cost less than US$30,000 may be procured using local or international shopping up to an aggregate of US$200,000. To ensure that the project obtains competitive prices for procurement of these and other goods, the project coordination unit is encouraged to use, for purposes of comparison, the price lists provided by the UNDP's Inter-Agency Procurement Service Office (IAPSO). 4.28 Consulting Services. Contracts for the recruitment of consultants for detailed design of rehabilitation works and surveillance of works expected to exceed US$100,000 would be subject to international recruitment through a short list of firms following Bank guidelines. Contracts for the control and surveillance of these works would be undertaken by the same consultants who will have prepared the detailed studies for the rehabilitation works (APD). The terms and conditions for the contracts for such surveillance would be submitted by the consultants with their original offer of services and would figure among the criteria for selecting the company for the detailed design work and studies (APD). All other short term technical assistance, both local and external, would be awarded in accordance with the Bank's guidelines for the selection of consultants. 4.29 Prior Review Procedures and Reporting. Each year, the Annual Work program will include details of the procurement procedures to be adopted during the year in question and to 32 be agreed by IDA. Requirements for IDA prior review of procurement operations would be as follows: (i) Procurement of goods and construction contracts: proposals for advertising, draft tender documents, bid evaluation and award proposals for all contracts exceeding US$100,000 would be subject to review by IDA prior to their execution. (ii) Contracts with consulting firms and consultants: prior review and approval of terms of reference, budgets, short lists, selection procedures, letters of invitation, proposals, evaluation reports and contracts would be required for contracts for the employment of consulting firms exceeding US$100,000 each and for all contracts for the employment of individuals exceeding US$50,000 each, as well as all sole source contracts. (iii) Contracts below the prior review thresholds under (i) and (ii) above would be subject to an ex-post review by IDA on a selected basis. 4.30 In the light of its experience under the first project, the project coordination unit will provide support to the technical services for the preparation of bidding documents and the launching of tenders. The longer term objective of the project should be to devolve procurement responsibility to the technical directorates, which will be the main beneficiaries of procurement orders. To the extent feasible these directorates (and eventually representatives of the directorates at the local level) will be responsible for preparing detailed technical specifications and for launching tenders for these goods and services. Recruitment of consulting services will be coordinated by the national coordinator's office. All committees established for the purpose of bid evaluation should regroup representatives of the Directorates participating in the project, including the project coordination unit and such representatives of the administration as required by national legislation. 4.31 A summary of procedures to be followed for procurement under the project is included in Table 4.1 below. A schedule of the procurement of goods, works and consulting services will be included in the annual work program agreed with project cofinanciers (para 8.2 (i)). 33 Table 4.1: Procurement Procurement Method International Local (USS 000) Competitive Competitive Consulting Bidding Bidding Other Services N.B.F. Total A. Works 13,193 1,202 - - 14,395 (11,737) (1,060) (12,797) B. Goods 976 459 293 - - 1,727 (662) (333) (205) (1,200) c. consultants Capacity building /a - - - 1,028 - 1,028 (1,000) (1,000) Implementation support & supervision /a - - 1.466 - 1,466 (1,363) (1,363) Studiesla - 11 883 - 894 (11) (824) (835) D. Training and Workshops - - 1,007 38 - 1,045 (1,007) (38) (1,045) E. Allowances and recurrent costs Incremental recurrent costs 155 1,434 - - 1,588 (131) (1,120) (1,251) Incremental allowances - - 998 - - 998 (998) (998) F. Other/b - - - - 1,712 1,712 G. PPF - - 839 - - 839 (739) (739) Total 14,169 1,816 4,582 3,415 1,712 25,693 (12,399) (1,524) (4,079) (3,224) - (21,227) Note: Figures in parenthesis are the respective amounts financed by International Development Association \a Services should be procured in accordance with World Bank 'Guidelines: Use of Consultants by World Bank Borrowers (Washington DC A \b Contributions from WUAs towards O&M - Not Bank-financed E. Accounting, Auditing, Budgeting, Disbursement and Reporting Procedures 4.32 Accounts and Audits. Separate accounts for project expenditures will be maintained, as under the first project by the national coordinator's office, which will be required to adopt an accounting system in line with the new directives to be issued shortly by IDA. The directives imply the introduction of a double-entry accounting system and the preparation of annual accounts covering the full activities of the project. However, given the objectives of decentralization of financial budgeting, management and accounting of the MEADR's activities and of the transfer of irrigation scheme management to WUAs, a simple system of decentralized double-entry accounting will be introduced at the level of the local coordinators and for each irrigation scheme. There is also a need for a more simplified system of document filing than existed under the first project. For both these needs, the project would fund training and short term technical assistance. Approximately 12 agents would receive training, three from the national coordinator's office and nine accountants from the local offices. Details of the training and terms of reference 34 for the consultants to carry out such training are included in the Implementation Volume which accompanies this report. Project accounts will be audited annually by independent auditors acceptable to IDA, in conformity with the norms of International Standards on Auditing. Audits of financial accounts will be submitted to IDA not more than six months after the close of the financial year to which they refer. Agreement on the above was obtained during negotiations (para 8.2 - ii). 4.33 Budgets. Special emphasis will be placed on the preparation of annual budgets and their respect during the course of implementation. The National Coordinator will play a major role in ensuring the timely preparation of annual budgets by all levels of the institutions benefitting from the project. Assurances were received during project negotiations that a project execution program, budget and financial plan will be submitted annually before October 31, 1994, for the following year's activities, to be reviewed and approved by IDA and the other cofinanciers (para 8.2 - i). Semi-annual reports of project implementation progress would be submitted to IDA and other cofinanciers before March 31, and September 30 each year. (para 8.2 - iv) 4.34 Disbursement Procedures. Once agreement has been reached on the annual work program and budgets, withdrawals from the Credit will be claimed using the standard procedures, including reimbursement of expenses prefinanced by the Ministry of Agriculture against presentation of a withdrawal request, direct payment to suppliers or replenishment of the special account. 4.35 To facilitate disbursement locally, IDA will advance the sum of US$300,000 into a special account in convertible currency to be opened with a commercial bank acceptable to IDA. Upon effectiveness 50 percent of the initial deposit amounting to US$150,000 will be deposited to the special account. The remaining balance will be paid as needed. Reimbursements should be grouped for replenishment of this special account and submitted for reimbursement every month or whenever the special account balance is reduced to one third of the original amount, whichever comes first, through the standard channel of the Ministry of Finance. The account may be used to pay for expenditures in either local currency or foreign exchange through the commercial banking system, against any category of expenditure. All replenishments should be fully documented except in the case where statements of expenditure (SOEs) are authorized. 4.36 A project advance account in local currency, operated by the office of the project coordinator, would be established in a commercial bank, with an initial deposit of the equivalent of US$100,000 to cover Government's contribution to project expenditures. This advance account would be replenished from a deposit account opened either at the Treasury or at the Central Bank, which would be created to receive the totality of the Government's counterpart contribution each year. The advance account would be replenished by the government, against presentation of supporting documentation, every two months or when the balance falls to one third of the original amount, whichever comes first. Agreement on these procedures was reached during negotiations. The deposit of an initial amount equivalent to US$100,000 shall be a condition of effectiveness of the credit (para 8.4). 35 4.37 Disbursement by IDA against Statements of Expenditure (SOEs). The project coordination unit would claim reimbursement on the basis of statements of expenditure (SOEs) for: (i) civil works costing less than US$50,000 equivalent (Disbursement Category I); (ii) goods, training, consultants contracts, studies and operating costs for single items costing less than US$20,000 equivalent (Disbursement Categories II, III, IV, V and VI). 4.38 The Coordinator's office will retain all the relevant supporting documentation for reimbursement claims under SOEs for inspection by supervision missions. During the course of the annual audit of the project's accounts, the auditor will certify that all expenditures claimed under SOEs during the year have been properly incurred for the project. 