Document of The World Bank FOROFFICIAL USEONLY ReportNo 37200-MG PROJECTAPPRAISALDOCUMENT ONA PROPOSEDCREDIT INTHEAMOUNTOFSDR20.2 MILLION (US%30.0MILLIONEQUIVALENT) TOTHE REPUBLICOFMADAGASCAR FORAN IRRIGATIONAND WATERSHEDMANAGEMENT PROJECT INSUPPORTOFTHE FIRSTPHASEOFTHE IRRIGATIONAND WATERSHED MANAGEMENT PROGRAM October24,2006 Environmental,Ruraland Socially SustainableDevelopment-AFTSl Country Department1,Madagascar Africa Region This document has a restricted distribution and may be used by recipients only in the performance o f their official duties. Its contents may not otherwise be disclosed without World Bankauthorization. CURRENCY EQUIVALENTS (ExchangeRate Effective September 2006) Currency Unit = Ariary (MGA) 1MGA = US$0,00047 SDR = US$1.48620 US$ = SDR0.67286 FISCAL YEAR January 1 December 31 - ABBREVIATIONSAND ACRONYMS ASC Agricultural Service Center AFD AgenceFranFaise de Dbeloppement (French Development Agency) APO Agricultural Professional Organization BV Bassin versant (Watershed) BV-PI Bassin versant -PdrimBtre iwigud (Watershed -IrrigationScheme) CAS Country Assistance Strategy CDP Communal Development Plan csc Communal Support Centre DFB Directorate for Finance and Budget DGDR Direction Gdndrale de Ddveloppement Rural (General Directorate for Rural Development) DRDR Direction Rdgionale de Ddveloppement Rural (Regional Directorate for Rural Development) DSI DepartmentofInformation Systems DTA Decentralized Territorial Authorities FERHA Fonds d 'Entretien deRdseam Hydro Agricoles (Irrigation Maintenance Fund) FMG Malagasy Franc (F)WUAs (Federation of) Water Users Associations GDP Gross Domestic Product GEF Global EnvironmentalFacility GELOSE Gestion Locale Sdcuriske (Community LandTenure Management) GoM Government of Madagascar IDA InternationalDevelopment Association MAEP MinistryofAgriculture, Animal HusbandryandFisheries MDAT Ministryof Decentralization andLandDevelopment MEEF MinistryofEnvironment, Water andForest Resources MEM Minister o f Energy and Mines NAP National Action Plan NSC National SteeringCommittee O&M Operation andMaintenance PC Performance Contract PLOF Local LandOccupation Plan PN/BV-PI Programme National Bassins Versants-PdrimBtres Irriguds (National Irrigation andWatershedManagement Program) PMU Project Management Unit PRSF Poverty ReductionStrategy Framework PSRP Poverty Reduction Strategy Paper W D Research-Development RDFB Regional Directorate for Finance andBudget RDP Regional Development Plan RMC Regional MonitoringCommittee RPMU Regional Project Management Unit SLM SustainableLandManagement TT Tranoben'ny Tantsaha-ChambersofAgriculture UNCCD UnitedNations Conventionto CombatDesertification WB WorldBank WUA Water Users Association WDP WatershedDevelopmentPlans WMP WatershedMaster Plan Vice President: GobindNankani . Country Director: htvaReinikka Sector Manager: RichardScobeyIFrankByamugisha Task Team Leader: IJsbrandde Jong MADAGASCAR Irrigation and Watershed Management Project TABLE OF CONTENTS A. STRATEGIC CONTEXT AND RATIONALE.......................................................... -4 Country and Sector Issues.......................................................................................... -4 eligibility ................................................................................................................... Rationale for Bank involvement. relation to Country Assistance Strategy and GEF Higher level objectives to which the project i s contributing......................................... -7 8 B. PROJECT DESCRIPTION......................................................................................... 9 LendingInstrument.................................................................................................... Program Objectives and Phases................................................................................... -9 9 11 Project Components .................................................................................................. Project Development Objectives and Key Indicators................................................. Lessons Learnedand Integrated into the Project Concept.......................................... 12 16 Alternatives considered and Reasons for Rejection ................................................... 17 C. IMPLEMENTATION............................................................................................... 17 Partnership Arrangements......................................................................................... 17 Monitoring and Evaluation of Outcomes/Results ...................................................... Institutional and Operational Arrangements ............................................................. -18 20 CriticalrisksandPossible ControversialAspects ...................................................... Sustainability and Replicability................................................................................. 20 21 D. PROJECT BRIEF SUMMARY................................................................................ 24 Economic and Financial analysis ............................................................................. -24 -25 Fiduciary .................................................................................................................. Technical................................................................................................................. 26 Environmental analysis ............................................................................................. Social analysis .......................................................................................................... 27 27 -28 Policy Exceptions and Readiness .............................................................................. Safeguard Policies ................................................................................................... 30 Annex 1: National. Sectoral and Program Context.............................................................. - 31 Annex 2: Major RelatedProjects Financedby the Bank and / or Other Agencies.................40 Annex 3: Results Framework and Monitoringand Evaluation .............................................. Annex 4: Detailed Project Description................................................................................. 42 Annex 5: Project Costs ......................................................................................................... 48 60 Annex 6: Institutional and Implementation Arrangements ................................................... Annex 7: Financial Management and Disbursement Arrangements...................................... 61 74 Annex 9: Economic and Financial Analysis......................................................................... Annex 8: ProcurementArrangements .................................................................................. 83 Annex 10: Safeguard Policies Issues ................................................................................. 88 105 Annex 12: Documents inthe Project File........................................................................... Annex 11: Project Preparation and Supervision .................................................................. 112 113 Annex 13: Statement o f Loans and Credits......................................................................... Annex 14: Country at a Glance........................................................................................... 114 116 Annex 15: IncrementalCost Analysis................................................................................ Annex 16: Technical Annex Land DegradationinMadagascar.......................................... 118 128 140 Annex 18: APL Triggers ................................................................................................... Annex 17: STAP Reviewand Response to Review............................................................ 153 Annex 19: Maps (IBRD 35115, IBRD35116) ................................................................... 155 MADAGASCAR IRRIGATIONAND WATERSHED MANAGEMENT PROJECT PROJECTAPPRAISAL DOCUMENT AFRICA AFTSl Date: October 24,2006 Team Leader: IJsbrandHarko de Jong Country Director: Ritva Reinikka Sectors: Irrigation anddrainage (50%); Agro- Sector ManagedDirector: Michel Wormser industry (30%); Crops (20%) Project ID: PO74086 Themes: Other rural development (P); Other trade LendingInstrument: Adaptable Program Loan andintegration(S) Environmental screening category: Full Assessment Global Supplemental ID: PO88887 Team Leader: IJsbrandHarko de Jong LendingInstrument: Specific InvestmentLoan Sectors: Irrigation anddrainage (50%); Agro- FocalArea: LandDegradation industry (30%); Crops (20%) SupplementFullyBlended?: No Themes: Ruralmarkets (P); Other trade and integration (S) For Loans/Credits/Others: 1 1Borrower: Government o f Madagascar: Ministryo f Economy, Finance and Budget B.P. 61 Antananarivo 101, Madagascar Tel: 261 20 22 382 86 Fax: 261 20 22 34530 ResponsibleAgency: MinistryofAgriculture, Livestock andFisheries B.P. 301, Anosy Antananarivo 101, Madagascar Tel: (261-20)-22-27227 Fax: (261-20)-2226561 infoamaep.gov.mg 1 FY 7 8 9 10 11 1 Annual 1.45 4.34 6.95 I 10.56 6.70 Cumulative 1.45 5.79 12.74 23.30 30.00 FY 7 I 8 9 10 11 IAnnual 0.15 0.36 1.74 1.98 1.77 Cumulative 0.15 0.5 1 2.25 4.23 6.00 Does the project depart from the CAS incontent or other significant respects? Ref: [ ]Yes [XINO Does the project require any exceptions from Bank policies? Ref: PAD D.7 [ ]Yes [XINO Have these been approved by Bank management? [ ]Yes [ IN0 I s approval for any policy exception sought from the Board? [ ]Yes [ IN0 Does the project include any critical risks rated "substantial" or "high"? Ref: PAD rL.Jc [ ]Yes [XINO Does the project meet the Regional criteria for readiness for implementation? Ref: PADD.7 [XIYes [ ] N o Project development objective Ref: PAD B.2, TechnicalAnnex 3 The development objective o f the project is to sustainably increase agricultural productivity infour high potential watersheds and their associated irrigation schemes. Global Environment objective Ref: PAD B.2, TechnicalAnnex 3 The global environmental objective o f the project i s to improve the environmental sustainability of land management practices infour targeted watersheds. Project description [one-sentencesummary of each component] Ref: PAD B.3.a, TechnicalAnnex 4 The proposed project comprises three technical components covering major strategic orientations: (i) Development o f Commercial Agriculture, (ii) Irrigation Development and (iii) Watershed Development. The fourth component is Program Management. In accordance with the 'growth poles' approach, the project proposes similar investments inthe four regions concerned (Annex 1). Component 1 "Development o f Commercial Agriculture" aims at improving access to markets and supporting the development o f commercial agriculture value chains and providing demand-based support to private investment; Component 2 "Irrigation Development" aims at improving the quality o f irrigation services and Operation and Maintenance (O&M), and puttinginplace a more favorable environment for agricultural intensification and diversification; Component 3 "Watershed Development" aims at sustainably managing watersheds including irrigated agriculture, preserving the natural heritage, benefiting from the production potential o f the natural resources, and contributing to improved living conditions and incomes o f the rural population; and Component 4 "Project Management" aims at managing and using resources in accordance with the project's objectives and procedures, and puttingin place a policy framework that is favorable to upscaling o f the project at the national level. 2 Whichsafeguardpolicies are triggered, ifany? Re$ PAD 0.6, TechnicalAnnex 10 The following safeguardpolicies are triggered: EnvironmentalAssessment, Natural Habitats, Pest Management, Involuntary Resettlement and Forests. Significant, non-standardconditions, if any, for: Re$ PAD C.7 Credit effectiveness: 0 adopted the Project ImplementationManual, the Project Monitoring and Evaluation Manual and the Project Administrative, Financial and Accounting Manual, all in form and substance satisfactory to the Association; 0 established a Financial Management System (FMS), in form and substance satisfactory to the Association, to ensure proper executionand monitoring of Project activities; it is understoodthat this includesrecruitment of financial managementTA; appointed the independent auditors; 0 prepared the tenderhidding documents to carry out the procurementrelatedto the preparationof the Watershed andIrrigationManagement Schemes programmedfor the first Project year; and 0 establishedthe National SteeringCommittee (NSC) and Regional MonitoringCommittees (RMCs) in form and with functions satisfactoryto the Association. Covenants applicable toproject implementation: 0 The Government will adopt, by December 31,2006, and thereafter implement, a national strategy on fertilizers; 0 The Government will issue, by December 31, 2007, and thereafter implement, legislative guidelines for the applicationof the new seeds legislation; a baseline survey for the Projectbe completedby December 31,2006; 0 the Watershed Master Plans (including Watershed Development Plans and Scheme Development Plans), oneper each Project Area, be completedby September 30,2007; 0 operational modalities of management and replenishment of Fonds d'Entretien de Rkseaux Hydro Agricoles (Irrigation Maintenance Fund) (FERHA) be definedby September 30,2007; 0 the irrigation related legal framework, including but not limited to Law no.90-016 and associated legal guidelines, will be harmonizedwith the Irrigation and Watershed Management Policy Letter by December 31, 2007. Such harmonization to include, inter alia: the management of autonomous irrigation schemes and partner schemes, the process for irrigation management transfer and the membership of the W A S . 3 A. STRATEGIC CONTEXT AND RATIONALE Country and Sector Issues 1. Background.Madagascaris one of the poorest countries inthe world, with per capita income o f about US$298 per year (2005). The economy is basically rural, with agriculture as one o f the main engines o f economic development. According to the latest household survey (2004), the poor represent about 69 percent o f the total population, but 77 percent o f the rural population. After a long economic decline during the 1970's and 80's structural reforms were undertakeninthe late 1990sthat resulted in the resumptionofeconomic growth (4.6 percent on averagebetween 1997-2001) and someprogress in poverty reduction, essentially limited however to urban areas. Poverty did not decline in rural areas where three fourths o f the population live. Economic growth, however, was short-lived. Contested presidential elections at the end of 2001 led a to six monthpolitical crisis which brought the country to a virtual standstill.The economic collapse and discontinuation of many public social services in 2002 caused widespread social suffering, in particular in rural areas where farm producer prices collapsed, especially inisolated areas, causing an unprecedenteddrop inrural incomes. 2. The new government elected in 2002 moved aggressively to restore public services and macroeconomic stability. After a sharp decline o f 13 percent in 2002, economic growth rebounded to 9.8 percent in 2003 and 5.3 percent in 2004. It is projected at 4.6 percent in 2006. With continued political stability and sound macro-economic policies, prospects for sustained economic growth are good. The macroeconomic projections forecast an average growth rate o f around 5.6 percent over the medium term (2006-2010). There is enormous growth potential in several growth sectors, including mining, construction, tourism (supported by the country's naturalbeauty and uniquebiodiversity) and, most important for the poor, agriculture. 3. Poverty Reduction Strategy Framework The development objective of Madagascar, as defined in the Poverty Reduction Strategy Paper (PRSP, July 2003) is the promotion of a rapid and sustainable development in order to reduce poverty by half within ten years. It is organized around three strategic orientations: (i)to restore the rule o f law and good governance; (ii) to foster and encourage broad-based economic growth; and (iii)to foster and encourage human and material security and enlarged social protection. The Government's strategy for broad-based growth targets five general objectives: (i)to reach an economic growth rate of 8-10 percent per annum; (ii)to increase the level o f investmentto 20 percent; (iii) to promote the vitality of the private sector so that it participates inaninvestmentrate of 12-14percent; (iv) to open upMadagascar's economy to greater competition with a view to reducing costs and improving quality; and (v) to foster the willingness o f the population to participate. In agriculture, the PRSP aims at "ensuringfood security and making optimal use of resources" through, inter alia, increased agricultural productivity and cultivated areas. The Government is in the process o f formulating its second-generation strategy, the Madagascar Action Plan (MM)that gives renewedemphasisto ruralandagricultural development. Box 1:The MadagascarAction Plan The Madagascar Action Plan (MAP) 2012 is the Government's new vision that sets out a "roadmap" for development.The purpose of the MAP is to produce a quantum leap in the development process by having a five year plan that will mobilize the Malagasy people and the country's international partners to ignite rapid growth, thus leading to the reduction o f poverty. Its goal is also to ensure that the country develops inresponse to the challenges of globalization and in accordance with the national vision "Madagascar Naturellement" defined by the Presidentin November 2004 (describedbelow), the long-term vision for the country of which the MAP is a 5-year implementation program. The MAP is a direct follow-on from the PRSP and covers the period 2007-2011. The MAP will be finalized in October 2006 and presentedto Parliament together with the 2007 budget. Madagascar Naturellement is the statement of the country's development vision in 2015. It states that Madagascarwill be a newly industrialized country with maximized competitivenessby 2020. The core of growth will be derived from the country's unique natural resources and from the transformation of its natural products. The vision aims to develop a diversified and rich natural resource base (agriculture, livestock, fisheries, and 4 mining) that will contribute to the creation of products with high value added such as essential oils, agri- business, pharmaceuticals, and mining products. A broader impact of growth and aprogressiveredistributionof its benefits will help reduce poverty substantially. Madagascar will be known worldwide for the beauty of its rich and well-protectedbiodiversity and its environment will be cherished and protected and usedin a wise and responsible way to enhance development. The Malagasy people, both in rural and urban areas, will be healthy and well-educated, will be active participants in the development process and will be gainfully employed in agriculture, industryand the provision of services. Education and healthwill be accessible to the populationand infrastructurewill be developedallowing for free movementof goods andpeople. The MAP'Sgoals are ambitious targets that the Government is committed to achieving over the coming five years. It i s recognizedthat these targets are challengingbut, at the same time, it is expectedthat resultswill be delivered through the mobilization of the full array of the available resources (human, technical and financial), combinedwith strongleadership, hardwork, creativity and excellent coordination.The MAP outlines eight specific areas of focus; (i)good governance; (ii)educational transformation; (iii)health and family planning; (iv) infrastructure; (v) rural development; (vi) the economy and the private sector; (vii) environment; and (viii) nationalsolidarity.There are quantifiable goals set out for each of these areas of focus to be achieved by 2012; these goals will be further refined, costedandclearly linkedto the MDGs. 4. At the end of 2005, the Government of Madagascar (GoM) adopted a National Rural Development Planwhich is centered around five strategic axes: (i) to makethe institutional framework more effective by completingpublic administration reformand decentralization; (ii) to facilitate access to capital and production factors; (iii) to improve food security and production through increased productivity, diversification, and risk management; (iv) to promote better natural resources management; and (v) to develop markets and promote a value-chain approach by encouraging public- private partnerships. The proposed Irrigation and Watershed Management Project follows these principles closely. 5. Agriculture, Rice and Irrigation. Agriculture remains the foundation of Madagascar's domestic economy. It contributes about one third o f the total Gross Domestic Product (GDP) and 40 percent of total exports. About three quarters o f the population depend on agriculture for their livelihood. About one-half o f Madagascar's landarea i s cultivable, but little more than 5 percent o f the land is currently under crops, with a large part of the cultivated area under irrigation (about 40 percent). Rice is the main staple crop, accounting for 70 percent o f total farm output. Traditional export crop include vanilla, coffee and cloves (Madagascar is the largestproducer o f natural vanilla in the world). Livestock is widespread, with about 60 percent of rural families depending on it for their income. Fishing and aquaculture has become one o f the most dynamic elements in the primary sector o f the Malagasy economy. The shrimp sector is the second largest source o f foreign exchange earnings andprovides about 62,000 direct jobs and218,000 indirectjobs for the country. 6. Performance of the agricultural sector has been disappointing in recent years, despite the liberalization of the economy, the sharp devaluation o f the exchange rate and the privatization o f state enterprises. The under-performance o f the agricultural sector i s a major cause o f the deep poverty in rural areas. Farming systems are still very traditional. Two-thirds o f all rural households live at subsistence level and yields are generally very low. Weak infrastructure hampers the transport o f produce, whether for export or for the domestic market.Agricultural productivity i s also hamperedby the poor access to agricultural technology, inputs and other agricultural services. Extension services are all but lacking. Only 1.5 percent o f Madagascar's small farmers have access to credit, and a mere 5 percent of total lending goes to agriculture. Traditional land tenure systems do not give farmers sufficient security. 7. Rice represents nearly 70 percent o f agricultural production and accounts for 48 percent o f total calorie consumption. Rice productionhas increasedby 1.2percent per annumsince the 1980sbut average paddy yield at the national level remains low (about 2.6 tha). Annual production of paddy rice has virtually stagnated over the past ten years, stabilizing between 2.3 and 3.0 million tons. Area planted to paddy has increasedby only 0.4 percent per year from 1970 to 2004; yields have increased by 0.7 percent per year, much slower than in other major rice producing countries. With an annual 5 population growth o f 2.7 percent, production per person has fallen from 275 kglpersonin 1970 to 179 kglperson in 2004. Rice farming techniques are largely traditional and use o f inputs i s the exception in many places. e.g., fertilizer use has remained stagnant at 10 k o a on average, as compared to 14k o a insub-SaharanAfrica, and291 k o a inIndonesia. The vast difference inprices between wet and dry season i s explained by the lack o f fluidity inmovement o f goods from production areas to the markets due to a lack of road infrastructure and lack of management capacity o f storage facilities by farmers. On average, 28 percent o f the paddy production i s marketed (750,000 t), but rice sales are highly concentrated. In2001, the top 10 percent of rice farmers (by value o f sales) accounted for 73 percent o f total national rice sales. These farmers sold on average 2.2 tons per household. An estimated 48 percent o f rice farmers did not sell any rice in2001. 8. Irrigation occupies an important place inthe agricultural sector, supplying water to more than one million hectares, or 40 percent of cultivated lands (as compared to 6 percent on average in sub- Saharan Africa). Irrigated crops represent 15 percent o f Gross Domestic Product (GDP), whereas 70 percent of agricultural production and 88 percent of rice production originate from irrigated agriculture. It i s estimated that 85 percent of the active farming population are directly or indirectly employed by the irrigation sector. Since the 1950s, irrigation has benefited from public investment. However, the impact o f these investments on rural incomes is mixed, and sustainability is far from certain. The rapid degradation o f infrastructure requires frequent rehabilitation, and many schemes are caught in a vicious circle o f poor yields, low capacity o f water users to pay for Operation and Maintenance (O&M), and rapid degradation of the schemes. Weak capacity to pay i s accompanied by low willingness to pay, reinforced by institutional weakness o f Water Users' Associations ( W A S ) and a lack of support from local authorities. Moreover, erosion o f upstream watersheds is weighing heavily on cost o f maintenance o f downstream irrigation schemes. 9. Extension services have also failed to have a significant impact on productivity levels, and have demonstrated to be unsustainable. Reasons for these past failures include (i)a bias towards technical messages, (ii)inadequate consideration o f demand for extension services and economic constraints that farmers face; (iii) a too centralized approach, with inadequate attention for regional variation, (iv) inadequate capacity o f extension agents, and (v) unrealistic expectations about the volume o fpublic (humanand financial) resourcesavailable. 10. Land degradation, natural resources and land development. Land degradation is one o f the most serious and widespread problems for the agricultural sector in Madagascar. The degradation dynamics inthe uplands and lowlands are often linkedand reinforcing each other. With the stagnation o f yields in the irrigated lowland areas and demographic growth, farmers extend their agricultural activities on the hillsides. Upper watershed land use is often based on extensive and unsustainable management practices, the most important being lack o f erosion control and lack o f soil fertility management on agricultural plots, slash and burn agriculture (tuvy), and the frequent burning o f pastures. Landdegradation i s also causedby deforestation for agricultural purposes, causing increased carbon emissions, biodiversity loss and declining regulatory ecological services. These practices not only contribute to the degradation andlow productivity o f uplands but also impact lowlandagriculture significantly. Upland soil erosion and water surface run-offalso causes sedimentation for downstream infrastructure, contributing to the reduction of cultivated area under irrigation, local flooding o f rice paddies inthe rainy season andwater shortagesinthe dry season. 11. Lessons from past experience. There is a clear emerging consensus on the need to adopt a new approach to the development o f irrigation in Madagascar. A recent World Bank report on the impact o f public spending on the productivity o f irrigation' confirms stagnant paddy productivity at the national level, but also found that overall investments insmall-scale community irrigation schemes have had a positive localized impact on productivity. However, even on those schemes that have benefited from investments, an important yield gap remains and weak sustainability is identified as a * Madagascar:The Impactof PublicSpendingon the Productivityof Irrigation Schemes (1985-2004) 6 key factor in the disappointing performance of irrigated agriculture. Recommendations include (i) promotion o f green revolution and agro-ecological technologies in order to boost productivity beyond the existing level, and (ii)an appropriate incentive framework for operation and maintenance o f irrigation infrastructure, as well as financial mechanisms against hurricane damage. 12. The need to adopt an approach to agricultural intensification that reaches beyond mere rehabilitation o f infrastructure has been confirmed by the Economic and Sector Work 'Madagascar - Rural and Environment Sector Review (2003)'. Based on household interviews and econometric analysis, the report presents a comprehensive list o f constraints to increasing productivity, including access to finance, inputs, markets and equipment, problems associated with land degradation and sedimentation, and lack o f maintenance of irrigation infrastructure. Past experience thus strongly emphasizes the need to adopt an integrated approach to agricultural intensification in Madagascar's watersheds. This new approach should (i) aim at the establishment o f an appropriate incentive and financing framework for efficient operation and maintenance o f irrigation infrastructure, as well as for the mitigation o fdamage causedby the frequenthurricanes that affect the country; but also (ii) address a wide range o f issues in agricultural development as well as soil and water conservation in upper watersheds. 13. Global experience provides ample affirmation. A recent Bank report `Watershed management operations: Approaches, challenges and emerging lessons (ARD 2006)' proposes a new approach to watershed management that distances itself from previous top-down approaches that are based on traditional engineering, towards a new paradigm involving an integrated and participatory approach. The approach proposes to build strong local institutions, and to address a wide range o f goals across soil and water conservation and integratedrural development. 14. Irrigation and Watershed ManagementPolicy Letter. These lessons and the needto adopt a long-term, comprehensive approach to irrigation development are fully reflected in the Government o f Madagascar's new Irrigation and Watershed Management Policy Letter that summarizes the Government's policy objectives with respect to the development o f irrigation. The Policy Letter provides a clear description o f the reasonswhy there was little development inthe country's irrigation sector and o f the constraints that a new development strategy should address. Accordingly, the Policy Letter clearly states that while it is urgent to spur agricultural production in these high potential areas by rehabilitating existing perimeters, this should be done only after critical pre-requisites are met: (i) linking producers to markets through improvement in transport and connectivity infrastructure, (ii) establishing a supportive policy environment and ensuring adequate access to critical services; (iii) ensuringthe buy-infrom stakeholdersthrough adequateconsultations; (iv) improvingtheir capacity to pay and to manage; and (v) establishing the institutional framework and incentives that will ensure accountability for adequate use and maintenance o f the infrastructure. This Irrigation and Watershed Management Policy Letter will provide the framework for the proposed operation. Rationalefor Bankinvolvement, relationto CountryAssistance Strategyand GEF eligibility 15. The Government o f Madagascar has requested World Bank (WB) and Global Environmental Facility (GEF) funding for an Irrigation and Watershed Management Project to accelerate economic growth in rural areas, through an integrated effort aimed at increasing productivity in high potential production zones (benefiting from public irrigation systems). The WB has played a uniquerole among the donors community inMadagascar, with the largestportfolio interms of commitments, and is seen as the lead GoMpartner for poverty reduction. 