4.39 Credit Disbursement. The disbursement categories for the IDA credit are presented in Table 4.2 below. IDA will finance 100 percent of all categories of expenditures, excluding identifiable taxes and duties, up to the amount agreed in the annual work program. The proportion of the annual work program to be financed by IDA will be decided each year after review and discussion of the proposed budget by the national Coordinating Committee (para. 4.15). Agreed levels of funding will be closely linked to the commitment of the government and of other donors to meet their share of project costs, as identified in Table 3.2. The credit closing date is scheduled for June 30, 2000. Table 4.2: Disbursement of IDA Credit Amount Percentage Category (US$ '000) Financed by IDA 1. Civil Works 11,905 100 2. Consultants' Services 2,183 100 3. Vehicles and Equipment 1,064 100 of foreign and 85 of local costs 4. Studies 400 100 5. Travel Costs and Training 907 100 6. Operating Costs 2,088 100 7. PPF advance 750 100 8. Unallocated 1,930 Total 21,227 4.40 Based on the standard disbursement profile for agricultural projects in Madagascar, the IDA Credit would be disbursed over a period of six years. The expected disbursement profile is given in the table below. During the early years of the project the disbursement rate is expected to follow the standard disbursement schedule. Towards the later years of the project, assuming successful implementation, disbursements are expected to accelerate as new rehabilitation work is started. If successful, the project is expected to lead to a third phase covering further irrigated schemes. 36 Table 4.3: Estimated IDA Disbursements (US$ millions) FY95 FY96 FY97 FY98 FY99 FY00 FY01 6 ms 6 ms Annual 0.5 1.5 2.5 4.0 4.8 5.2 2.6 Cumulative 0.5 2.0 4.5 8.5 13.3 18.5 20.1 Percentage (%) 2.0 9.0 21.0 42.0 66.0 92.0 100.0 Agriculture Sector Disbursement Profile (%) 0.0 10.0 22.0 38.0 58.0 82.0 94.0 4.41 Reports on project implementation progress would be submitted to IDA and other cofinanciers semi-annually at the end of March and at the end of September each year. The September report would be submitted so as to coincide with the meeting of the Coordinating Committee called to discuss the following year's budget (para 4.15). Agreement on this would be reached at negotiations. F. Project Impact on Government Budget 4.42 During Project Implementation. The proposed contribution from Government to the project represents approximately the equivalent of US$ 550,000 per annum or 9 percent of the government's contribution to the 1993 agricultural public investment program. The proposed level of contribution from the Government takes account of the chronic shortage of local resources to meet counterpart contributions during the past years. During the annual budget approval process, the proportion of funding from local sources will increase, in line with realistic projections on the availability of local funds, based on experience with the preceding budgetary exercise. At the end of the project period, Government will assume the recurrent costs of the extension services and the services provided by the Rural Works Directorate, strengthened under the project, which represent approximately US$800,000 equivalent per annum. Of this amount about two thirds would be taken up by the National Agricultural Extension program, currently being funded on a pilot basis, but due to attract further funding as a national program by 1996. There would be savings to the Government budget of approximately US$500,000 equivalent per annum due to the assumption by water users of operation and maintenance charges currently the Government responsibility. 4.43 The close dialogue which the Bank will maintain with the Ministry of Agriculture on the public expenditure program will help to ensure that sufficient local funding is available to meet the Government's contribution. G. Legal and Regulatory Framework 4.44 Land Tenure. Most land is cultivated by the defacto owner/occupant on the majority of the schemes to be rehabilitated under the project. These ownership rights have evolved according to Madagascar's traditional land tenure rights. In some cases, land tenure contracts between the Government and the users were established at the time an irrigation scheme was 37 established but have since become outdated. Some owners have title deeds dating back to a land registry established prior to independence. According to a specific study undertaken on the subject in 1991 4/, and in spite of the different land tenure situations, there is little evidence of insecurity of tenure among occupants of irrigation holdings. The law does not provide for share- cropping or tenancy, although these are common practices in Madagascar. In spite of this lack of legislation, there does not appear to exist any major reluctance on the part of tenants or share- croppers to undertake modest investments on their holdings or any other specific constraints posed by prevailing land tenure which are likely to affect project implementation. 4.45 As under the first project, in selecting schemes for rehabilitation, land tenure arrangements will be reviewed and land tenure issues specific to the scheme will be assessed to determine whether or not they are likely to inhibit successful transfer of responsibility for operations and maintenance to users. In addition, the Directorate for Lands (Direction des Domaines) will continue to collaborate with the local coordination units to survey individual plots on irrigation schemes which have been rehabilitated. Once these surveys have been undertaken, individual farmers can obtain full title deeds through the local land registries. To the extent that the national environment program (Cr. 2125-MAG) is providing a major support to the land registry services, the two projects should coordinate their activities closely. 4.46 Water Users' Associations. The methods of association and limits of responsibility were defined for WUAs in legislation promulgated in 1985 and modified in 1990. The detailed regulations providing clarification on the role and responsibilities of the state and of the Water Users' Association, concerning in particular investment and maintenance charges, the level of financial autonomy of the associations, the role of local collectors, and the degree of autonomy of the WUAs, have been prepared and discussed during a special national seminar with the Presidents of existing WUAs. The signature of the Ministerial decree, incorporating these regulations together with the special conditions of the contracts to be signed between Government and the WUAs, in terms which are satisfactory to IDA, was a condition of Board presentation As a further condition, the Ministry of Agriculture has issued a specific decree providing formal recognition of the Water Users' Associations as separate legal entities. 4/ P-C. Bloch and C-W. Dickermann "Le r6gime foncier du secteur irrigu6 i Madagascar", Final report, November 1991 38 V. AGRICULTURAL DEVELOPMENT AND PRODUCTION IMPACT 5.1 Technological Production Improvements. Several technological packages are currently available for dissemination by the agricultural extension staff. For rice production, which constitutes the main crop on the irrigated schemes, the basic recommended technology (FN1 - use of organic and chemical fertilizers in the nursery, replanting in lines with application of an insecticide application and sufficient irrigation water) can be diffused by extension staff to the majority of farmers on the rehabilitated schemes without major risks. The more advanced rice production recommendations (FN2) represent a higher risk and should probably be limited to larger growers only. (FN2 involves selected and improved seeds and the use of chemical fertilizers after replanting). For crop intensification in the schemes, several technological recommendations exist to permit a double cropping of rice or other crops adapted to irrigated conditions in the off-season. The use of organic manure on the rice seedbeds and paddy fields is a technique which is relatively widely used on the central plateau, by those who have access to manure and which could usefully be propagated elsewhere. Improved varieties (such as 2787 in the south) can provide appreciable increases in yield under less intensive conditions. Other practices which have given positive results include protection against rice borer (pou du riz), clear seeding in the nursery, sowing in line in the paddy fields and respect of planting dates and could be encouraged by extension, taking account of local sociological context and existing agricultural systems already in place. Intensification of fish farming in conjunction with rice production also exists, which would increase farmer incomes and would encourage the adoption of more ambitious production practices. Fish raising under irrigated conditions has the added advantage of encouraging maintenance of the perimeters, to avoid escape of the young fish stock. 5.2 Table 5.1 presents a summary of potential technical themes which could provide the basis for more extensive agricultural extension on and around the rehabilitated irrigation schemes: Table 5.1: Potential Technical Improvements in Agricultural Production Practice\Area High Plateaux North and West South East Coast Rice production Organic manure, Treatment against Improved Improved azolla, chemical rice borer, varieties, azolla, varieties, fertilizers 5/ agricultural improved improved seedbed, implements, azolla, seedbeds, azolla, chemical improved seedbeds, chemical fertilizers 5/ chemical fertilizers 5/ fertilizers 5/ Off-season crops Wheat, potatoes, Vegetables, fish Vegetables, fish Vegetables, fish (other than rice) beans, pasture, vegetables, fish 5.3 To simulate the expected agricultural production increases in the project area, crop models for paddy, potatoes, triticale (wheat equivalent) and fish raising have been prepared for the different project areas. For paddy, without project yields have been assumed to be relatively low, under rainfed or poorly irrigated conditions. Fifty to sixty percent increase in yields of paddy are feasible given current levels of production on many schemes which are limited by the lack of water during the off season. For production models for potatoes and triticale, I/Only for larger rice producers 39 recommendations adopted by FIFAMANOR and KOBAMA projects have been adopted, which can provide yields of 16 tons/ha and 2.5 tons/ha respectively. Fish farmers can expect yields of 300 kg/ha according to techniques recommended by a UNDP/FAO project in Madagascar. 