16. The Bank has a comparative advantage in fhnding this operation due to its active role in the support for reforms in the irrigation sector. Specifically, privatization o f public and parastatal irrigation organizations in the early 1 9 9 0 ~rationalization o f public expenditure for the maintenance, ~ transfer o f the management of irrigation schemes to W A Sand capacity buildinghave beensupported by past Bank investment operations. More recently, the Bank supported the Government in the 7 establishment o f the Fonds d'Entretien de Rbseaux Hydro Agricoles (FERHA, the Irrigation Maintenance Fund). 17. The Country Assistance Strategy (CAS) for Madagascar is designed to support the implementation o f Madagascar's Poverty Reduction Strategy Paper (PRSP), which has the objective to reduce poverty by half in ten years. To carry out this objective, the CAS lays out three key priorities: (i) improve governance; (ii) promote broad based growth; (iii) human security. The Bank i s provide currently preparing its new CAS which will support implementation o f the MAP. The new CAS will reflect the GoM's focus on, among others, infrastructure, rural development and environment to underpin growth and poverty reduction. A second phase of the proposed APL is mentioned in the investmentprogramofthe CAS. 18. Madagascar is eligible for GEF support. It ratifiedthe UnitedNations Convention to Combat Desertification (UNCCD) in 1997, the Convention on Biological Diversity in 1996, the United Nations Framework Convention on Climate Change in 1999, and i s a contractingparty to the Ramsar Convention on Wetlands since 1999. GoM has also prepared and submitted a UNCCD National Action Plan in2001. Higher level objectives to which the project i s contributing 19. The proposed project constitutes a key element o f the Bank's strategy inMadagascar, and will contribute to achieving the priority objectives o f GoM and WB through the promotion of broad-based economic growth, the second key priority of the PRSP. In recognition o f the fact that growth in Madagascar will be derived from the country's unique natural resources and from the transformation of its natural products, and in accordance with the vision outlined in the MAP, the project would contribute to developing a diversified andrichnatural resourcebase that will contribute to the creation o f products with highvalue added. More specifically, the project aims to turn around a vicious cycle o f low productivity, deferred maintenance and poor water management into a virtuous cycle o f increased productivity, full cost recovery and acceptable O&M by clarifying responsibilities, providing an adequate incentive framework, encouraging accountability and promoting agricultural intensification and commercial agriculture. It would thus contribute to creating favorable conditions for accelerated agricultural and rural growth in a number o f clearly identified high potential rural growth poles. 20. The proposed project i s consistent with the GEF Operational Program 15, concerning the mitigation and prevention o f land degradation. The project will promote sustainable landmanagement across the watersheds that create long-term global environmental benefits within the context o f agricultural development, ecosystem services creation andpreservation, protection o f primaryhabitats, as well as rural livelihood improvement. Although the project will touch upon all four strategic objectives, the most relevant contribution will be provided to Strategic Objective 2: Demonstration and scaling-up successful Sustainable Land Management (SLM) practices for the control and prevention of desertiJication and deforestation. The project will promote best practices in agriculture based on agro-ecological principles, support improved and fire-less pasture management, favor the revegetation o f the watersheds through natural regeneration, reforestation and agroforestry, and support incentive systems (e.g. land tenure regulations) that are in favor o f natural habitat protection, especially the remaining rainforests in the upper watersheds. This will be achieved through the technical assistance, capacity building and providing access to investment hnds in Component 1 (Agricultural Development) for on-farm development and adaptation of agro-ecological cropping practices. In addition, participatory watershed management planningwill be supported in Component 3, that creates larger-scale and environmental awareness and provides the local stakeholders with the means to design an action plan that will be executed locally, and contributes simultaneously to the overall objective o f watershed protection and the creation of global environmental benefit. These include prevention o f natural habitat loss (including forests or wetlands), biodiversity conservation, reduction o f carbon emissions from wide spread fire use, and the increase o f carbon sequestration through increased vegetation coverage. SLM farmer groups will be provided training and given the 8 opportunity to access investments for innovative community-based SLM initiatives. Through this engagement, GEF will strengthen GoM's National Irrigation and Watershed Management Program. It will directly contribute to the implementation o f the UNCCD NAP (National Action Plan) and to most of its specific objectives, including (i)the promotion o f sustainable natural resources management, particularly forests, land and water, (ii) the promotion and adoption of improved farming techniques that respect the environment, (iii)the provision of support for a more profitable and sustainable management o f pastures and associated livestock, and (iv) the support o f an enabling environment of regulatory and financial incentives for local communities and the private sector to engage in activities that contribute to NAP implementation. The project will also directly contribute to the execution o f NEPAD's EAP (Action Plan for the Environment Initiative) and CAADP (Comprehensive Africa Agriculture Development Program). B. PROJECTDESCRIPTION LendingInstrument 21. The proposed lending instrument for this operation would be a three phase, 12 year APL (FY07-FY19). Only the first phase of the proposed project (FY07-FY11) i s presented here. An APL would provide the GoM with the necessary flexibility to implement the program in accordance with preferences and capacities o f users groups. It would also lay adequate foundations for scaling up o f the project's activities on the basis o f lessonslearned from earlier phases. 22. This operation (APL-1) aims to assist the GoMto implement innovative approaches in support o f sustainable investments in agricultural productivity in both irrigated and rainfed areas. The content o f the subsequent phases o f the APL are not yet known at this stage, and will be determined by the lessons learned from the experience during the first phase. The three phases provide the Bank with the possibility to support the long-term national Irrigation and Watershed Management program, while at the same time providing incentives for achieving the program's development objectives. ProgramObjectivesand Phases 23. Government's National Irrigation and Watershed Management Program (PN/BV-PI). The Government's National Irrigation and Watershed Management Program (PNIBV-PI) is a central part of the MAP and Government strategy for the development of agriculture. The global objective o f the PN/BV-PI program, as formulated in the Irrigation and Watershed Management Policy Letter of the Government is to sustainably improve the living conditions and incomes of rural populations in irrigation schemes and their surrounding watershedsand the management of natural resources. 24. The PN/BV-PI covers all medium- and large-scale irrigation schemes inthe country, and will include both newly prepared (including the proposed Irrigation and Watershed Management project) as well as on-going operations that will gradually be retro-fitted into the national program and its institutional framework. It is estimatedthat over the next 16 years, it will cover about 970,000 hectares o f irrigation schemes and their associated watersheds, at a total cost of about US$937.0 million. The PN/BV-PI will be supported by all interested donors. The French Development Agency (AFD), the African Development Bank (AfDB), the European Union, USAID, the Food and Agriculture Organization (FAO), AFD, International Fundfor Agricultural Development (IFAD), Kreditanstalt fir Wiederaufbau (KfW), Organization o f the Petroleum Exporting Countries (OPEC), the Japanese International Development Agency (JICA) and a number o f NGOs are all operating in the irrigation sector and/or the National Irrigation and Watershed Management Program (annex 2 presentsa detailed list o f current and expected donors). 25. The PN/BV-PI i s underpinned by the Irrigation and Watershed Management Policy Letter (available in the project file) that describes the objectives o f the Government with respect to the development o f irrigation schemes and surrounding watersheds. The Policy Letter also indicates the 9 approaches needed to achieve the objectives. The Policy Letter clearly defines its medium-term development vision regarding Irrigationand Watershed Management: (i)clear responsibilities for each of the actors in the management o f irrigation schemes and surrounding watersheds (farmers, water users, populations and their associations, Communes and Inter-communities, Regions, and central Government); (ii)effective participation o f the population in problem diagnosis and options identification; (iii)co-management o f irrigation schemes and watersheds by all the actors concerned; and (iv) adequate incentive systems and efficient mechanisms to ensure that all respect their commitment. 26. Proposed IDA/GEF APL-funded program (the Program): objectives and phases. The IDNGEF-funded Program will support the implementation of the Government's PN/BV-PI in six o f the country's main irrigation zones and associated watersheds (six "sites"). Four o f these have been included in the project (APL-1): Lac Alaotra (Sahamaloto), Marovoay, Andapa and Itasy. The sites that will be added for inclusion inthe Program's third phase will be selectedon the basis o f experience o f the first phase. Overall, the six sites will require the rehabilitation o f about 66,000 hectares o f irrigated perimeters and the managemento f about 200,000 hectares o f watersheds at an estimated cost o f about US$126.0 million (including US$18 million GEF and a 20 percent contribution from stakeholders). 27. The program sites have very different ecological environments, farming systems and social structure/ institutions. Thus: (i)the design of the Program's activities will need to be flexible to take into consideration the specificities o f each site; and (ii)Program outputs and outcomes will need to benefit from an efficient monitoring and evaluation system for assessing progress and impact and taking corrective action if necessary. Finally, the Program will be (i) fully integrated into Government's overall development strategy; and (ii)tightly coordinated with other national programs in infrastructure, water management, environment, or local development. 28. The objective of the Program i s to sustainably improve the living conditions and incomes of rural populations in six main irrigation sites and their surrounding watersheds, and the management of natural resources. This i s the same as the Government's PN/BV-PI but applied to six sites. Its global environmental objective is to improve the environmental sustainability of land management practices infour targeted watersheds. 29. The APL Program will be implemented over a period o f 12 years, through three phases which will combine both the gradual implementation and consolidation o f reform activities on a given site (each phase building on the achievements/lessons o f the previous phase and initiated only upon completion o f specific and monitorable triggers) and the geographical expansion to new sites. The proposed project is the first phase of this Program. 30. The focus o f GEF support under the Program will be to promote the sustainable development o f the watersheds' resource base through an integrated watershed management (WSM) approach with innovative, long-term approaches to deal with complex natural resource management issues (such as fire use, deforestation, and unsustainable farming practices). With that, GEF will support the development goals of local communities and secure global environmental benefits. GEF will also emphasize capacity strengthening in sustainable land management, and identify successful processes and outcomes and disseminate lessons learned in order to strengthen the National Program and facilitate its scaling up. 31. Phase 1 (2007-2011) will lay the foundations for scaling up investments in subsequent phases, will put in place the institutions and will improve the infrastructure that are required for the implementation o f the Program. The proposed IDAiGEF funding envelope for the first phase is US$36 million and would cover about 21,780 hectares o f irrigated and about 7,700 hectares o f rainfed agriculture. 10 32. It is envisagedthat phase 2 will consolidate the achievements of phase 1 inthe four sites and lay the foundations for scaling up in phase three to additional sites through detailed assessments and stakeholder participation. Consolidation implies the establishment of a proper functioning supportive institutional framework as described in the Policy Letter, the satisfactory management o f irrigation schemes and watersheds by sustainable land management groups, and satisfactory project management. The IDMGEF funding envelope for phase 2 i s estimated at US$31 million, and will cover about 20,000 hectares o f irrigated and about 7,000 hectares o frainfed agriculture. 33. Phase 3 will, on the basis o f methodologies and tools that have been developed during the first and second phase o f the project, scale up investments intwo additional sites. The sites will be selected during the second phase on the basis of experience and taking into consideration the characteristics o f the various main irrigation areas. Phase 3 will have an IDA envelope of US$41 million and will cover an additional 25,000 hectares. 34. Annex 3 shows both the Program's expected final impact and the milestones and outcomes which will be usedto monitor development and implementation progress. Project DevelopmentObjectivesand Key Indicators. 35. Inthe context of the broader APL objectives described above, the development objective of the Project (the first phase of the Program) is to establish a viable basisfor irrigated agriculture and natural resources management in four main irrigation sites and their surrounding watersheds: (i) Andapa (Sava Region), (ii)Marovoay (Boeny Region), (iii)Itasy Region, and (iv) Lac Alaotra (Alaotra Mangoro Region). A detailed description of the project zones i s included in Annex 1 and Annex 16. 