5.4 The increased yields and production resulting from adoption of improved techniques under the project are included in Table 5.2 below. Table 5.2: Increases in Production and Yields with and without Project Project Area/Crops Yield t/ha Cropping intensity % Without With Increase % Without With Project Project!' Project Project!' Antananarivo 95.0 140.0 . Rice (paddy-Vakianb) 1.6 2.6 63.0 - - . Triticale (off season) 0.0 2.5 - - - . Potatoes (off season) 0.0 16.0 - - - Fianarantsoa 91.0 140.0 . Rice (paddy-Vakiamb) 1.6 2.6 63.0 - - . Rice/fish farming = Paddy 1.6 2.6 63.0 - - Fish 0.04 0.3 525.0 * Triticale 2.5 2.5 - * Potatoes 16.0 16.0 - - - Mahajanga 85.0 130.0 Rice (Paddy Jeby) 1.4 2.2 57.0 - - Rice (Paddy double cropping) 1.4 2.2 57.0 - Toamasina 83.0 125.0 Rice (Paddy main season) 1.5 2.3 53.0 - - Rice (Paddy double cropping) 1.5 2.3 63.0 - - Toliara 90.0 125.0 Rice (Paddy main season) 1.6 2.4 50.0 - - Rice (Paddy double cropping) 1.6 2.4 50.0 - - ! by Year 5 of the project 5.5 Financial Impact of the Adoption of New Techniques. To estimate the financial viability of the agricultural development proposals recommended for adoption on the rehabilitated irrigation schemes, five farm models have been developed which are representative of the five Faritany, where the project will intervene. These models illustrate, as Table 5.3 shows, how the irrigated paddy fields are used in the different provinces. Since the project is not expected to have a major direct impact upon rainfed agriculture or on livestock, these activities have not been taken into account in the presentation of the improvements to be introduced through the project. Each model shows the evolution of the type of use of a hectare of irrigated land before and after the rehabilitation works. Thanks to better water management and control, the rice fields can be used fully during the rainy season and off-season cropping could occupy between 125-140 percent 40 of the total available area, compared with actual cultural practices which only use 83-96 percent of the area available. 5.6 The significant increase in the levels of production can be attributed to a significant increase in the crops produced during the rainy season and during the off-season and from an increase in yields. Once an improvement in water management is achieved by rehabilitation, important increase in production can be expected from double-cropping both rice and off-season crops. The above table shows the clear financial advantage of adopting the recommendations to be generalized under the project. The increases vary between the US$195 and US$420 per hectare according to the above models. The farm models demonstrate also that increases in farm revenues will require a considerable increase in the farmer's labor requirements. However, the small size of individual holdings and the lack of alternative sources of remunerated work suggest this should not represent a constraint, given that improved farm production is the only way in which farmers can improve their living conditions. This conclusion is confirmed by the increases in crop intensity that has been observed on certain schemes, once the rehabilitation work has been finished. Table 5.3: Cropping Intensity and Net Revenue Per Model Farm (ha) Antananarivo Fianarantsoa Mahajanga Toamasina Toliara Without With Without With Without With Without With Without With Project Project Project Project Project Project Project Project Project Project Fallow 0.10 0 0.10 0 0.15 0 0.2 0 0.1 0 Cropping intensity/ ha Rice 0.85 0.85 0.8 0.8 0.85 1.0 0.8 1.0 0.9 1.0 (main season) Rice (off- 0 0 0 0 0 0.3 0.03 0.25 0 0.25 season) Potatoes 0.03 0.2 0.05 0.2 0 0 0 0 0 0 Triticale 0.03 0.2 0.05 0.2 0 0 0 0 0 0 Fish 0.05 0.15 0.01 0.2 0 0 0 0 0 0 breeding Cropping 0.96 1.40 0.91 1.4 0.85 1.30 0.83 1.25 0.9 1.25 intensity Net 0.469 1.211 0.482 1.238 0.332 0.709 0.349 0.719 0.404 0.756 revenue/ ha FMG millions 41 5.7 Aggregation of the production on approximately 20,000 ha, which will benefit from more intensive agricultural extension and improved water management shows a total increase in output of agricultural production as follows: Without Project With Project Increase (tons) Paddy 38,000 52,800 14,800 Potatoes 1,280 7,960 6,720 Triticale 225 1,240 1,025 Fish 4 120 116 VI. MARKETS AND PRICES 6.1 The marginal increase in production generated by the project should have no difficulty in finding a market. Madagascar depends to a large extent still on imports of rice, wheat and triticale. Between 1985 and 1992, paddy production reached an average of 2.4 million tons, which covered only 90 percent of domestic consumption. Increased rice production generated by the project would enable Madagascar to reduce its imports of this crop by 10 percent. The production of wheat, barley and triticale under the production would have only a nominal impact of imports (less than 1 percent). The additional production of potatoes and fish would readily find a local market in the areas of production. 6.2 Marketing of agricultural production is assured by private sector traders with the exception of a number of specialized crops, such as wheat and cotton, which still remain in the hands of state monopolies. Marketing does not normally pose a problem to the extent that an irrigated scheme has relatively easy access to a road outlet. Since one of the criteria for the selection of irrigation schemes for rehabilitation will continue to be the proximity of a local road (para 3.18), it is not anticipated that the project should encounter major marketing problems. 42 VII. BENEFITS AND RISKS A. Financial and Economic Benefits Institutional Development Benefits 7.1 The project will make a major contribution to the reinforcement of the technical services of the Ministry of Agriculture, both at the central level and more especially at the provincial and local level. The services of the Rural Infrastructure Directorate and the Directorate for Agriculture are principal beneficiaries. This institutional strengthening will benefit agricultural development in Madagascar in general and the irrigation sub-sector in particular. Direct Quantifiable Benefits 7.2 Through the improvement of water management on existing irrigation schemes, the project will have a direct impact on increasing the income of approximately 40,000 small scale farm families, assuming an average holding of 0.5 ha per family. By introducing the possibility of cropping during the off-season, the area cultivated may be increased by up to 50 percent. The introduction of better organized agricultural extension services and the application of improved techniques for rice and non-rice crops will lead to increased yields and hence output. An estimated incremental annual production of 15,000 tons of paddy, 6,700 tons of potatoes, 1,000 tons of triticale and 116 tons of fish are expected to be achieved by the project within the next five years. Indirect Benefits 7.3 With the transfer of the responsibility for operations and maintenance of the medium-sized irrigation schemes, the recurrent costs of maintenance of these schemes from public funding sources will be reduced. Once rehabilitation has been completed and the schemes have been taken over by the water users, it is estimated that there will be a saving of the equivalent of US$500,000 per year to the public exchequer. Economic Rate of Return (ERR) 7.4 The economic analysis of the project has taken into account the costs of complementary work funded under the project directly related to the rehabilitation of the irrigation schemes, started under the first project, as well as the costs of the new rehabilitation work, currently limited to 4,000 ha. In addition, the costs of the environment component have been taken into account. Only fifty percent of the costs of strengthening the local services of the Ministry of Agriculture have been included in the calculation of the economic rate of return, given that the benefits would not be limited to the irrigation sub-sector. Benefits are based upon the incremental production of paddy, potatoes, triticale and fish, on the basis of the farm models for each of the project areas, on the full 20,000 hectares which would benefit from rehabilitation investments on 4,000 ha included under the project and the improved provision of extension and other services on the irrigated schemes rehabilitated under the present and on the earlier project. 43 7.5 The rate of return for the project calculated over 10 years is 27 percent. The net present value of the cost and benefit streams applying a discount factor of 12 percent is US$3.3 million. The switching value for the project benefit stream at an opportunity cost of capital of 12 per cent is 19%. The sensitivity analysis of the project costs and benefits indicates that the break even point (the point at which the ERR falls below 12 percent) is reached by a 23 percent increase in costs or a 19 percent decrease in benefits or alternatively a combined 10 percent increase and decrease in costs and benefits. The project is relatively more sensitive to a fall in the benefit stream than to an increase in the cost stream as Table 7.1 demonstrates. Analysis of the projections for increases in rice yields indicates that the project is relatively insensitive to shortfalls in the achievement of increased rice production. A 10 percent shortfall in incremental rice yields would bring the project ERR down to 22 percent. A 35 percent shortfall on projected increased rice yields would be required to bring the ERR down to 12 percent, all other things being equal. Sensitivity against slow achievement of project benefits indicates that a full year's delay in achieving project benefits would be required to bring the ERR down to 13 percent. Table 7.1: Sensitivity Analysis of the ERR to Changes in Costs and Benefits Increase in Decrease Increase in Costs in Benefits Costs and Decrease in Benefits 10 percent 19% ERR 19% ERR 12% ERR change 15 percent 16% ERR 15% ERR 7% ERR change 23 percent 12% ERR 10% ERR 7.6 One of the main benefits from improved water management on the newly irrigated schemes is the increased potential for production of non-traditional crops. The farm models which have been prepared for the above analysis include only potatoes, triticale and fish farming as diversification from traditional rice production. Approximately 20 percent of the project benefits are attributed to production of such crops. The potential is however considerably greater and should be the object of strengthened ties between the extension and research services as project implementation advances. B. Environmental Impact 7.7 Irrigation has a major positive impact in terms of increasing yields and increasing national productivity and has the indirect benefits of improving local standards of living and local diets. In Madagascar, irrigation also has a positive impact outside the irrigated schemes in that it alleviates agricultural pressure on the marginal tanety (hillside) lands. For an irrigation scheme to fully benefit from a rehabilitation program, it is important that rehabilitation is carried out in a manner that is technically and environmentally sound. The direct environmental impacts of rehabilitation include problems due to errors and omissions in technical designs and the damage 44 caused by construction during the actual physical rehabilitation. With proper supervision such impacts can be reduced to negligible proportions. The direct impacts most often associated with intensive irrigation such as soil degradation, surface water contamination by fertilizers and pesticides and the increased risk of water borne diseases appear insignificant on the rehabilitated irrigation schemes in Madagascar and should pose no major threat with sound farming and maintenance practices. 7.8 External impacts of rehabilitation include the degradation of watersheds upstream of the rehabilitated schemes, which has had a moderate to high adverse impact on certain schemes during and after rehabilitation. Cyclones also have had on occasions a moderate to very high impact on the environment, both during the rehabilitation phase and during actual perimeter operations. 7.9 Soil Fertility. Rehabilitation which permits the possibility of two crops per annum might be considered, if fertilizers are used, to pose a threat to the soil fertility, which could be considered a directly negative environmental impact of the project. However, to date there is no evidence to suggest that there is diminishing soil fertility on the rehabilitated schemes. On the contrary, there have been observations suggesting that the introduction of crops during the off- season with fertilizers brought a complementary benefit to rice production the following season. 7.10 Health Risks. Increased activity in the rehabilitated schemes could contribute to increasing the risks of contamination of diseases for which the vector requires an aquatic phase such as malaria and bilharzia, two diseases endemic to Madagascar. However, the potential impact of this risk is low given (i) the important proportion of lands already under irrigation and (ii) the compensatory impact of improved drainage resulting from rehabilitation. However, it would certainly be possible to reduce the impact of these diseases by introducing measures which lead to greater availability of prophylactics, such as Nivaquine, in the project area. 7.11 Effects of Intensification of Agricultural Production. The effects of rehabilitation of the irrigation schemes should contribute to improve farmer incomes on the schemes. This in turn should have a marginal effect on reducing the impact of clearing land elsewhere for new production, although the growing population pressure probably nullifies any such benefits. Nonetheless, rehabilitation of irrigation schemes provides an opportunity to negotiate a contractual arrangement for the use of other areas, such as catchment areas From this point of view, the impact of the project can be considered to be beneficial. 7.12 Environmental Actions. By conducting an environmental analysis prior to any new rehabilitation, the project will make a major effort to avoid adverse effects of rehabilitation experienced to date. For the rehabilitated schemes under the first project, a scheme-specific evaluation will be carried in one retrospective evaluation and limited environmental protection works carried out subsequently. This evaluation will provide mitigation recommendations which will serve as a reference in the development of future mitigation plans. C. Risks 7.13 The success of the project depends, among other things, on the successful transfer of the management and the maintenance of the rehabilitated schemes to local WUAs. The project has incorporated the principle of progressive rehabilitation, with an assessment of physical and monetary contributions at each stage of the rehabilitation. In addition, a more intensive process 45 of local feed-back and participation in project execution by beneficiaries will permit project staff to assess the levels of local commitment of the WUAs at different stages. Given the progress registered under the first project, there is a high probability that management and maintenance responsibility will be successfully transferred. 7.14 The lack of intensified agricultural extension support provided by the Directorate for Agriculture following rehabilitation was one reason for the only modest improvements in agricultural productivity and production during the first project. The risk of a lack of collaboration by the Department of Agriculture is reduced in the light of the (a) the renewed political commitment by the Ministry of Agriculture on improving the effectiveness and professional quality of its national agricultural extension service; and (b) the resources which will be mobilized specifically for the Department of Agriculture to improve the quality of extension support on and around the rehabilitated schemes, unlike the previous project, when such resources were not specifically identified. 7.15 In order to ensure longer term sustainability of the rehabilitation program, virtually no recourse is anticipated to long term technical assistance. In its place, short term assignments by consultants, both national and expatriate, are envisaged. There is a risk that this might result in a higher incidence of technical errors in rehabilitation design. 7.16 Insufficient funding of the Government's contribution to the project could have a negative impact on execution. However, the project budget approval and financing arrangements are such as to minimize the impact of delinquency by any cofinancier. IDA funds will be made available on a lender of last resort basis, and annual work programs and budgets will be designed accordingly. An intensified dialogue between the Bank and the Government on macroeconomic questions will focus attention on the public expenditure program, through which priority counterpart funding obligations would be flagged. The risks of poor Government funding performance seriously affecting successful project implementation are low. 7.17 The risk of environmental damage to water catchment areas feeding newly rehabilitated schemes, as encountered under the first project, would be minimized by the environmental assessments proposed in the selection of additional schemes for rehabilitation. On the basis of historical evidence, there is nonetheless an five to ten percent possibility that rehabilitated schemes will suffer from cyclone damage or other natural disaster within ten years of rehabilitation. 7.18 On the basis of an analysis of the foregoing risks, there is a better than seventy percent chance overall that the project will meet its development objectives. 46 VIII. AGREEMENTS REACHED WITH THE BORROWER, CONDITIONS AND RECOMMENDATIONS 8.1 A draft "Irrigation Sector Policy Statement" has been prepared by the Ministry of Agriculture (see Annex 2) and was transmitted to the Bank officially with a letter from the Minister for Agriculture dated .... 1994. 8.2 The following assurances were obtained during negotiations: (i) A project execution program, budget and financial plan will be submitted annually before October 31, 1994 for the following year's activities, including training proposals and details of procurement procedures to be followed (para 3.35); (ii) Project accounts would be audited annually by independent auditors acceptable to IDA, in conformity with the norms of International Standards on Auditing (para 4.31); (iii) Audits of financial accounts will be submitted to IDA not more than six months after the close of the financial year to which they refer (para 4.31). (iv) Reports on project implementation progress would be submitted to IDA and other cofinanciers semi-annually at the end of March and at the end of September each year. (para 4.32). (v) The MEADR would fix a common policy on field allowances for all projects before the end of 1994 (para 3.24). (vi) The MEADR confirmed its agreement to the maintenance of an irrigation sector coordinating committee and the appointment of a coordinator (para 4.15). (vii) Government would carry out, in collaboration with donors, a mid-term evaluation of the project not later than October 31, 1996 (para 4.22). (viii) Government would carry out in consultation with all concerned donors, an annual review of the Public Expenditure Program for the agricultural sector (para 3.35) 8.3 The following would be conditions for presentation of the credit to the Board: (i) The signature of a ministerial decree providing recognition of the Water Users' Associations as legal entities. (ii) The adoption and the submission to the Bank of the final version of the agreed policy statement. 8.4 The following would be a condition of effectiveness of the credit: The deposit of an initial amount equivalent to US$100,000 to the project advance account (para 4.35). 8.5 With the above assurances and conditions, the above project is suitable for an IDA Credit of SDR 15 million. 