36. The global environmental objective of the project i s to improve the environmental sustainability o f land management practices in four targeted watersheds. 37. Triggers. Triggers for moving to the second phase o f the APL include attainment o f the following targets: Watershed Master Plans (WMP, including Scheme Development Plans (SDP) and Watershed Development Plans (WDP)) and associatedPerformance Contracts executed satisfactorily'; an acceptable institutional mechanism for the funding o f non-transferable irrigation infrastructure (FERHA) establishedand operational; private sector investments in agriculture increased as evidenced by disbursements under the matching grant mechanism; Agricultural Service Centers (ASCs) establishedand operational inthe four project sites; guichetsfonciers established and operational inthe four project sites. Specific information on what is implied inattaining the triggers i s presentedinAnnex 18. 38. The project will be implemented in four rural `growth poles' - four zones characterized by medium- and large-scale public irrigation where a number of conditions have been met for a rapid kick-off o f growth, including relatively easy access by road, and better access to finance, inputs, markets and equipment. A more reliable access to water puts a high premium on the use o f productivity enhancing inputs, provides more flexibility, diversity, reliability, quality and product uniformity to satisfy the requirements of markets, and enables farmers to capture higher seasonal prices. In addition, the sites are similar in the sense that institutional issues such as a clarification o f roles and responsibilities through irrigation management transfer represents a high priority for improving performance o f irrigated agriculture. WatershedMaster Plans, Scheme DevelopmentPlans, WatershedDevelopmentPlans and Performance Contracts are discussedin annex 4. 11 39. The first phase will cover about 21,780 hectares o f irrigation schemes (out o f a total o f 66,000 hectares for the sites to be included inthe IDA-hnded APL in six sites). Direct beneficiaries include about 30,000 smallholder households producing irrigated and rainfed crops, and farmers' groups and private operators providing services, selling products and performing various functions in the value chain. 40. The four sites meet a number o f conditions for rapid growth, including relatively easy access to markets for outputs and inputs and good agricultural development potential. However, they are also suffering from severe institutional weaknesses for the management o f the irrigation perimeters and significant upstream watershed degradation. Correcting these weaknesses in support o f fast and sustainable development will be one o f the main objectives o f the project, thus buildinga strong base for the subsequent phases o f the Program. The Project would focus on: 0 Establishing the policy and institutional framework adopted for the PN/BV-PI and reflected in the Policy Letter: (i) the respective roles and responsibilities o f national stakeholders clarify (central and decentralized government, Region, communes, Federations o f W A S ((F)WUAs), local communities and private sector) inthe management o f irrigation schemes and the natural resource base o f upper watersheds; (ii)build the capacities (technical, managerial, financial) o f these stakeholders; and (iii) establish the right incentive framework for efficient operations and sustainability; Supporting rapid agricultural growth by (i) providing farmers with improved access to markets and agricultural services; (ii)undertaking rehabilitation and improving management o f irrigation schemes; and (iii) promoting productive investments inparticular in diversification; and; 0 Supporting the equitable development o f the upper watershed resource base through prevention o f severe erosion, the promotion o f suitable farming systems and agricultural practices and the conservation o f fragile ecosystems. 41. Finally, to ensure their success, the rehabilitation o f irrigation schemes and the development activities in upper watersheds will be undertaken only when they satisfy specific eligibility criteria. These criteria are presented inthe detailed description o fproject components below. 42. Project total cost i s estimated at US$40.4 million, to be financed by: (i) US$30.0 million IDA: (74 percent o f total cost); (ii)GEF: US$6.0 million(15 percent); andbeneficiaries: US4.4million (11 percent). The expected project results include (i) dissemination o f innovative technologies and equipment to 30,000 beneficiaries through extension, capacity strengthening and targeted cost sharing, (ii)improved management o f about 21,780 hectares o f irrigation infrastructure through investments in rehabilitation, training and institutional reforms, (iii) percent increase o f landarea under sustainable 20 landmanagement and 15 percent improvedvegetation cover as a percentage o f the baseline intargeted watershed areas (iv) improved management o f about eight sub-watersheds through capacity strengthening and investment in watershed infrastructure, and (v) increased government support for agricultural intensificationinirrigatedandrainfed areas through increased.public expenditures. Project Components 43. The proposed project comprises three technical components covering major strategic orientations: (i)Development o f Commercial Agriculture, (ii)Irrigation Development and (iii) Watershed Development. The fourth component i s Program Management. In accordance with the `growth poles' approach, the project proposes similar investments in the four regions concerned (Annex 1). A more detailed description o f the components and activities is attached in Annex 4. Information on GEF hnded activities within the components can be found in the Incremental Cost Analysis (Annex 15). 12 44. The project concept is based on the following principles: (i) refocusing of government intervention on its core mandate; (ii)a market-driven approach to agricultural supply chain development, (iii)contractualization of partners and service delivery (iv) a participatory approach based on stakeholder-demand (v) clear roles and responsibilities; and (vi) an appropriate policy framework for agriculturalintensification. 45. The project i s expected to be completedover a four-year period, with a closing date on March 1,2011. Component 1:Development of CommercialAgriculture (US$12.68 million, including an IDA Contribution of US$7.45 million, a GEF contribution of US$2.72 million, and a beneficiaries contribution of US$2.51million) 46. The objective for this component i s to lay thefoundations for improved market access and sustainable intensification and diversification of irrigated and rainfed agricultural systems in the project's watersheds. 47. The `Development of Commercial Agriculture' component includes the project area as a whole: both irrigated and upland or tanety areas. Its specific objective will be achieved through an approach focused on market-driven demand, agricultural technology development and dissemination, private sector initiative and vertical integration of supply chains, as well as promotion of partnerships among stakeholders (including public-private partnerships, PPP). 48. The component aims at improving, all alongthe targeted supply chains: Access to market and marketing systems in order to reduce costs and increase farm gate prices; 0 Addedvalue through diversification into higher addedvalueproductsand ago-processing; 0 Capacities of farmers, farmers groups andprofessionalorganizations; 0 Agricultural productivity throughbetter access to extension, improvedtechnology, inputs, and credit. 49. The component includes two sub components: one involving activities that largely depend on public/collective initiative; the other one depending essentiallyondemandfrom stakeholders: (9 Support to agricultural services. The sub-component aims at improving access to markets and supports the development of commercial agriculturevalue chains, through innovative technologies for production, storage and processing, and a stronger enabling environment at the site level. The project will find services, work, equipment, training and operational costs. Activities will be adjusted to specific needs of each site, and would include the following: (a) support to the development of dynamic market-drivensupply chains, particularly by creating and strengthening links betweenproducers and markets, (b) building up of farmers capacities and strengthening professional organizations, as well as establishing Agricultural Service Centers (ASC), and (c) dissemination of technologies for agricultural intensification and diversification, including support and advisory services for the implementation of agro- ecological and agroforestry techniques in the upper parts of the watersheds. These services would beprovidedby strategic partners and specialized serviceproviders. (ii) Support to private investment. This sub-component will provide demand-based support to private investment by operators, farmers and farmer groups at all levels of the agricultural activity. The sub-projects funded under this sub-component would be essentially private in nature and would be initiated upon request by a farmer, a farmer group or a private sector operator, with financial support from the project if government considered them a priority and wanted to promote them. Project support would be provided to priority new investments 13 through a cost sharingmechanism according to a pre-established positivehegative list. Private operators would be responsible for implementing the sub-projects and related activities according to procedures approved by the project. Sub-projects considered could include investment incollective storage, market research and supply chain development, technical and managerial advisory services, new technology demonstration and dissemination (including ago-ecological cultivation techniques), support to seed production, private distribution networks for inputs and equipment and microfinance institutions, and support to contract farming and integrated sub-projects initiated by commercial or agro-industrial partners and involving small scale producers. The project would take a gender sensitive approach and would specifically support vulnerable groups in their demands. In addition to investment in infrastructure and equipment, sub-projects could include studies andmarket tests andresearch, extension and advisory services, appliedresearch, training, and study tours. 50. The financing modalities are described in further detail in Annex 4. Implementation responsibilities are detailed in Annex 6. Eligibility criteria for activities funded under this component include the willingnessto cover part o f the associated costs and to commit to develop and implement a capacity strengthening plan. Component 2: Irrigation Development (US$l7.47 million total, IDAfunding US$15.67million, and a beneficiaries' contribution of US$1.80 million) 51. The objective o f this component is to lay the foundations for improved management, maintenance and sustainability of irrigation services provision infour large-scale irrigation schemes through rehabilitation of irrigation infrastructure, capacity strengthening of stakeholders and clarflcation of roles and responsibilities, and establishmentof an appropriate incentiveframework. 52. The component will contribute to the project's development objective by investing in irrigation infrastructure, improving the quality o f irrigation services and O&M, thus puttinginplace a more favorable environment for agricultural intensification and diversification. The project will adopt a contractual approach that empowers stakeholders and clarifies their respective roles, based on the principle that investments in infrastructures enhance and at the same are conditioned by the performance o f stakeholders. T o that end, Irrigation Management Transfer agreements will be signed between DRDR and WUAs that specify mutual rights and responsibilities. Investments will be allocated competitively among the four sites, providing more resources to sites where targets are more ambitious and where key indicators are being met. The instrument for clarifying and formalizing commitments and responsibilities is the Performance Contract signed between the (F)WUAs, the Communes and Regions, and the Ministry o f Agriculture, Livestock and Fisheries (MAEP). Achievement o f all performance indicators will pave the way for a subsequent phase o f the Performance Contract. 53. Specifically, the project will finance the following sub-components: (0 Support to Irrigation Development. As part o f an overall W,participatory preparation o f a Scheme Development Plan (SDP) and an annual Performance Contract (PC), negotiated between (F)WUAs, the Communes and Regions, and Direction GCnCrale de DCveloppement Rural (DRDR) .The SDP and PC will provide the overall framework for support to irrigated agriculture, including possible investments in the rehabilitation o f irrigation infrastructure. The project will also support stakeholders during implementation o f the PC, through capacity strengthening, mobilization o f water users, annual evaluation o f performance indicators and user satisfaction surveys. Studies will be conducted, among others into O&M costs and hurricane damage to irrigation infrastructure; (ii) IrrigationInvestments.Rehabilitationofirrigationandappurtenantinfrastructure, including technical design studies, implementation o f works, and their supervision. As many as possible o f these contracts will be co-signed by (F)WUAs. Investments will be determined in a 14 competitive way between the sites, with those sites performing better (in terms o f O&M cost recovery and DRDR expenditures for irrigation) benefiting from higher investment levels. The project will also promote low-cost individual irrigation technologies. 54. Eligibility criteria for support under this component include pre-project recovery levels for overall O&M costs, existence and functioning o f a WUA, current government (including DRDR and Communes/Districts) expenditure levels insupport of irrigation. Component3: WatershedDevelopment (US$4.33 million, including IDAfunding of US$1.82 million; GEF contribution of US$2.42 million, and benejciaries contribution of US$O.09 million) 55. The objective o f the component i s to lay the foundations for sustainable management of watersheds including irrigated and rainfed agriculture, the conservation of the natural heritage, and improved productivity of the natural resources. 56. The component will adopt an integrated and participatory approach to watershed management to make ruralpopulations more accountable and encouragethemto manage land and natural resources on a more sustainable manner. Thus, the component would contribute to: (i) protect watersheds by reducing erosion and sedimentation; (ii) increase the productivity and sustainability o f agricultural production based on agroecological and agroforestry technologies; and (iii)strengthen the management o f natural resources to improve the environment and living conditions. The component would concentrate on investments with long-term environmental impacts, and support to SLMgroups. The project will finance the following sub-components: (i) Planning and capacity building for sustainable management of watersheds, including (a) preparation, as part of Watershed Master Plans (WMP), of Watershed Development Plans (WDP) in the four project areas; (this will be financed by a PPF o f an additional 0.77 million US$) (b) preparation o f participatory plans for managing approximately eight sub-watersheds (each o f about 10-500 kmz);(c) support to communication and negotiation platforms, (d) training and capacity strengthening o f SLM groups; and (e) support to improvement o f land tenure security; (ii) Sustainable investments in watersheds, including (a) determining, through participatory negotiations, local strategies for controlling erosion, arresting gullies and reducing the sediment load o f river runoff. The project will finance investments in strategic anti-erosion works (through, among others, biological methods and technologies); and (b) interventions, through matching grants, on communally owned land to improve plant cover, reforestation and pasturesthrough strengthenedtechnologies andmanagementtransfer of naturalresources. 