47 MADAGASCAR SECOND IRRIGATION REHABILITATION PROJECT Annex 1 DETAILED COST TABLES Table 1: Project Components by Year - Base Costs (USs U0) Beas cost 94/ 9M 36197 97/NI 8S Total A. Transfrt des pdrknkres r6habilits aux AUE Renforcement des Instilutions Paysannes 744 2,189 60 61 61 3,094 Tmvau Confortafs 620 413 - - - 1,033 Subtotal Transfit des pdrkntres r6habilit6s aux 1,364 2,582 60 61 61 4,128 B. Protection de rErwronnerwt 304 333 315 230 276 1,458 C. Rdhabilitation des nouveaux p6rm6trs - 104 2,158 4,004 1,950 8,215 0. Appul aux Services du Minagri Coordinaton centrale 410 141 108 91 40 791 Renforcemernt des Services Regoneux 98 1,101 1,322 1,221 694 4,438 Subtotal Appul aux Services du Minagri 509 1,243 1,430 1,313 734 5,229 E. Formation 247 122 121 96 79 666 F. SuM-EvWluien at Audit 53 39 28 28 17 165 G. Entraden des P4rmres per les AUE 138 276 276 310 345 1345 Total BASELINE COSTS 2,616 4,698 4,388 6,041 3,462 21,204 Physical Contingencies 262 470 439 604 346 2,120 Price Contingencies Inflation Local 127 443 673 1,103 944 3,290 Foreign 41 139 192 387 251 1,009 Subtotal Inflation 168 582 865 1,489 1,195 4,299 Devaluation -32 -254 -490 -660 -493 -1,930 Subtotal Price ContIngencies 136 328 374 829 701 2,369 Total PROJECT COSTS 3,013 5,496 5,201 7,475 4,509 25,694 Taxes 292 637 606 873 491 2,900 Foreign Exchange 1,857 3.187 2,820 4225 2,205 14,295 Table 2: Components Project Cost Summary (FMG VW) (us$ WO) % % Ttal % % Tea Foreign Base Foreign Base Local Foreign Total Exchange Costs Local Foreign Total Exchange Costs A. Transfort des p6rkmtres rdhabilit6s aux A Renforcement des institutions Paysannes 3,222,000 5,751.667 8,973,667 64 15 1,111 1,983 3,094 64 15 Travaux Confortat 951,750 2,044,500 2,996,250 68 5 328 705 1,033 68 5 Subtotal Trans~rt des p6rntres r6habilttis 4,173,750 7,796,167 11,969,917 65 19 1,439 2,688 4,128 65 19 B. Protection de rEnvironneenet 1,525,889 2,701,511 4,227,401 64 7 526 932 1,458 64 7 C. Rdhabilitation des nouveaux p6rim6tres 8,260,370 15.563,333 23,823,704 65 39 2,848 5,367 8,215 65 39 D. Appul aux Services du Minagri Coordination centrals 606,393 1,687,204 2,293,597 74 4 209 582 791 74 4 Renforcement des Services Regionaux 6,746,095 6,123,076 12,869,171 48 21 2,326 2,111 4,438 48 21 Subtotal Appul aux Services du Minagri 7,352,488 7,810,280 15,162,768 52 25 2,535 2,693 5,229 52 25 E. Formation 1.085,000 845,833 1,930,833 44 3 374 292 666 44 3 F. Suivi-Evaluation at Audit 168,925 308,367 477,292 65 1 58 106 165 65 1 G. Entretien des P6ftrAtres par es AUE 3,900,000 - 3,900,000 - 6 1,345 1,345 - 6 Total BASELINE COSTS 26,466.423 35,025,491 61,491,914 57 100 9,126 12,078 21,204 57 100 Physical Contingencies 2,646,642 3,502,549 6,149,191 57 10 913 1,208 2,120 57 10 Price Contingencies 10 754,746 11,139,467 21,894,213 51 36 1,360 1,009 2,369 43 11 Total PROJECT COSTS 39,867,811 49,667,508 89,535,318 55 146 11,399 14,295 25,694 56 121 ( 49 Annex 1 Table 3: Expenditure Accounts Project Cost Summary % % Tota % % Total Formsgn Bas Forelgn Base Local Forign Total Exchange Cost Local ForMign Total Exchange Costa 1. Bnvestm Cook A. Treveux Teraln Hydraulque 10.113.750 21,725,833 31,839.583 68 52 3,488 7,492 10,979 68 52 Envrnnmen 952.000 2.088,000 3.040,000 69 5 328 720 1,048 69 5 Survmlkontrole 988,000 1,005,333 1,993,333 50 3 341 347 687 50 3 Subtotal Trvax Teran 12,053,750 24,819,167 36.872,917 61 T 4.156 8,558 1715 67 60 L Baud"ent Rhah n bure u 304,657 443,056 747,712 59 1 105 153 258 59 1 RtiahmlaUn logment 443,137 644,444 1,087 582 59 2 153 222 375 59 2 subtotalk mnt 747,794 1,087.500 1.835.294 59 3 258 375 833 59 3 C. Vehåculk~ Tout t~nai 880,952 805,556 1,686,508 48 3 304 278 582 48 3 Lager 352,381 322.222 674,603 48 1 122 111 233 48 1 Motocye~ths 376,650 352,350 729.000 48 1 130 122 251 48 1 Bicycletta 126,857 116,000 242,857 48 - 44 40 84 48 Subbotal Vehicula~ 138 1.596,128 3,332,968 48 5 599 55 1,149 48 5 D. Equipment MobilrdtEqup.mnt 296,475 582,900 879,375 66 1 102 201 303 66 1 E. Asssac Tchniqu* Coufttrme - 3.608.889 3,608,889 100 6 1,244 1,244 100 6 Local 752,000 - 752.000 - 1 259 - 259 - 1 Subtotal Acianc Techniqus 52,00 3.,,889 4,360,889 3 T 259 1,244 1,W4 83 7 F. Etude. Bursurtrn9er - 294.833 294.833 100 - 102 102 100 - Bureuurineanr 922,074 618,667 1,540,741 40 3 318 213 531 40 3 Fris tndrnce de rET 25,000 2500 - - 9 - 9 - Subtotal Etudes 947,074 913,500 1,860,574 49 3 32T 315 642 49 3 0. Formeln Formrbanlocale 1,559,000- 1,559,000 - 3 538 - 538 3 Formtio a reS - 792,667 792,667 100 1 - 273 273 100 1 Frans de ubsäan 108,000 108,000 37 - 37 - Subtotal Formaon 1,667,000 9 4 7 73 8 3 4 H. Entren paul AUE 3,900,000 3900,000 6 1,345 1,345 - 6 Total Invetmnt Cost 22,100,934 33,400.750 55,501,684 60 90 7.621 11,518 19,139 60 90 Ii. Recurrnt Costs A. Indemnte Ingeneua 475.800 475.800 - 1 164 - 164 - 1 Techr~cien 324,000 - 324,000 - 1 112 112 - 1 Vulga ~ur 1.440,000 - 1,440.000 2 497 - 497 - 2 Chaulleur 33,000 - 33,000 - 11 11 - - Subtotal indenta2,8 2,272,800 - 4 784 - - 4 L Entrden des Vehicule Tout terrn 755,630 590,730 1,346,360 44 2 261 204 464 44 2 Leger 270,486 211,458 481,944 44 1 93 73 166 44 1 Motocyc~ette 276,545 216,195 492.740 44 1 95 75 170 44 1 SubtEtalEnr~tin dmVehcules 1,302,661 1,018,383 2.321,044 44 4 449 351 800 44 4 C. Entreln badumenta Bure u 279,176 406,000 685,176 59 1 96 140 236 59 1 D. Fouirnt 9ueux BurenuD 1con 180,416 70,760 251,176 28 - 62 24 87 28 - Burau Terran 279.498 109,620 389,118 28 1 96 38 134 28 1 Docunentaon 50.937 19,978 70,915 28 - 18 7 24 28 - Subtota Foum~tur Bureaux 510,851 200,358 711.209 28 1 176 69 4 28 1 Total Recurrent Coet 4,366,489 1.624,741 5,990,230 - 2 10 1,505 50 6 27 10 Total BASELNE CotTs 26,466,423 35,025,491 61,491,914 57 100 9.126 12.078 21 _=3T 100 Ph~yscaltCaningences 2.646.642 3,502,549 6,149,191 57 10 913 1.208 2,120 57 10 Pric Catigencies 10754746 11,139,467 21.894,213 51 36 1,360 1009 2389 43 11 Total PROJECT COSTS 39a67,811 89,53,318 5 4 11,399 14,96 4121 50 Annex 1 Table 4: Expenditure Accounts by Years - Base Costs (US$ 000) as Coft Foreign Exchange 949 96196 9MU 97 9796 9M96 Total % Amount . Invegint Coa A. Travaux Terraln Hydruque 1.328 2.538 1,778 3,556 1,778 10,979 68.2 7,492 Envann* uiem 44 306 262 218 218 1,048 68.7 720 SurveWonrole - - 172 344 172 687 50.4 347 Subtotal Travaux Trran 1,372 2,844 2.212 4.118 2,168 12,715 67.3 8,558 8. Batiments Re.hliftn bure~u - 94 94 70 258 59.3 153 Råb~linaogem.ni. - 141 141 94 - 375 59.3 222 Subt~ Mmnta - 234 234 164 - 633 59.3 375 C. Veail s Touttnnn 116 140 140 140 47 582 47.8 278 Lagere Bl 70 70 - 12 233 47.8 111 Mo ~oaste - 68 68 68 47 251 48.3 122 Bicycls~. 31 31 21 84 47.8 40 SubtaVehicules 198 300 309 229 105 1,149 47.9 550 D. Equipument MobNer et Equipmnt 43 98 98 64 - 303 66.3 201 E. Aza'tanc Techniquc Court terme 467 289 228 178 83 1,244 100.0 1.244 Local 79 26 42 60 53 259 - - Subtotalalstanca Technique 546 314 270 237 136 1,504 82.8 1,244 F. Etudes Bure~uxtrangers 102 - - - 102 100.0 102 Buraux locaux 15 119 264 119 15 531 40.2 213 Frak d'nnr de rET 2 2 2 2 2 9 Subtotalhua.a 118 120 268 120 17 642 49.1 315 G. Formatdon Fornadon ~ocal 88 129 129 107 84 538 - - Foryaon a rad@~eiu - 64 100 91 18 273 100.0 273 Frais de cubak-n 4 8 8 8 8 37 - Subtotal Fonmation 92 201 238 207 110 848 32.2 273 H. EntrUen perl AUE 138 276 276 310 345 1,345 - - Total Investment Costa 2,507 4.398 3,903 5.450 2,881 19,139 60.2 11,518 Ili. Recurrent Costå A. Indemntes Ingenieu 18 22 42 42 42 164 Technicien 9 19 28 28 28 112 - - Vulgwisteur 41 83 124 124 124 497 - - Chauffeur 1 2 3 3 3 11 - - Subtotal undmntes 68 125 197 197 197 784 - - 8. Entretian dås Vehcues Toul terin 24 67 110 139 124 464 43.9 204 Leger 16 29 40 44 38 166 43.9 73 Motocyel~. - 18 35 53 65 170 43.9 75 Subtotal Entreden desVöices 39 113 185 236 227 800 43.9 351 C. Entretan batimenta Bureux - 30 44 81 81 238 59.3 140 D. Founiture Bureaux BureeuDiredon 1 11 21 27 27 87 28.2 24 Bureu Terrain 17 32 42 42 134 28.2 38 Docurentati 1 3 6 8 7 24 28.2 7 subtotal Fourniture Bureaux 2 31 59 78 76 245 28.2 69 Total Rcurrent Costa 109 300 485 591 581 2,066 27.1 560 Total BASELIE CO8TS 2,518 4,698 4,388 6.041 3,462 21,204 57.0 12.078 Physical Conngencies 262 470 439 604 346 2.120 57.0 1,208 Prlc Contngfni~ innaton Local 127 443 673 1,103 944 3,290 - - Foreign 41 139 192 387 251 1000 100.0 1,009 sutalnnation 168 582 866 1,489 1.196 4,299 23.5 1,000 Devumnn -32 -254 -490 -60 -493 -1,930 - SubMotal Prims Condngencie 136 328 374 829 701 2,369 42.6 1,009 Total PROJECT COSTå 3,013 5,46 5,201 7,475 4,500 25,694 55.6 14,296 Taxs 292 637 606 873 491 2,900 - Foreign Ex=hang 1,857 3,187 2,820 4,225 2,206 14,295 - 51 Atxnex i ;U3 *r.e~ I - ~ ::: jI~ ~ ~ Nf4 ,-.- ~- -a ~ Oa wo 0 0 I I e. ~ ~e. I- il i I I I ,. I .- I ~ 4 - E 0 h I- glj - .111 liii ~ I .2 In 1ff .1 ~ 11.11 fl ui- i- ~- i'- 52 Annex 1 1 aI a st W~ 0 0f 0 2.. kk 2 -2- - 66c 21 12 1 . lnfn l nn~ - fl 53 Annexl -,―名鑼―召呈寫必劃呂k；寫。劇8呂日劉。―刨文合。露：―悶，文變―斗瀾；萬州糾引用 I】l啊鳥！g；二凡一―g&，憫一，―闖啊一＃豐―零―觀9：零一―之寫亡訌―冷n•雲、―勾―細―毒 I蔆■口1 1 11州1111→〕i ―一l&I&’婪’―婪”&&&-‘病豐悶‘―審―森不悶兀們―寫9啊也！叩召久零、―乏1綱．萬 I•―變萬！名審’乏兀―尺曦零必h劊荔8豐胄乏―豐1芝不胄兀們―電叩啊斗―召不久響、―細l剽悶 l，酒卜•1黑：&&―騙，&―•甲一州｀―︰―論h門。―甲：n頃I又•、，―、―潺―文 11.1名；l萬響史召不】呂：8•曦豐―蠶8妝；乏1望―哥萬胄兀n―沼細不瞬―調哥尺認的I話―寫l悶 1一l名；I萬響變露萬l合；g”必叩I寫8久；森l習l婪9巴斗刊l了；文甲l石寫9釁內l寫―荔I日 l,I一黑I夾了｀，門―另，•一―。9內，&1丰―荔”•―;，州一―•們一一―h―妝19 -,-&&I&&&&-&&&&-&&&&l’〕”&‘一〔‘&&’〔‘&&&l&l&l& ,―尋藝寫蘊旦畏日曆添學盡論：藝日巨喜藝藝逐規絕藝C豐絕 居召― &1，〕’&&&&&&&“邑”&&，界牌：安！也：· --&&&‘叩”&&&&8婿‘量韃暈騷叉彎；，坼騙’ !.―一‘ -―另旦’‘文軍”日”&&&8尋？員馴遷騷‘,,,，騙’ ,&&＿。． 黑――賽多‘’文呂’,&&&’呂§”遷革霆屈，,&&”寫’ 糒一―鑒―&’網 頂―一― 蒲――賽拯”“多’叩”&&“騷’邑呈里騷”&,“日－ 嘗l變l內 量―返〕 亡．--&’為”&&&&&’騷屈醫婿’&& 個口細■■．9&j i}_.“二 邑’-&,&~&“纏‘’罐”&,,,,，結 巨 O 啼d !.,：。：＿。；,,，。。。！:，。＿！。 遛〕．1 11．。．！。j．名要喜I,．。！!!l渥。，．韋＿!I。。！看11蘆纏！I ,!!11!!!,l，〕！l！。！}I!}!!!!i}i!&!!,i!!,!i!,!!!!i〕， Table 10: Technical Capacity Building - Detailed Costs Quantities mase cost unit 64M as= Wit elm 1041m Total Unit Cost --gM 91M1 OW m7m qeM TOW L investffwd Coda Fmmmilon des Agents do Wnlin 31 31 62 62 62 248 onnedon des services god" 34 34 26 17 112 Fonnedw gedlon at budgets kww 7 7 - - - 14 AT - 101 3 2 1 1 17.5 NMI 175 50 33 17 17 m Total 247 W 121 79 m Table I I: Monitoring/Evaluation and Audit - Detailed Costs (US$'000) Ouar"I" Ba" cod unit "m 96196 96197 97/98 9"S Total Unit Cost SQ116 96M 96197 97011 911419 Total L Investinad Costs A. SulviOlEvakiation k-n maidrial emquate forfak 3 - - - - 3 r- Frals crwqufteurs forfak 2 2 2 2 2 9 Mirioo"lifumuns rib a - - - 8 1.509 12 - - - - 12 AT do m" tVm 2 2 1 1 6 11.111 22 22 11 11 - 67 Audit dm cmWes forfak 15 15 15 Is is 75 TOW 53 39 28 28 17 165 Table 12: WUAs Contribution to Maintenance - Detailed Costs (US$ DW) Unit Bass Cost Link 94M Ofim NMI 97NO Total Cod "M 91W%G UN7 VM -MM 1. inves&nwd Costs on nwidw bal6ft 8.000 16.000 16.0w 18.000 20,ODO 78,000 OV7 138 276 276 310 345 1345 0 TOW :3 138 276 276 310 34 1.34 ID 55 Annex 2 Declaration of Irrigation Sector Policy MADAGASCAR Second Irrigation Rehabilitation Project (Draft) I. Agricultural Development Strategy 1. Within the framework of its overall development policy, the Government of the Republic of Madagascar aims both to improve the standard of nutrition of the population, develop and diversify exports, and supply its domestic industries with raw materials, while at the same time improving management of the natural environment. 