57. Eligibility criteria for support under this component include the severity of land degradation, and the willingness of stakeholders to cover part of the associated investmentcosts. 58. An MOU, one per Project Area, will be signed by MAEP through the National Irrigation and Watershed Program and MinEnvEF through Third Environment Program (EP3) to ensure adequate integration, inall project areas, o f the project and EP3 financed under Credit from the AssociationNo. 4362. The M O Uwill specify indetail the activities that will be financed by eachprogram. Component4: Program Management. (UB4.31 million, including IDA finding of US$3.45 million and a GEF contribution US$0.86 million) 59. The objective o f this component i s to manage and use resources in accordance with the project's objectives and procedures, and to put in place a policy framework that is favorable to upscaling of the project at the national level. This component will finance the following sub- components: 15 (i) Management of the project, including (a) provision of technical assistance, training, office equipment and vehicles, minor ofice upgrading works, auditing and evaluation studies, and incremental operating costs in support o f project management; (b) overall project planning, quality oversight, procurement, financial management, and monitoring o f project activities; and (c) outsourcing o f quality oversight through independent financial and technical audits, and evaluation o f project activities. Project management will encompass all four target watersheds as well as national level coordination; (ii) Support to national policies, including (a) provision o f technical assistance, studies, training, information campaigns, exchange visits and workshops for the development o f major national policies, regulations, and plans considered critical to the Government's National Irrigation and Watershed Management Program; and (b) provision o f support to emerging professional groups, in particular the Platforme Consultative de Riz and the Association Malgache de Producteurs de Semences, (iii) Monitoring and evaluation. Data collection and reporting on key performance output and impact indicators, including targeted data collection, surveys, participatory assessments and mid-term and final evaluations. 60. The scope o f this sub-component would be national. The improved policies are expected to benefit all key operators andproducers involved inthe sub-sector. LessonsLearnedand Integratedintothe ProjectConcept 61. The design o f the project is based on lessons drawn from evaluations3 of programs and projects inthe irrigation sub-sector that were often unsuccessful. Despite significant investments inthe rehabilitation o f irrigation infrastructure, there has been little diversification to higher value added crops, and sustainability has been questionable because o f lack o f maintenance. 62. Some o f the reasons for the failure identified by the different studies are lack o f market access (remoteness leading to high transport costs); lack o f access to extension services and input supply; failure to take upstream watersheds into account; unclear responsibilities; weak stakeholder capacity; land tenure constraints; non-respect o f commitments by users and government; and indiscipline and impunity. 63. The conditions o f success identified by these same studies include the following: (i) An integrated approach that contributes to increased productivity and incomes in irrigation schemes and surrounding watersheds, safeguards natural resources in watersheds, improves the provision o f agricultural extension and inputs, and actively supports emergence o f a private sector; (ii) A conducive economic environment including a price policy for products and inputs, market access in terms o f road infrastructure and information; promotion o f the private and associative sectors for marketing o f products and supply o f inputs; access to appropriate and efficient agricultural services; and access to rural finance; (iii) An unambiguous institutional framework with clear responsibilities in accordance with policies such as decentralization and legislation (land, water and forestry codes) for farmers and their associations; communes, inter-communes and regions; decentralized government services; and agencies and private operators; This comprises, among others (i) Madagascar - RuralandEnvironmentSector Review (WB, 2003), (ii) Watershed ManagementOperations: Approaches, Challenges andEmergingLessons (WB/ARD, 2006), (iii) Madagascar: The Impactof Public Spending on IrrigatedProductivity, 1985-2004 (WB, 2004), (iv) ICR PPI-2(WB, 2000), (v) Agriculture, Pauvrete Rurale et Politiques Economiques a Madagascar (Mintenet al, 2003), and (vi) Review of Madagascar's Rice sub-sector (Bockel, 2002). 16 (iv) An approach that emphasizes capacity strengthening of all stakeholders to help them play their respective roles andresponsibilities; (v) A participatory approach, coordinated decisions and respect for commitments, including stakeholders with established and acknowledged rights and obligations, adequate resources and capacities, who filly participate in decision-making; incentives and mechanisms in place to encourage appropriate behavior and respect for commitments made; and interfaces for cooperation and dialogue inaccordancewith decentralization policies. Alternatives consideredand Reasonsfor Rejection A numbero falternatives were consideredand rejected inproject design: Develop the program as a Sectoral Investment Loan (SIL). However, it was felt that the investment part of the project requires a flexible implementation mechanism with an appropriate incentive framework that can respond to different preferences and capacities o f stakeholders; Splitting the project into three separate projects - (a) an agricultural productivity project focusing on irrigation and agricultural services; (b) a community based natural resource management project focusing on watershed management and decentralization; and (c) a land reform project focusing on implementation o f the recent economic and sector work findings. However, it was felt that this design would fail to capture evident synergies and create implementationgaps; Putting in place a sector-wide multi-donor approach similar to the EP3. However, discussion with other donors suggestedthat more flexible donor collaboration, possibly inpreparation for close collaborationthereafter, was more appropriate; Expandedfocus on complementary rural development activities like ruralJinance reform, and land administration. However, it was felt that this would exacerbate project complexity and create implementationrisks; Reduction in the geographic scope of the project to three areas. However, this would not minimize complexity and would be at odds with the government's scaling up objective; and; Designing the project to respond to the government's nascent decentralization program, transforming the project into a multi-sectoral demand-driven operation. While the project will be implemented through decentralized and "de-concentrated" authorities, it was felt that the policy, institutions and disbursement mechanisms associated with effective decentralization were not yet sufficiently clear and mature. C. IMPLEMENTATION PartnershipArrangements 65. The project is a partially blended operation between GEF and IDA. It contributes to the National Irrigation and Watershed Management Program, for which the GoM has prepared a policy letter (available in the project file). The National Program i s also supported by a large number o f other donors (see Annex 2). 66. In each of the four project areas, the project will work with regional partners. These may include PLAE in Marovoay, WWF in Andapa and Durell in Lac Alaotra for watershed activities; and BAMEX and CTHT/CTHA for marketingandbusinesspromotionactivities. 17 67. The project will benefit from the Memorandum o f Understanding signed between MAEP through the National Irrigation and Watershed Management Program and MinEnvEF through the multi-donor EP3 program to ensure coherence and synergies between activities inthe lower and upper watersheds. 68. Project Manuals, including a Project Implementation Manual, a Project Monitoring and Evaluation Manual and a Project Administrative, Financial and Accounting Manual have been prepared during project preparation. 69. The conceptual design of the ASCs and "guichets fanciers" has been elaborated in close collaboration with FAO, EU and AFD. The EU is expected to provide significant support to MAEP in the establishment andcapacity strengthening ofASCs. 70. Linkages to other GEF programs will also be established, through harmonized approaches and integration o f lessons learned, in particular with the Strategic Investment Program for Sustainable Land Management in Sub-Saharan Africa (SIP), with emphasis on mutual learning via regional knowledge sharing, with the World Bank supported Medium Sized Project Institutional Strengthening and Resource Mobilizationfor Mainstreaming Integrated Land and Water Management Approaches into Development Programs in Africa; the UNDP-GEF funded Stabilizing Rural Populations through theIdentijlcation of Systemsfor Sustainable Management and Local Governance of Lands in Southern Madagascar, and the UNEP regional project Addressing Land-based Activities in the WesternIndian Ocean. Institutionaland OperationalArrangements 71. The project will be implemented under the responsibility o f the Ministry o f Agriculture, Livestock and Fisheries. A national Project SteeringCommittee and Regional Monitoring Committees will be establishedat the nationallevel andineacho fthe four project areas. 72. The National Steering Committee will be chaired by the SG o f the Ministry o f Agriculture andwill includerepresentativesfrom: (i) Other central ministries involved at SG level - Ministry of Decentralization and Land Development (MDAT), Ministry of Environment, Water and Forests, Ministry o f Economy, Finance and Budget, Ministry o f National Education and Scientific Research, Ministry o f Industrialization, Trade and Private Sector Development - to ensure consistency o f project actions with national policies; (ii) the Chairperson ofthe PermanentSteering Teamofthe RuralDevelopment Action Plan; (iii) The mainprofessional organizations such as the Chamber o f Agriculture and associations/fora involved inthe mainvalue chains such as the ((Rice Platform)). 73. The National Steering Committee will be supported by a technical secretariat under the responsibility of the Director General for Regional Development at MAEP. It will be responsible for (i) programmingofprojectactivitiesandapprovaloftheworkplanandbudget,(ii) annual monitoring implementationand results, inparticular the analysis and approval o f activity reports and financial and operational audits, and (iii) recommending corrective measures that may be necessary. The National Steering Committee will meet twice a year. 74. Regional Monitoring Committees will be established in each of the four project areas. They will be chaired by the Head of the Region and made up of members of GTDR4. The Regional The Working Group for RuralDevelopment(Groupe de Travail de DBveloppementRural, GTDR) is made up of five local stakeholdergroups (farmers organizations,privatesector, decentralized authorities, NGOs, andprojects/ programs active in 18 Monitoring Committee will be supported by the GTDR's Technical Secretariat, and will be responsible for (i) ensuring consistency o f project actions with both national strategy and policy, and regional development priorities and programs; (ii) preparing and validating detailed work plans and budgets at the regional level; (iii) reviewing project progress andperformance, andthe implementation o f corrective measures ifnecessary. The Regional Monitoring Committee will meet twice a year. 75. The overall coordination o f the project will be ensured by the Directorate General for Rural Development (DGDR) at MAEP, as follows: 0 The Director General for Regional Development will be responsible for project coordination at national level; 0 The Regional Director for Rural Development (DRDR) will be responsible for project coordination and project investments intheir respective regions; 0 To help them in these tasks, the project will (i)finance recruitment o f an international technical assistant for operations at the national level, and (ii)four national technical assistants, as advisors to the Regional Director for Rural Development for operations at the site level; 0 The DGDR and DRDR will select from within their respective units one staff member who will provide support for coordinationandproject monitoring. 76. The project financial management will be the responsibility, at the national level, o f the Directorate o f Finance and Budget (DFB) from MAEP and, at regional level, o f the DRDR Department o f Finance and Budget (RDFB). The project will recruit a specialized national financial management and procurement agency that will provide technical financial management assistance to MAEP's Finance Director. The project will also recruit, for each o f the four sites, a regional financial manager, who will work closely with the DRDR and who will be in charge o f financial management. This person will work closely with MAEP's Department o f Administration and Finance and will report to the national financial management and procurement agency. 77. The project financial management will be strengthened by the following salient features: 0 The organizational structure in place defines the lines o f responsibilities and authority that already exist, and seems appropriate for planning, directing and controlling operations; 0 MAEP has an administrative manual o f procedures describing the internal control system applied withinthe Ministry; 0 MAEP also has an Internal Audit Department which can collaborate with the Financial General InspectiodState General Inspectionwhile carrying out internal audits. 78. MAEP's Directorate o f Finance and Budget will be responsible for: (i) consolidation o f work programs and budgets; (ii) maintenance o f records and accounts for all transactions made at the central level; (iii)timely preparation o f quarterly Financial Management Reports (FMRs), consolidated project financial statements and other required reports; and (iv) cash management and replenishment applications for the Designated Account. The Regional Department o f Finance and Budget at each o f the four sites will manage disbursements from the sub-Designated Accounts, maintain records and accounts for all transactions related to the regions, and prepare financial and other basic information onproject management`monitoring as required by the MAEP Financial Directorate. the region). Its activitiesinclude: (i) developing andupdatingregional developmentplans, (ii) updatingregionaldatabases; (iii)establishingregionaldevelopmentindicators andtheir monitoring; (iv) organizing meetingsfor exchanginginformation relatedto ruraldevelopment; (v) preparing and monitoringrural development programs/projectsinthe region. 19 79. To meet efficiently the challenges o f the project, the capacity o f the DFB/RDFB needs to be strengthened by: (i)improving the information system in place to ensure timely delivery o f information on project activities; and (ii) recruiting a specialized technical assistance agency to provide qualified and experienced accountants, and assist DFB/RDFB staff in performing financial management tasks including budgeting, accounting, financial reporting, and disbursement operations. 80. Procurement will be ensured, at central level, by the Person Responsible for Public Procurement (PRMP) o f MAEP and, at regional level, by relevant units o f the DRDR. The project will recruit (i)a national financial management and procurement agency that will provide technical assistance to the PRPM, and (ii) at the level o f each region, an additional staff, under contract to the national agency, who will be full time in charge o f project procurement. This staff will work closely with the PRMF' and will benefit from the project support in procurement technical assistance at the national level. 81. Technical assistance. Recruitment o f technical assistance (TA) will be done under two separate contracts (one for financial andprocurement management, and one for operational assistance) with specialized firms. The international "Operations" TA will be in charge o f (i) the DGDR advising and DRDRs on operational strategy, project implementation and monitoring; and (ii) training and operational support to MAEP staff involved in project implementation. Four national "Operations" TAs will be posted at the level of DRDRs to advise and support them inproject implementation and ensure coordination o f all project components at the regional level. National consultants in financial management and in procurement will be responsible for financial management and procurement and for providing technical support to DRDR staff. The four financial and four procurement consultants at region level will be responsible for financial management and procurement at the regional level. They will be recruited under one contract with the national level financial management and procurement specialist, and will report to the national specialists. Monitoringand Evaluationof Outcomes/Results 82. Monitoring and Evaluation (M&E) will be done by MAEP's Department o f Information Systems (DSI). A specialized project M&E system and procedures for data collection and reporting will be prepared to the satisfaction o f IDA. M&E will be based on direct reporting by institutions involved in project implementation, relevant data collected on a systematic basis for other purposes, participatory assessments, user satisfaction and income surveys, and targeted data collection (among others through satellite photos), as established in the project implementation manual. DSI will commission two evaluations o f project output and impact indicators, at mid-term and at completion. The project will establish a baseline before or soon after project effectiveness. The results framework i s presented in Annex 3. Performance indicators are linked directly to the CAS goal o f promoting broad-based social and economic growth. SustainabilityandReplicability 83. Sustainability o f project investments will be achieved inthe following manner: (0 Inlinking soil and water management in upstream watersheds to irrigation, the project will contribute both to more profitable rainfed agriculture, and more sustainable and cost-effective irrigation management. In so doing, the project will seek to set o f f a cycle o f increased productivity, higher income and improved capacity to pay for irrigation services. (ii) InviewoftheexperienceinMadagascar,prioritywillbegiventocapacityandinstitutional strengthening. The project will not establish new institutions, but will build on GoM's priorities and on what has already been established. Investments will only be done if conditions associated with institutional performance and governance have been met. (iii) Clientdemand,contributionincashorinkindandownershipwillbethedeterminingfactorin deciding to go ahead with investments inagriculture, irrigation andwatersheds. 20 (iv) Past experience provides abundant confirmation that irrigation schemes that depend on pumping are not sustainable. The project will therefore select irrigation sites that depend on gravity flow only. Technical alternatives will be designed or those areas o f the schemes that cannot be reachedthrough gravity flow will be abandoned. 84. Successful project outcomes and lessons learned can be disseminated through the National Program and replicated to other regions. The fact that the project will work in four distinct sites will allow for replication o f lessons learned within each region, taking into account local specificities and conditions. If successful, the project will also have a good potential for transferability to other countries in the Africa region. Dissemination o f good practices and successful approaches would be essential in facilitating the scaling-up process. A detailed replication strategy would be proposed after the mid-termevaluation of the project. Criticalrisks and PossibleControversialAspects 85. The potential risks o f the project are presented in the table below. A potential controversial issue i s the fact that the project intends to put in place incentives for sustainable irrigation management. This will require up-front buy-inand commitment from all levels, but inparticular from the local authorities. The (F)WUAwill be responsible for taking action against defaulters, which may include legal court action, closing down irrigation water supply, and imposing financial penalties. able 1:Critical risks and mitigationmeasure Risks Risk rating RiskMitigation Measures Operational Substantial The projectwill adopt acontractual approach Farmers are unwillingto pay for the full that confersmaximumresponsibilityto costs of irrigation services even though (F)WUAs andthat reviewsannualprogresson they have indicatedinthe Scheme keyperformance indicators DevelopmentPlanthat they prefer Work with the (F)WUAs to follow-up on fee rehabilitatingthe irrigationinfrastructure, payment; and have agreedto the associated O&M InvolveCommunes and Regionsin managing costs. irrigation schemesto help (F)WUAs put in placean appropriateincentiveframework; Improvelinkages insupply andmarketing chains to increase farmers' incomeand improve their capacityto pay. Cyclone damageunderminessustainability Substantial The projectwill help operationalizethe of the irrigation schemes. Irrigation Maintenance Fundestablishedby MAEP, to undertakecyclone damage repair works. The projectwill also helpdevelop climate-proof designs for irrigationrepairworks. Policy Moderate The implementation ofthe GoM fertilizer and GoM doesnot follow a sound seed and seedpolicy is a covenantunder the project. fertilizer policy based on privateproviders, as well a favorable environment for private agrobusiness development. 21 Inherentrisks: Country Level. Substantial These issues arebeingaddressedthrough the Delays in the productionofpublic ongoingPFMreformssupportedby IDA (through accounts, and audit maynot be conducted the Governanceand InstitutionalDevelopment in compliance with internationalauditing Project) and other donors. Inthe meantime, the audi standards due to: weak capacity of the ofthe WB-financedprojectswill be carriedout by accountingprofessionin Madagascar, and; internationalaccountingfirms or by international ii)inadequatenumberofskilledand accountingfirms associatedwith local auditing experiencedauditors at the "Chambre des firms, with effective participationofthe former comptes" in particular. (internationalaccounting firm) in the fieldwork. Entity Level Substantial Developmentpartnerswill continueto support the Accounting, financialreportingand GoMpriority actionplan for public finance reforms auditingnot timely. includingtrainingand capacity buildingof key line ministries(includingMAEP) inthe area of public financial management.This trainingwill include planning, budgeting, accounting, reportingas well a monitoringand evaluation. Recruitment of an internationalauditingfirm acceptable to IDA to carry out the annualaudit of the project financial statements duringthe first phas ofthe Project. ProjectLevel Substantial Recruitment of aFinancialManagementAgency The MAEP FinancialDirectorate(DFB) (FMA) inconformitywith Bankproceduresto: i) andthe three regionalspendingunits providethe DFB and RDFB with qualifiedand (RDFB) have noexperiencewith experiencedaccountants and; ii)assist the implementingan IDA- financed project. accountingstaff of these Directoratesinperforming FMtasks. Controlrisks: Accounting/ Internal Controls Substantial Implementationof an accounting system acceptable Projecttransactions maynot beproperly to IDA, including: accounted due to absence of a Chart of accounts. 1 Preparationof aChart of accounts. 1 Elaborationof an accountingmanual of Policies and control proceduresmaynot proceduresacceptable to IDA to provideclear be applied efficiently due to absence of . guidanceto staff. appropriatedocumentationcovering Users trainingto ensure constant applicationof accounting, managementof finances and policies/procedures. financial reporting. Financialreporting Substantial Recruitmentof a consultantto designand implemen Riskof delays inthe productionand 3 computerizedsystem to allow for timely submissionof financial reports required xoduction ofreliablefinancial statementsand for managingandmonitoringproject FMRs. activities. 3rganizationof atraining sessionto ensure efficient Jse o f the new computerizedsystem by the project - - iccountingstaff. Auditing Substantial iecruitment of an internationalauditingfirm The quality ofthe audit maynot be icceptableto IDA to carry out the annual audit of acceptable andthe report not deliveredin heprojectfinancialstatements duringthe first phasc time due to weak capacity ofthe IftheProject. accountingprofession. 22 Conditions 86. Effectiveness Conditions: e adopted the Project Implementation Manual, the Project Monitoring and Evaluation Manual and the Project Administrative, Financial and Accounting Manual, all inform and substance satisfactory to the Association; e established a Financial Management System (FMS), in form and substance satisfactory to the Association, to ensure proper execution and monitoring of Project activities; It is understood that this includes recruitment of financial managementTA; e appointed the independent auditors; e preparedthe tenderhidding documents to carry out the procurement related to the preparation of the Watershed and Irrigation Management Schemes programmed for the first Project year; and e establishedthe NSC and RMCs inform andwith functions satisfactory to the Association. 87. Other Special Covenants e The Government will adopt, by December 31, 2006, and thereafter implement, a national strategy on fertilizers; e The Government will issue, by December 31, 2007, and thereafter implement, legislative guidelines for the application o fthe new seeds legislation; e abaseline survey for the Project be completed by December 31,2006; e the Watershed Master Plans (including Watershed Development Plans and Scheme Development Plans), one per eachProject Area, be completed by September 30,2007; e operational modalities of managementand replenishment o f FERHAbe definedby September 30,2007; e the irrigation related legal framework, including but not limited to Law no.90-016 and associated legal guidelines, will be harmonized with the Irrigation and Watershed Management Policy Letter by December 31, 2007. Such harmonizationto include, inter alia: the management of autonomous irrigation schemes and partner schemes, the process for irrigation managementtransfer and the membershipo fthe W A S . Disbursement Condition 88. No disbursement shall be made for any Irrigation Rehabilitation Work and their supervision under component 2, unless the relevant SDP and PC, and as the case may be, Transfer Agreement, has been finalized and signed, respectively, between the relevant Implementing Institution andthe WUA, interms anconditions satisfactory to theAssociation. 23 D. PROJECTBRIEFSUMMARY Economic and Financial analysis Summary of Benefitsand Costs: 89. Project benejts. For each watershed, two types o f benefits were identified: (i)additional agricultural production inirrigatedperimeters anduplandsor tanety areas and (ii) reduced siltation and avoided cyclone damages to irrigation infrastructure. The benefits were quantified and valued using hypotheses on (i)delay and increment in generating additional agricultural production in irrigated areas (mainly paddy) and uplands (mainly cassava, maize, and tomatoes) and (ii) delay and increment in reducing siltation and damages, and (iii)rent values associated with increased productivity and reduced O&M costs. The results are presented in the table below. The gross benefit value o f the project is US$62 million. 90. Economic Analysis. For the purpose o f the economic analysis, the irrigation and watershed management project was divided into four watersheds, each of which was assessed separately: Marovay, Itasy, Andapa and Lake Alaotra. For each watershed, economic costs are grouped into three categories: (i) investment costs related to commercial agricultural development (the public aspects), irrigatedperimeters, watershed development, andproject management; (ii) physical contingencies; and (iii)incremental recurrent costs. The results are presentedinthe table below. Table 2: EconomicCostsper Watershed Typeof Costs (PV, $Thousand) Marovoay Ztasy Andapa Lac Total Alaotra Commercial Agricultural Development $2,308 $2,3 12 $2,312 $2,3 12 $8,910 IrrigatedPerimeters $3,390 $2,815 $2,138 $5,799 $14,142 Watershed Development $1,171 $1,018 $980 $903 $4,072 Project Management $836 $882 $882 $882 $3,482 Physicalcontingencies $323 $313 $250 $488 $1,374 Total $8,027 $7,340 $6,563 $10,385 $31,979 Recurrent Costs (included) $1,119 $703 $692 $1,887 $4,401 91. For each watershed, two types o f benefits are identified: (i) additional agricultural production in irrigated perimeters and uplands or tanety areas; and (ii) reduced siltation and avoided cyclone damages to irrigation infrastructure. The benefits were quantified using hypotheses on (i) delay and increment in generating additional agricultural production in irrigated areas (mainly paddy, but also off-season crops), and uplands (mainly cassava, maize, and tomatoes); (ii)delay and increment in reducing siltation and damages; and (iii)rent values associated with increased productivity and reduced operation and management (O&M) costs. Results are presented in Table 3, below. The estimated value o f the project's gross benefits is US40.7 million. At 73 percent, the contribution o f the additional irrigated areas to agricultural production (in well-, partially- and off-season-irrigated areas) is the larger project benefit. Table 3: Economicbenefits, NPV andERRper Watershed WatershedBenefits (PV, $Thousand) Marovoay Ztasy Andapa Lac Total Alaotra Well-IrrigatedAreas Production $3,968 $2,391 $2,235 $7,382 $15,977 Partially-IrrigatedAreas Production $1,207 $1,810 $754 $201 $3,973 Off-seasonirrigation $428 $6,422 $2,676 $535 $10,061 Tanety Production $3,546 $2,828 $2,246 $499 $9,118 SiltationReductioninIrrigation systems $170 $204 $127 $153 $653 24 Avoided Cyclone Damagesin Irrigationsystems $282 $263 $170 $298 $1,013 Total Project Benefit $9,602 $13,918 $8,208 $9,068 $40,795 Total Project Cost (investment andrecurrent) $8,027 $7,340 $6,563 $10,385 $31,979 ProjectNet PresentValue (NPV 10%) $1,574 $6,578 $1,645 41,317 $8,816 Economic Rate of Return (ERR) 13% 20% 13% 8% 14% 92. The calculations o f the Net Present Values (NPV) and Economic Rates o f Return (ERR) for each watershed (show that Itasy i s the most economically viable watershed for the project, with an ERRo f20 percent (see Table 3, above). The ERRSo fall the 4 watersheds, with the exception o fLake Alaotra, are above 10 percent which is regarded as the opportunity cost o f capital inMadagascar. The ERR of Lake Alaotra, at 8%, has been underestimated by excluding positive economic and environmental benefits associated with water storage in the dams, incuding benefits from reduced flooding o f the Lake Alaotra watershed, a designated Ramsar site under the Ramsar Convention on Wetlands. Investments in the Lake Alaotra watershed are justified on the ground that they are considered by the government to be highly strategic for food security purposes and also to reduce flooding. All together, the project has an overall ERR o f 14 percent and an NPV o f US$8.8 million, computed at a discount rate o f 10 percent. Both the ERR and NPV are conservatively estimated because: (i) the current producer price for rice i s assumed to remain constant for the next 25 years while international forecasters are predicting an increase in international rice prices over the same period; (ii) not all environmental benefit have been captured in the economic benefit stream; and (iii) some benefits from investments in policy and capacity buildingmost o f which will be reaped by the future phases o f the APL have not been included inthe analysis. The variables that influence most the net economic benefits from the project are: (i)the producer price o f paddy; and (ii) the ability o f W A Sand the government to maintain irrigation infrastructure (including if a cyclone hits any perimeters) beyond the life o f the project. These are the variables that have been used inthe sensitivity analysis. First, ifthe producer price o fpaddy declines by 18 percent from the current price o f about 21 cents per kgto 17 cents per kg inthe second year o f the project and stays at this level, the project will not be profitable. However, this is unlikely given the current trend on the international rice market where a long term price increase i s forecasted. Second, ifWUAs do not maintain productivity on well- irrigated areas for more than 7 years after project completion, or if a cyclone hits Marovay's 6,000 hectares o f well-irrigated perimeters after four years o f project implementation without being repaired, the project's NPV declines to zero. The institutional arrangements built inthe project and the level o f project benefits should be able to mitigate against these risks. The main beneficiaries o f the project, about 500,000 people, that will see their incomes grow through gains in agricultural productivity, will be in a position to pay the incremental recurrent costs through their W A Sto maintain infrastructure andfinance insurancemechanisms. Technical 93. Irrigation investment operations have had a mixed experience in Madagascar. While investments were generally justified in terms o f increase in production, sustainability has been far from sure. The project will focus on increased production and higher value, but in particular on translating higher income into better maintenance o f infrastructure through capacity strengthening and improving governance o f hydraulic assets. In addition, the project will invest in upper watersheds to promote sustainable land use practices, which i s expected to deliver higher production o f rainfed agriculture, while at the same time reducing sedimentation and thus maintenance costs. The project will thus adopt a three-pronged strategy: (i) increase production and farmers' income, (ii) inplace put mechanisms for sustainable irrigation maintenance; and (iii) reduce irrigation maintenance costs. The strategy is based among others on the experience o f the BV-Lac project in PC15/Marianana (Lac Alaotra) and the GTZ-funded project inMarovoay. 94. Extension is a critical ingredient to increase agricultural productivity. The P N V A project (completed in 1999) adopted a `training and visit' approach that resulted inincreasedproductivity, but with limited sustainability and high staff costs. Based on international experience, the project will 25 support a demand-driven approach to extension services that are, ultimately, to be provided by private service providers on a commercial basis. Establishment o f Agriculture Service Centers (ASC) will be supported by the project as a platformto bringtogether supply anddemand for extension services. 95. Agricultural value chains in Madagascar are characterized by weak linkages between actors. Agricultural marketing has not traditionally been part o f irrigation development, despite the fact that many o f the benefits of irrigation (higher quality, more uniformproducts, possibility to more precisely target harvest dates) are more relevant in a market environment. Based on the experience o f similar projects in West Africa, the project intends to capture the synergies between markets and irrigated agriculture to improve quality andvalue, and increase farm gate prices through targeted investments in linkages between operators inthe chain. 96. Watersheds form integrated spatial management units with irrigation schemes. Failure to address synergies between the two has ledto missed opportunities andreduced returns on investments. The project proposes to address productivity o f agriculture in both irrigated low lands and rainfed watersheds, while capturing the environmental externalities associated with more sustainable land use and management. The integrated design o f the project is based on similar projects in Madagascar financed by FA0 and AFD, and on an Africa Landand Water Initiative pilot project inAnjepy. Fiduciary 97. Procurement: The third Country Procurement Assessment Review (CPAR) for Madagascar was conducted in November 2002, followed by a workshop in June 2003 for the validation o f a joint CPAIUCFAA action plan to ensure rapid implementation o f procurement reforms. Key elements o f these reforms are: (i)revision o f the draft procurement code to ensure transparency, simplify procedures, and comply with international standards, (ii)establishment o f effective procurement institutions to ensure that the new regulations will be adequately applied, and to provide sufficient oversight and control; and (iii) adequate training and capacity building to ensure the sustainability o f the procurement reforms. A new procurement code was enacted in July 2004 but since the texts for regulatory application are still under preparation, the existing Procurement Code o f 1998 will continue to be applied. The World Bank ascertained that the deficient features identified in the 2003 CPAR have been properly addressed. IDA standardbiddingdocuments are also widely used. 98. A remaining area o f concern is the Government's cumbersome and overly bureaucratic approval process for contract signing, which causes unnecessary delays. In addition, insufficient procurement planning contributes to delays in project implementation which results in slow disbursement. To mitigate the risk o f delays, proper prerequisites for the use o f Bank standard bidding documents, including evaluation reports for National Competitive Bidding procedures (NCB) have been agreed upon with Government during negotiations. The procedures manual will be updated as a part o fthe Project ImplementationPlan. 99. A Procurement Capacity Assessment of the MAEP, including training needs and arrangements, was conducted as part o f the project preparation. On the basis o f the initial assessment, an action planwas drafted to address areas where MAEPneeds to be strengthened. This includes (i) a specific section on procurement in the Project Implementation Plan to be finalized or updated before Credit effectiveness; (ii) improved filing organization o f procurement-related documents (including in the regional offices); (iii)procurement training sessions for project staff; (iv) the recruitment o f technical assistance to help MAEP handle the project procurement load, and (v) the financing o f independent procurement and technical audits on a regular basis. 100. Financial management: The overall conclusion o f the MAEP's financial management assessment carried out during the pre-appraisal mission indicates that the financial management arrangements inplace do not yet satisfy the Bank's financial management requirements as specified in OP/BP 10.02. An agreed action plan was developed with MAEP to strengthen the existing financial management system and build capacity to produce quarterly Financial Monitoring Reports. The main 26 measuresagreedare the following: (i) recruitment, under terms and conditions acceptable to IDA, of a Financial Management and Procurement Agency to provide the DFB and the four RDFB with qualified and experienced accountants, and assist the accounting staff o f DFB/RDFB in performing financial management tasks; (ii) elaboration and implementation o f an accounting procedures manual providing sufficient information to facilitate adequate record keeping and the maintenance o f proper control over assets; (iii)design and implementationof a computerized systemto satisfy MAJ3PProject requirementsand ensure timelyproduction of financial statements and Financial Management Reports (FMRs).These recommendations shouldbe implementedprior to credit effectiveness. The content and formats o f financial statements and FMRs will be determined during the forthcoming appraisal mission and agreed at negotiations. 101. To mitigate the risks posed by the limited capacity of the Auditor General (Chambre des Comptes), includinglack o f adequate number of skilled and experienced auditors commensurate with its needs, the audit of the project financial statements will be carried out during the first three years by a qualified private auditing firm acceptable to IDA. During the second half o f the implementation period, and following a favorable IDA assessment, MAEP could use the service o f the Auditor General. The audits will be conducted in accordance with International Auditing Standards, and the auditors recruited prior to credit effectiveness. The audit report should be submitted to IDA no later than6 months after the endof each fiscal year. Social analysis 102. The large rice producing irrigation schemes constructed over the last fifty years have attracted migrants from other parts o f the country. Some o f the farmers who have land inthe irrigation schemes also often have land in the watersheds surrounding the irrigation schemes. Other farmers have only land in the surrounding watersheds. Degradation of agricultural production systems in the irrigation schemes and in the watersheds has led to reduced agricultural production and consequently to increasedpoverty. Degradation inthe watersheds, inparticular, has been dramatic and may over time lead to abandonment o f the land. The project will aim to sustainably increase agricultural production, diversification and revenues in the four sites. Ago-ecological agricultural practices, which have the potential to triple agricultural production, will be promoted in the watersheds to increase farmers' income, but also to reduce or stabilize man-made erosion, increase soil fertility, improve vegetation cover, and reduce bush fires. The project is also expected to contribute to increased land security in bothproduction irrigated and watershed systems. 103. The project will examine carefully the position o f sharecroppers in the irrigation schemes, where share cropping is most common. It will ensure that the capacity o f the private operators is not strengthenedat the expense o f smallholders, marginalizing vulnerable groups. 104. The project will strengthen V A Sin order to improve the management and maintenance o f the irrigation schemes. It will also establish or strengthen communication and consultation platforms in each watersheds (which will include WUA representatives) to improve the management of natural resources and develop sustainable agricultural systems. It is expected that these activities will have a positive environmental and social impact on the sustainable use o f the natural resource base and reduce siltation on the downstream irrigation schemes, which inturn would have a positive impact on poverty reduction inbothproduction systems. Environmental analysis 105. Madagascar is a mountainous country with a relatively low population density. The country has abundant land and water resources, which are only partlydeveloped, and biodiversity resourceso f global significance. Madagascar has a high natural erosion rate, as a consequence o f its soil types and heavy rainfall, often exacerbated by cyclones. This high natural erosion rate has been exacerbatedby deforestation of erosion prone fragile soils, frequent bush fires (many o f which linked to livestock grazing) and unsustainable agricultural practices inthe watersheds, which made most o f the watershed 27 soils infertile and marginal for agricultural and livestock production. This pattern o f severe land degradation has lead over the years to reduced agricultural production and increased poverty. This, together with increased land scarcity in the four highpotential sites, has increased the pressure on the watersheds and has lead to increased deforestation and pressure on the globally important biodiversity resources in the watersheds in three project sites: Marojejy National Park, the South Anjanaharibe Special Reserve, and the Makira Conservation Site, all located in the upper watersheds around the Andapa irrigation scheme; the Ankarafantsika National Park located in the upper Marovoay watershed; and the Lac Alaotra Ramsar site. In Itasy, agriculture i s practiced on very steep slopes, which in other places are kept under a mandatory forest cover to minimize erosion. Slash and burn agriculture i s still practiced, particularly in Andapa. These unsustainable agricultural practices have exacerbated the already high natural erosion rates and led to sedimentation and flooding o f downstream irrigation schemes, severely hampering irrigated rice production and increasing poverty. The impact o f the degraded environment on the agricultural production systems i s significant. This situation was made worse by the absence of adequatemaintenance o f the schemes. 106. The project would seek to reverse this trend by rehabilitating and improving the management of the existing irrigation schemes, as well as by stabilizing or reversing land degradation in the watersheds through the promotion of more sustainable agro-ecological practices. These improved practices should, over time, reduce soil erosion and sedimentation in the downstream schemes. Over the short term, it is expected that these improved practices will significantly increase agricultural production of traditional and new crops inthe watershed areas, and thereby help reduce poverty. One of the requirementsfor increased production will be the integration of agriculture and livestock (such as use o f dung as fertilizer and organic soil conditioner). It is also expected that intensified agricultural practices will reduce or stabilize agricultural expansion and thus reduce the pressure on the remaining highbiodiversity resourcesinthe watersheds. 107. The project is expected to have mostly beneficial environmental and social impacts, as demonstratedby GoM's Regional Environmental and Social Assessment (RESA). The main positive environmental impact will be the improvement o f environmental services o f the watersheds through the adoption of agro-ecological production systems and better management of pastures, which will stabilize or reduce erosion rates. 