2. The low growth rate of the agricultural sector in recent years may be attributed to four factors: an unfavorable macroeconomic environment, an unbalanced pattern of public investment, inadequacy of the legal and regulatory framework, and poor natural resource management. The agricultural sector has the potential to become one of the driving forces of accelerated economic growth. Given the geography of the country and the constraints linked to settlement and development of the underpopulated regions, such growth is going to come less from an expansion of activities in those regions than from the intensification and diversification of production in already developed areas, although this does not imply that other areas would be neglected. The Government intends to gradually redeploy the public resources in such a way as to promote the establishment of the conditions that will favor the sector's accelerated development. 3. Macroeconomic Environment. The Government confirms its commitment to a liberal economic policy favoring the free play of market forces. It is pursuing its objective of direct divestiture from the production, processing and marketing sectors. To offer the necessary economic incentives, it will include among its new policies: - an exchange policy that will enable exports to remain competitive by preventing overvaluation of the currency and will ensure freedom of access to foreign exchange (both for goods and services and for capital transactions); - a tariff and fiscal policy that will not discriminate against the agricultural sector in favor of the other sectors of the economy; the restructuring of the customs tariff is designed to eliminate intersectoral distortions in this area; - a financial policy (reduction of the budget deficit and improvement of the banking system) that will increase the availability of credit to the private sector, thereby facilitating private investment and the development of competition. 4. Public Investment. As regards public investment in the agricultural sector, the Government intends to make better use of its limited financial and human resources by gradually reducing the number of independent projects existing in the field. It intends to combine these 56 development efforts into a small number of national programs. These programs will seek greater harmonization of the efforts so far made in the field by adopting those formulas and approaches that have yielded the best results and the best cost/benefit ratio. It will mobilize domestic and external funding to ensure that a greater number of farmers have access to such basic specialized services as agricultural research, extension and animal health. It proposes to streamline the public expenditure budget for the agricultural sector by reducing the share of capital and operating expenditure in the irrigation sector in favor of improved maintenance of public physical infrastructures such as roads and marketplaces and cofinancing of the programs mentioned above. It is understood that the maintenance of certain structures (such as those found on the GPIs) is beyond the technical and financial capacities of the users and will therefore be performed by the Government as a public service. 5. This new approach will be accompanied by the decentralization of government actions, as provided under the Constitution of the Republic of Madagascar. The Government intends to promote greater participation by local beneficiaries in the infrastructure construction, rehabilitation and maintenance (O&M), focusing in particular on getting them to assume the responsibility for operation and maintenance of the irrigation works that presently lies with the Government. It intends to expand the role of the local authorities and economic operators in the work of repairing and maintaining farm roads and to promote the development of local markets. 6. Legislative and Regulatory Framework. Certain reforms have already been undertaken, notably those involving rewriting of the Investment Code and of legislation on free trade zones, which measures are intended to stimulate investment throughout the entire economy. The Government intends to pursue these efforts by improving the Labor Code and introducing one- stop windows to streamline the procedures involved in the setting up of a new business. It is also in process of drafting new legislation concerning such matters as quality promotion and control, occupational organizations, inter-occupational associations, and the procedures for transferring responsibility for maintenance and management of irrigated areas. These drafts will shortly be submitted to the National Assembly for ratification. 7. Natural Resource Management. Within the framework of its Environmental Action Plan, the Government places a strong emphasis on the need for effective natural resource management and protection of the irrigation infrastructure as essential to the achievement of sustainable development. New investments will not take place until the pertinent environmental impact assessment has been made. Formulation of the procedures to be followed in such assessments is nearing completion. 57 II. General Irrigation Sector Development Policy 8. Being aware of the importance of irrigated agriculture to the economy in general and to the agricultural sector in particular, the Government has for several years been mobilizing large amounts of money to stimulate the productivity and efficiency of this subsector. Despite these efforts, the irrigation sector is characterized by a low level of performance that falls far short of its potential. This is due to: (a) lack of diversification, since the principal crop grown in the irrigation schemes is rice, produced largely for on-farm consumption; (b) stagnating agricultural yields; (c) little progress in the transfer to the users of the responsibility for managing and running the classified schemes; and (d) lack of irrigation infrastructure maintenance and poor water management. Nevertheless, the economic and financial analysis has confirmed that, based on the existing technologies and crops, intensification of irrigated agriculture remains a proposition that is both financially attractive for the farmers and profitable for the country's overall economy. 9. The Government's strategy in support of the irrigation subsector is to continue to provide the favorable framework that is necessary to the intensification and diversification of irrigated farming. More precisely, it will encourage the WUAs to assume the O&M costs of the schemes that have up to now been managed essentially by the Government, providing them with the appropriate expert services and technical assistance and supporting the creation of farmer groups and associations. In particular, the Government proposes to: (a) concentrate efforts and resources on improving the institutional operation of the irrigation subsector, defining the respective institutional responsibilities and capacities of subsector operators, both public and private. In particular, the farmers will be encouraged to participate and to make their wishes known; (b) curb transfers of public funds to irrigated farming, reducing operating subsidies to a level that will nonetheless allow the farmers to meet their obligations. This reduction of subsidies will help promote both the optimum use of existing irrigation infrastructure and the intensification of agriculture by freeing up public funds for use in the support of such projects as roads, agricultural extension, agricultural research and marketing information services, all of which will facilitate agricultural intensification and diversification; (c) make all new public investments in the irrigation subsector (regardless of source of financing) subject to the following principles: (i) full participation by the beneficiaries of the recommended investments in the management and maintenance costs of irrigation infrastructure; (ii) systematic attention to the environmental aspect, which will add to the sustainability of the investments; and (iii) the assurance of a satisfactory return on the investments, both for the individual beneficiaries and for the irrigation scheme as a whole. 10. The following actions will be taken to implement this policy: (i) consultation and agreement between the public authorities and the users concerning their respective responsibilities as regards scheme operation and management; (ii) improved resource mobilization at both the central and the decentralized levels, through coordination and concentration of local and external financing in support of sectoral public service programs; (iii) improvement of the programming and planning process at the local level, with active WUA participation; (iv) decentralization and 58 strengthening of the human and material resources of the Ministry of Agriculture, so as to enhance coordination and improve the quality of the technical support services provided to the farmers; (v) strengthening of training activities targeting both the public authorities and the beneficiaries of the irrigation scheme rehabilitation programs; and (vi) enactment of legislation governing the procedures for transfer to the users of the responsibility for the rehabilitated irrigation schemes. 11. The volume of financing needed to revamp the national network of classified irrigation schemes outstrips the Government's present financial capacity. It is therefore understood that future financing priorities for both GPI and PPI rehabilitation projects will be allocated in priority to those schemes where the water users have already demonstrated the commitment and capacity to form themselves into organizations capable of assuming responsibility for the operation and management of a rehabilitated network. This demonstration of commitment and capacity on the part of the farmers is thus a condition for the execution of any new rehabilitation project or program. The criteria to be adopted for the selection of additional schemes will be identified and described on the basis of experience gained in the next phase of the irrigation rehabilitation program. 12. The extent of the various maintenance and management responsibilities under any new rehabilitation programs will be described in the terms of reference prepared for each individual scheme at the time of the rehabilitation study, in consultation with the water users and other beneficiaries. These terms of reference will in particular define the role of the government officials and the WUAs and the annual levels of financing needed to enable both sides to meet their respective commitments. The principles governing the transfer of responsibility for O&M costs are covered by ministerial decrees Nos. ..... dated ..... 13. The specific details of implementation of this policy are presented in the attached table. November 12, 1993 OBJECTIVE STRATEGY ACTION PLAN STATUS 1. Ensure sustainability of 1.1 Creation of and support for 1.1 1 Improve and enforce the regulatory framework governing the establishment, Ongoing irrigation infrastructure, in the WUAs in all the irrigation operation and legal capacity of the WUAs. order to: schemes - foster intensification of 1.1.2 Provide the WUAs with all necessary technical, organizational and logistic Ongoing agriculture support. - promote diversification of agricultural systems 1.1.3 Evaluate reaulta obtained in the process of empowering the WUAs to take Iterative - promote the region's specific over management and maintenance of irrigation networks. advantages 1.2 Land-holding security 1.2.1 Facilitate access to the land (land-tenure and cadastral surveys, ODOC). In progress 1.2.2 Enact regulations governing owner/farmer relations, clearly spelling out In progress respective maintenance responsibilities 1.3 Redefinition of tasks and roles 1.3.1 Redefine the roles of Gnie Rural in the area of irrigation scheme By July 1995 of the departments of the Ministry management and maintenance, as follows: of Agriculture and Rural to provide advice to the WUAs; Development to manage and take action to safeguard structures when public safety io in jeopardy; - - actions to safeguard the infrastructure (closing of networks whose sustainability is jeopardized by defective management and/or maintenance). 1.3.2 Strengthen local agencies in all aspects of their work, pending effective By end 1966 implementation of the decentralization process. in the Project areas 1.3.3 Prepare regulations and manuals of procedure deriving from the development By end 1997 of government functions and roles in the irrigation sector. 1.4 Resource orientation in light 1.4.1 Perform diagnostic study of WUF1A operations. By October of the WAs' capacity to assume 1966 responsibility for their schemes 1.4.2 Determine levels of contribution. Management contract 1.5 Effective user participation in 1.5.1 Ensure effective user participation in all study phases and in the development Ongoing development of the irrigation activities. achemes 1.5.2 Conduct in-depth orientation studies. Ongoing 1.5.3 Favor a gradual approach to the development activities. Ongoing 1.6 Environmental protection for 1.6.1 Systematize the diagnostic study of the irrigation schemes' environmental Ongoing the irrigation schemes situation and its impact on their sustainability, and identify necessary conservation starting 1997 or restoration activities. 1.6.2 Implement and monitor the proposed activities with effective participation by Ongoing the target population. starting 1997 1.6.3 Participate in the drafting of environmental protection regulations (PAE Completed Program). 2. Reduction of Government's 2.1 Government divestiture from 2.1.1 Improve and enforce regulations on transfer of management taking account Ongoing financial contribution to the management and of specific features of the schemes and the regions. starting 1994 irrigation sector maintenance of irrigation networks 2.1.2 Promote alternatives for maintenance work that cannot be performed by Studies to be users. conducted in 1995 2.1.3 Facilitate operator access to the purchase or rental of network maintenance MEADR equipment. seminar June 1996 2.2 Mobilization of resources of 2.2.1 Set up a system to finance maintenance of the irrigation systems with Studies in all sector operators contributions from all parties concerned (WUAs, economic operators, decentralized progress authorities, Government). 2.2.2 Set up a permanent emergency [intervention d'urgencel fund as a sustainable By end 1997 source of financing to cover cyclone damage. 2.2.3 Promote good donor coordination in the irrigation sector. Coordinating Committee 61 Annex 3 ENVIRONMENTAL ASSESSMENT SECOND IRRIGATION REHABILITATION PROJECT 1. The environmental component of the Madagascar Irrigation Rehabilitation Project includes all recommendations for environmental mitigation developed by the Environmental Analysis (EA) of the project. For EA purposes the project is rated category B in view of the rehabilitation aspect of project interventions and the limited extent of small-perimeter irrigation. The project's environmental analysis is one of the first in the agriculture sector in Madagascar and is expected to be helpful in design of environmental policies and assessment criteria for other agricultural projects as well. Project environmental actions will be coordinated with specialized agencies, notably ONE and ANAE. ONE has broad oversight responsibilities in environmental assessment which are currently being defined more precisely. The mandate of ANAE is to coordinate ail activities in erosion control, deforestation and watershed management. Environmental Impacts of Rehabilitation 2. Irrigation has a major positive impact in terms of increasing yields and increasing national productivity and has the indirect benefits of improving local standards of living and local diets. In Madagascar, irrigation also has a positive impact outside the irrigated perimeters, in the sense that it alleviates agricultural pressure on the marginal 'tanety' and 'tavy' lands. 3. For a perimeter to fully benefit from a rehabilitation program, it is critically important that rehabilitation is carried out in a manner that is technically and environmentally sound. Two categories of environmental impacts of rehabilitation are distinguished: direct impacts and external impacts. The direct impacts concern the effects of the rehabilitation process on the environment, and the external or indirect impacts pertain to the influence of the environment on the irrigated perimeters. In many ways, the external impacts tend to aggravate the direct impacts. 4. Direct Impacts. Two direct impacts are of special importance in the rehabilitation of perimeters: (a) the problems due to errors and omissions in technical design and (b) the damage caused by construction during the actual physical rehabilitation. With proper technical inputs and sufficient care during construction, those impacts can and must be reduced to negligible. The direct impacts that are normally considered in intensive irrigation such as soil degradation, surface water contamination by fertilizers and pesticides, and the increased risk of water born diseases appear insignificant in the rehabilitated perimeters in Madagascar and should remain at that level with sound farming and maintenance practices. 5. External Impacts. Degradation of watersheds upstream of the perimeters has a moderate to high adverse impact on some vulnerable perimeters. In addition, cyclones have on some occasions a moderate to very high impact on the environment, in particular during the rehabilitation phase but also during actual perimeter operations. 62 Environmental Actions 6. The project will make an important effort to remedy or to avoid the adverse environmental impacts of rehabilitation, of past as well as present rehabilitation. A key step in this process is to conduct an environmental analysis prior to the rehabilitation of new perimeters. The project will institute specific EA for all rehabilitation perimeters and promote the assessment process. Three studies will be financed and institutional support will be provided. 7. In the case of existing rehabilitated perimeters, perimeter-specific environmental analyses will be done in one single retrospective evaluation, one perimeter after the other. For the new perimeters, individual environmental analyses will be part of the standard perimeter selection and planing process which is currently being revised. To assist in preparation of these perimeter- specific analyses, a series of mitigation recommendations have been developed by the project. These recommendations are intended to serve as a reference in the development of environmental plans. 8. Environmental Analysis and Actions for Rehabilitated Perimeters. The immediate purpose of the environmental analysis is (1) to identify the environmental problems of all 30 perimeters that were rehabilitated in the first phase of the project, and (2) to propose corrective measures, including "travaux confortatifs". It is best to carry out this analysis before end 1994, in order not to delay continued rehabilitation efforts. The irrigation infrastructure improvements will be integrated in the project's "travaux confortatifs" programs as defined in the Hydro- Agricultural component of the project. Watershed management works will not be limited to damage control, but will also include preventive measures. 9. Urgent protection measures in rehabilitated perimeters will be included, as far as practical, in irrigation infrastructure work already assigned to construction companies. Simple conservation measures, which can be implemented by farmers themselves without much project support, will be identified. In the case of the latter, the technical specifications, work schedules and maintenance requirements will be determined jointly by farmers and Government agents (ADR/CA/CH). 10. Environmental Analysis and Actions for New Rehabilitation. The strengthening of environmental aspects during the studies phase, the implementation and perimeter operation will be ensured by systematic environmental analysis of each new perimeter. Most environmental measures will aim at prevention and avoidance of adverse environmental impacts. Environmental analyses for the new perimeters will become part of the preparation process of each perimeter as they will be integrated into the newly developed "Etudes Approfondies d'Orientation (EAO)" which will be carried out for each perimeter. The analyses will be conducted by the environmental specialist of the EAO team. Each of these evaluations will include an assessment of the impact of the upstream watershed on the perimeter and consideration of the impact of rehabilitation on neighboring perimeters. 11. The new EAO procedure for new perimeters evolved from a more restricted approach used in phase one of the project to strengthen farmer participation and to handle enviromnental concerns. As a result, the EAO now comprises a series of parallel studies on the following subjects: irrigation infrastructure, agriculture-livestock, environment, social organization and institutional requirements, and financial feasibility. The integration of environmental analysis in the EAO ensures that environmental concerns in perimeter rehabilitation are identified at an early 63 stage, and that they can be incorporated into rehabilitation design, which is expected to lead to more sustainable perimeter rehabilitation. Studies and Institutional Support 12. The environmental analyses will be supported by a series of studies which are scheduled to be carried out at the beginning of the second project phase. Three studies are involved: (1) a study of the legal aspects of cyclone damage, (2) a study of the vulnerability of irrigation perimeters to erosion, and (3) a synthesis of the approaches and techniques of watershed management currently known and applied in Madagascar's four main natural regions. 13. The start-up, supervision, and monitoring of environmental analyses, actions and studies will be entrusted to a special Environmental Unit at the "Direction du Genie Rural (DGR)". An environmental specialist will be responsible for the Environmental Unit. The coordination with other Government and private agencies necessary to implement environmental tasks will be in collaboration with the National Coordination office of the project. 14. Malgache consulting firms need to strengthen their environmental assessment capability in order to effectively contribute to perimeter rehabilitation. The project will provide limited support to these firms to assist them in building up their environmental competence. Two environmental training workshops are scheduled to be organized in collaboration with ACMA (Association of Malgache Consultants) to focus on environmental assessment of agricultural projects. Cost 15. Total basic costs (without inflation) of the environmental component of the project are US $1,682,000, or about 3,200 million FMG. About 82% of this total amount is allocated to environmental analyses and improvements which will benefit approximately 40 perimeters of both project phases, 8% of the total amount is allocated to support studies, and 11% to institutional support. 64 Annex 4 MADAGASCAR SECOND IRRIGATION REHABILITATION PROJECT Rehabilitation of Medium-Sized Irrigation Schemes (PPIs) by Source of External Funding during the First Project (PPI-I) (hectares) Ralisations Surface 1986 1987 1988 1989 1990 1991 1992 1993 Total (ha) Banque mondiale 1 Andreambe 130 - - - - 130 - -- 130 2 Mananantanana 733 - - - 500 133 100 - 733 3 Ankona 250 - - - 250 250 4 Basse-Ankona 437 - - - - - 437 437 5 Alakamisy Ambohi 457 - - - - - - 457 457 6 Ivakoanina 182 - - - - - - 182 182 7 Anjiajia 452 - - - - 452 452 8Ankijabe 120 - - - - 120 - 120 9 Ampary 101 - - - - - 101 - 101 10 Analavory 140 - - - - - - 140 140 11 Antanimenakely 88 - - - - - 88 - 88 12 Ambalanirana 113 - - - - - 113 - 113 Sous-total IDA 3 203 630 133 662 1 778 3 203 Caisse fran;aise de d6veloppement 13 Soavina 1 200 600 600 - - - - - 1 200 14 Ambohibary 2 380 - - 1 200 1 180 - - - -2380 15 Manandona 820 - - 300 520 - - - - 820 16 landratsay 630 - - 130 500 - - - 630 17 Vinaninony 710 - - 350 360 - - - - 710 18 Onive 1 040 - - - - - - - 800 800 19 Vatambe 436 - - - - - - 200 236 436 20 Ikianja 379 - - - - - - 379 379 21 Ivato 360 - - - - - - - 300 300 22 Lot 1 Manakara 1 095 - - - - - - 500 500 22 Lot 2 Manakara 1 368 - - - - - - 300 1 068 1 368 Sous-total CCCE 10 418 600 600 1 980 2 560 - - 500 3 363 9 523 FED 23 Behara 800 200 600 - - - - - - 800 24 Belamoty 1 200 300 900 - - - - - 1 200 25 Taheza R.D. 1 200 - - - - 600 600 - - 1 200 26 Manombo Ranozaza 935 - - - - - - - 935 935 27 Ankerika 252 - - - - - 126 126 252 28 Andengondroy 52 - - - - - - 26 26 52 29 Maroadabo 115 - - - - - - 50 65 115 Sous-total 4 554 500 1 500 - 600 600 202 1 152 4 554 1 PNUD/FAO/PAM 32 Andakana 225 - - - 2 - 225 - - 225 33 Mangamda 180 - - - - 80 100 -180 Sous-total 405 - - - - 305 100 - 405 PNUD/FAO0PAM TOTAL GENERAL 18 580 1 100 2 100 1 980 2 560 1 230 1 038 1 464 6 21317 685 65 Annex 5 MADAGASCAR SECOND IRRIGATION REHABILITATION PROJECT Calculation of Incremental Benefits/Costs Stream and Internal Economic Return RICE PY1 PY2 PY3 PY4 PY5 PYa PY7 PYs PYS PYlo W,rhour p4scr Area (he) 16000 1a000 16000 18000 20000 20000 20000 20000 20000 20000 Y,eid 1.9 1.9 1.9 1.9 1.9 1.9 1.9 1 9 1.9 1.9 Producl3w (tonnes) 30,400 30400 30400 34200 38000 38000 38000 38000 38000 380CO W1 r pfq- Area the) 15000 16000 16000 18000 20000 20000 20000 20000 20000 20000 Yiwd 2 2.1 2.2 2.4 2.4 2.4 2.4 2.4 2.4 2.4 Increa,se l croppng inteneity % 1.05 1.1 1.1 1.1 1.1 1.1 1.1 1 1 1.1 1.1 Production (tonnes) 33600 38960 38720 47520 52800 52800 52800 52800 52900 528C Incremental producton 3200 6560 8320 13320 14800 14800 14800 14800 14800 14800 POTATOES Withou preoa Aru& (Hal s0 80 80 80 80 80 80 80 80 80 Yield 1d 16 15 16 16 15 16 16 16 15 Producton Itomnes) 1280 1280 1280 1280 1280 1280 1280 1280 1280 1280 WrA Pq Area the) 150 300 450 500 500 500 500 500 500 500 Y.Jd 16 1a 1d 18 le id la 16 16 16 Production lionnes) 2400 4800 7200 8000 8000 8000 8000 8000 8000 8000 Incremental production 1120 3520 5920 6720 6720 6720 6720 8720 6720 6720 TRITICALE wrhour pr4ecr Area ihs) 90 90 90 90 90 90 90 90 90 90 Y,ed 2.5 2.5 2.5 2.5 2.5 2 5 2.5 2.5 2 5 2.5 Production (tomesl 225 225 225 225 225 225 225 225 225 225 Wth pf*err Area tha) 150 300 450 500 500 500 500 500 500 50 Yad 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2 5 Production (tonnool 375 750 1125 1250 1250 1 250 1250 1250 1250 1250 lncrmental production 150 525 900 1025 1025 1025 1025 1025 1025 1025 FIS H Wm0our pect Area iha) 100 100 100 100 100 100 100 100 100 100 Y,od 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0 04 Producton (tonnaol 4 4 4 4 4 4 4 4 4 4 With pocrt Area he) 150 200 300 400 400 400 400 400 400 400 Y".id 0.04 0.1 0.2 0.25 0.3 0.3 0.3 0.3 0.3 0 3 Producon itonnes) a 20 60 100 120 120 120 120 120 1.0 Incremental production 2 16 50 96 116 116 115 116 116 16 Economac vsakm of knaernWtal pro&Kxion a/ Paddy @ 224 715800 1469440 1863580 2983680 3315200 3315200 3315200 3315200 3315200 3315200 Potatoes O 83 9290 292160 491360 557780 557780 557750 557750 557750 557750 557760 rtcate @ 248 37200 130200 223200 254200 254200 254200 254200 254200 254200 254200 Fsh 0 857 1714 13712 47992 82272 99412 99412 99412 99412 99412 99412 TOTAL 848674 1905512 2625232 3877912 4225572 4225572 4226572 4225572 4226572 4226572 Total n ' 000s 1000 849 1906 2820 3878 4227 4227 4227 4227 4227 4227 1 yew delay n benefits 0 849 1906 2526 3878 4227 4227 4227 4227 4227 Etnnmo prec Costs (not of taxse 184 2504 4835 6481 3552 685 685 685 685 185 Ned aef"it /Costa -835 -598 .2010 -203 675 3542 3542 3542 3542 3542 Net B/C wth I yew deey nB .1684 .1655 -2731 -3854 326 3542 3542 3542 3542 3542 Internl rate of rerum 27% IER wth yar deay In benefits 13% al detaAled caculaton of econome pnces taken from FAO/CP preoaration report 9193CP-MAG 35 -nnex 7. 66 Annex 6 SECOND IRRIGATION REHABILITATION PROJECT Monitoring and Evaluation Performance Indicators Indicator PY 1 PY 2 PY 3 PY4 PY5 Water Users' Associations Execution of maintenance works by WUAs (%) 100 100 100 100 100 Membership dues (percentage collected) 80 90 100 100 100 Maintenance dues (percentage collected) 50 65 80 90 90 Average number of meetings by WUAS 2 2 2 2 2 Performance of Irrigation schemes Water availability - percentage covered regularly 50 60 70 80 80 Agricultural production Increases in rice yields (%) 5 5 10 10 10 Increase in area planted to potatoes (ha) 150 300 450 500 500 Increase in area planted to triticale (ha) 150 300 450 500 500 Increase in the area of fish farming (ha) 150 200 300 400 400 Agricultural extension Number of demonstration plots per ADR 40 40 40 40 40 Rate of adoption of technical messages (%) 10 20 30 40 50 Harmonization of agricultural extension systems 5 10 15 30 30 in the project area (cumulative number of schemes) New rehabilitation schemes Indicators to be fixed during the mid-term 0 0 0 review 67 Annex 7 MADAGASCAR SECOND IRRIGATION REHABILITATION PROJECT Estimated Schedule of Disbursements (US$ million) IDA FY/Quarter End DISBURSEMENT ACCUMULATED % DURING QUARTERS DISBURSEMENTS FY95 September 30, 1994 0.2 0.2 1 December 31, 1994 0.5 0.7 2 March 31, 1995 0.3 1.0 1 June 30, 1995 0.3 1.3 1 September 30, 1995 0.4 1.7 2 December 31, 1995 0.5 2.2 2 March 31, 1996 0.5 2.7 2 June 30, 1996 0.5 3.2 2 September 30, 1996 0.7 3.9 3 December 31, 1996 0.8 4.7 4 March 31, 1997 0.8 5.5 4 June 30, 1997 1.0 6.5 5 September 30, 1997 1.0 7.5 5 December 31, 1997 1.2 8.7 6 March 31, 1998 1.2 9.9 6 June 30, 1998 1.2 11.1 6 September 30,1998 1.2 12.3 6 December 31, 1998 1.2 13.5 6 March 31, 1999 1.3 14.8 6 June 30, 1999 1.3 16.1 6 September 30, 1999 1.3 17.4 6 December 31, 1999 1.3 18.7 6 March 31, 2000 1.2 19.9 6 June 30, 2000 1.2 21.1 6 68 Annex 8 Documents Contained in the Project File 1. Rapport d'évaluation 1985 à 1992 - Coordination Nationale du Projet PPI - Cellule de Suivi-Evaluation - Mars 1993 2. Project de Réhabilitation des Petits Périmètres Irrigués - Phase 2 - Mission de Préparation Vol I et Il. Centre d'Investissement Programme de Coopération FAO/Banque Mondiale - janvier 1993. 3. Plan d'Action 1991-1995 (actualisation Janviuer 1993) - Secretariat Général, Ministère de l'Agriculture - Comité de Coordination du Projet de Réhabilitation des Petits Périmètres Irrigués 4. Contribution à la Revue du Sous-Secteur Irrigation - Henri Boumendil - Consultant en irrigation - janvier 1991 4. Evaluation de la Perception des Associations d'Usagers de l'Eau par leurs membres et diagnostic sur l'évolution des AUE - Paul Mathieu, Consultant socio-organisateur (LBI) - Août 1991 5. Etude de Sociologie du Fonctionnement des Périmètres Irrigués - A. Razakarivony et E.P. Brown. Mars 1990 6. Assessment of Environmental Issues Realting to Irrigation Investments in Madagascar - Jeffrey Metzel and Katherine Baird for Associates for International Development, Dec 1990 7. Textes Législatifs AUE - Vol I et Il - Edition Bilingue - Ministère de l'Agriculture - PPI Février 1991 69 Map: IBRD No25595 - Second Irrigation Rehabilitation Project    믐：&