108, Intensified agricultural production may require increased use o f chemical fertilizers and pesticides. GoM has thus prepared a Pest and Pesticide Management Plan (PPMP) to mitigate the health and environmental impacts of increasedpesticide use. It is at present not clear iffarmers will be able to afford and maintainthe financing o f increasedinputs. 109. Irrigation schemes in Madagascar are main sources o f waterborne diseases, such as malaria and urinary and intestinal bilharzia and diarrhea. The four selectedproject sites are no exception. The Environmental and Social Management Plan (ESMP) has included measures to reduce these diseases inorder not to impair theproduction capacity ofthe farmers andimprovetheir quality oflife. 110. The major potential environmental risk posed by the project would be the potential attraction of an influx o f migrants from other areas o f Madagascar should the project be successful inincreasing agricultural production inthe watersheds. These migrants would increase the already highpressure on land in the four project watershed areas, which could lead to further deforestation o f the sites, increaseduse of steep hillsfor agriculture production, and further clearing o f reed lands in Lac Alaotra for rice production. Land zoning, transfer o f land management to existing social groups, and empowerment of farmers and farmer's groups to manage these lands will therefore be o f fundamental importance duringproject implementation. Safeguard Policies 111. The Safeguard Policy issues raised by the project have been briefly discussed above and below and are further detailed inAnnex 10. 28 112. The project has been categorized as a Category A project, since three o f the project sites are located in areas with globally important biodiversity resources, which increases the reputational risk for the Bank. As stated above, the project activities themselves will have mostly positive environmental and social impacts, with environmental management measures fully integrated into project design. However, increased use o f fertilizers and pesticides may have negative impacts on the Lac Alaotra Ramsar site, Lac Itasy, the mangrove habitats inthe Marovoay area and the Lokoho River inAndapa. Inmany areas, river andlake water is also used for drinkingpurposes. 113. The following World Bank Safeguard Policies were triggered: Table 4: Safeguard Policies Triggered bythe Project SafeguardPoliciesTriggered by the Project Yes No Environmental Assessment (OP/BP/GP 4.01) [XI [ I Natural Habitats(OP/BP 4.04) [XI [I Pest Management(OP 4.09) [XI [I Cultural Property (OPN 11.03,being revised as OP 4.11) [I [XI Involuntary Resettlement(OP/BP 4.12) [XI [I IndigenousPeoples(OP 4.10) [I [XI Forests(OP/BP 4.36) [XI [ I Safety of Dams (OP/BP 4.37) [I [XI Projectsin DisputedAreas (OPIBPIGP 7.60)' [I [XI Projects on International Waterways(OPIBPIGP 7.50) [I [XI 114. Environmental Assessment, Natural Habitat and Forests. G o M has prepared a Regional Environmental and Social Assessment (RESA) which has been disclosed at the project sites, at the national level, and in the Infoshop in Washington prior to appraisal. Agro-ecological production systems and improved pasture management will be promoted in degraded and deforested soils in the watersheds. Sites where large amounts o f sediments originate and which affect the downstream irrigation schemes will be given priority. By preparing and implementing a land use zoning plan and transferring the management o f landinthe watersheds to communities it is expected that landuse will change from an open access situation to a regulated access natural resource, where migrants cannot settle freely any longer. Intensification o f the watershed agricultural systems and a change to higher productive and less erosion prone agro-ecological practices it also expected to reduce the pressure on the globally important biodiversity resources in the upper watersheds. This approach satisfies the Environmental Assessment Safeguard Policy OP/BP 4.0 1, Natural Habitat Safeguard Policy OP/BP 4.04 and the Forests Safeguard Policy OP/BP 4.36. 115. The project will also finance sub-projects, such as check dams, anti-erosion structures, small irrigation dams, markets or other structures. These sub-projects will be screened for environmental and social impacts by the Technical Secretariat o f the Matching Grant Mechanism (to be financed under the project), that will also identify if a Resettlement Action Plan (RAP) and/or a small Environmental Assessment study will be needed as part o f the feasibility analysis. 116. Pest Management. G o M has addressed the requirements o f the Pest Management Policy OP/BP 4.09 by preparing and disclosing a Pest and Pesticide Management Plan (PPMP) acceptable to IDA. The PPMP includes a number o f actions which will reduce the exposure o f the farming community to pesticides used in the agricultural production systems as well as pesticides used for malaria control in the project areas. The PPMP will also promote the development and establishment o f Integrated Pest Management Practices (IPM). By supporting theproposed project, the Bank does not intend toprejudice thefinal determination of theparties' claims on the disputed areas 29 117. Involuntary Resettlement. GoM has also met the requirements of the Bank's Involuntary Resettlement Safeguard Policy (OP/BP 4.12) by preparing and disclosing a Resettlement Policy Framework (RPF). It is expected that any potential resettlement, land acquisition or loss of access to traditional natural resources will occur at a limited scale. Should this happen, a Resettlement Action Plan (RAP) will be prepared to ensure that people are fully compensated (at replacement costs) and will not be worse off than before the project intervened. Sub-projects will be screened to identify whether a RAPwill be required(see also EnvironmentalAssessment, above). 118. Safety of Dams. The Safety of Dams Safeguard Policy is not triggered. The project will rehabilitate a scheme that is served by an irrigation reservoir. At the same time the safety o f the dam (less than 15 meter) will be inspected and ifneededbrought up to international dam safety standards. 119. Analysis of alternatives. Feasible alternatives are (i) not implement the project; or (ii) to to implement it without a watershed management component. The "no project" alternative would allow further deterioration o f the irrigation schemes and the watersheds with consequentnegative impacts on poverty, agricultural production, and globally significant biodiversity sites. The alternative "without watershed management" would leave the irrigation systems exposed to large sediment loads, which would endanger and potentially underminethe investments. 120. Public consultation. Public consultations have beencarried out on the Terms o f Reference of the RESA, on the draft report, as well as during the preparation of the RPF.This is inconformity with the requirementso f OP 4.01 and OP 4.12. 121. Borrower Capacity and Implementation and Monitoring of the ESMP. The Borrower's capacity to supervise and monitor the implementation o f the Environmental and Social Management Plan (ESMP) has to be strengthened. One o f the Technical Assistants to be financed under the project will be qualified in environmental and social management and will be made responsible for the adequate implementation and monitoring o f the ESMP. Parts o f the ESMP will also be implemented by contracted service providers. Ifneededthe capacity of these service providerswill be strengthened. 122. Disclosure. The Regional Environmental and Social Assessment, the Pest and Pesticide Management Plan and the Resettlement Policy Framework have been disclosed at the four project sites, inAntananarivo, and in the Infoshop in Washington prior to appraisal. Policy Exceptions and Readiness 123. The project requires no exceptions to Bankpolicy. 30 Annex 1: National, Sectoral and Program Context A. National and Sectoral Context 1. The Island of Madagascar covers a total area of 588,841 km'. The population, estimated at 16.4millioninhabitantsin2003, is increasingat an annual rate of about 2.8 percent. Nearly78 percent of the population lives in the rural area. The country is characterized by major biodiversity and considerable cultural and socio-economic diversity. The economy is essentially rural-based and agricultureremains the main engine of economic development.Per capita income is US$290.Poverty affects 68.7 percent of the total population and 73.5 percent ofthe rural population. Poverty Reduction Strategic Framework 2. The development objective for Madagascar as defined in the Poverty Reduction Strategic Framework (PRSF/PRSP, July 2003), is to promote a rapid and sustainable development with the aim to reduce by half the poverty rate within ten years. It is organizedaroundthree strategic orientations: (i)restoringthe rule oflaw and awell governed society; (ii)foster and encouragebroad-based economic growth; and (iii) foster and encourage systems for ensuringhumanand material security and enlarged social protection. The second strategic orientation of the PRSFIPRSP targets five general objectives:(i)to reach an economic growth rate of 8-10 percent per annum; (ii) increase the level of to investment to 20 percent; (iii) foster the dynamism of the private sector so that it participates in an investment rate of 12-14 percent to the investment rate; (iv) to open up Madagascar's economy to greater competitionwith a view to reducingcosts and improving quality; and (v) foster the willingness of the populationto participate. 3. The PRSF/PRSP implementation programs concerning agriculture will essentially aim at "ensuring food security and making optimal use of resources", through five objectives: (i) increase to agricultural productivity and cultivated areas; (ii)to promote small-scale investments in rural areas and partnershipbetweenfarmers' associations and the private sector; (iii) promote agricultural and to ago-food exports and improve their quality; (iv) to ensure transparent and rational management of resources to guarantee their sustainability; and (v) to facilitate producers' access to land capital. Each specific objective corresponds to an investmentprogramregroupingseveral clearly identified actions. Programs under the first objective include: development of irrigated schemes and surrounding watersheds. 4. The Madagascar Action Plan (MAP) 2012 i s the Government's new vision that sets out a "roadmap" for development. The purpose of the MAP is to produce a quantum leap in the development process by having a five year plan that will mobilize the Malagasy people and the country's international partners to ignite rapid growth, thus leading to the reduction of poverty. Its goal is also to ensure that the country develops in response to the challenges of globalization and in accordance with the national vision "Madagascar Naturellement" defined by the President in November 2004 (described below), the long-termvision for the country of which the MAP is a 5-year implementationprogram. The MAP is a direct follow-on from the PRSP and covers the period 2007- 2011. The MAP will be finalized in October 2006 and presentedto Parliamenttogether with the 2007 budget. 5. Madagascar Naturellement is the statement of the country's development vision in 2015. It states that Madagascar will be a newly industrialized country with maximized competitiveness by 2020. The core of growth will be derived from the country's unique natural resources and from the transformation of its natural products. The vision aims to develop a diversified and rich natural resource base (agriculture, livestock, fisheries, and mining) that will contribute to the creation of products with high value added such as essential oils, agri-business, pharmaceuticals, and mining products. A broader impact of growth and a progressive redistribution of its benefits will help reduce poverty substantially. Madagascar will be known worldwide for the beauty of its rich and well- 31 protected biodiversity and its environment will be cherished and protected and used in a wise and responsible way to enhance development. The Malagasy people, both in rural and urban areas, will be healthy and well-educated, will be active participants inthe development process andwill be gainfully employed in agriculture, industry and the provision o f services. Education and health will be accessibleto the population and infrastructure will be developed allowing for free movement o f goods andpeople. 6. The MAP'Sgoals are ambitious targets that the Government is committed to achieving over the coming five years. It is recognized that these targets are challenging but, at the same time, it is expected that results will be delivered through the mobilization o f the full array o f the available resources(human,technical and financial), combinedwith strong leadership, hard work, creativity and excellent coordination. The MAP outlines eight specific areas o f focus; (i)good governance; (ii) educational transformation; (iii) and family planning; (iv) infrastructure; (v) rural development; health (vi) the economy and the private sector; (vii) environment; and (viii) national solidarity. There are quantifiable goals set out for each o f these areas o f focus to be achieved by 2012; these goals will be further refined, costedand clearly linkedto the MDGs. Agriculture, rice, and irrigation 7. Rice represents nearly 70 percent o f agricultural production and accounts for 48 percent o f total calorie consumption. Rice production has only increased by 1.2 percent per annum since the 1980s and average paddy yield at the national level is still low (about 2.4 t/ha). Annual production o f paddy rice has virtually stagnated for about ten years, stabilizing between 2.3 and 3.0 million tons. Area planted to paddy has increased by only 0.44 percent per year from 1970 to 2004; yields have increased by 0.71 percent per year, much slower than in other major rice producing countries. With population growth o f 2.7 percent per year, production per person has fallen from 275 kglperson in 1970 to only 179 kgperson in2004. Rice farming techniques are largely traditional and use o f inputs i s the exception in many places. E.g., fertilizer use has remained stagnant at 10 kglha on average, as compared to 14 kgha in sub-Saharan Africa, and 291 kg/ha in Indonesia. Vast differences in prices between wet and dry season are explained by the lack of fluidity in movement of goods from production areas to the markets due to a lack o f road infrastructure and lack management capacity o f storage facilities by farmers. On average, 28 percent o f the paddy production is marketed (750,000 t). Rice sales are highly concentrated. In 2001, the top 10 percent o f rice farmers (by value o f sales) accounted for 73 percent of total nationalrice sales. These farmers sold on average 2.2 tonshousehold. An estimated48 percent ofrice farmers didnot sell any rice in2001. 8. Irrigation occupies an important place in the agricultural sector, supplying water to more than one million hectares, or 40 percent o f cultivated lands (as compared to 6 percent on average in sub- Saharan Africa). Irrigated crops represent 15 percent o f GDP, whereas 70 percent o f agricultural production and 88 percent o f rice production originate from irrigated agriculture. It i s estimated that 85 percent o f the active farming population are directly or indirectly employed by the irrigation sector. Since the 195Os, irrigation has benefitedfrom public investment. However, the impact o f these efforts on rural incomes i s mixed, and sustainability i s far from certain. The rapid degradation of infrastructures requires frequent rehabilitation, and many schemes are caught in a vicious circle o f poor yields, low capacity o f water users to pay for O&M, and rapiddegradation o f the schemes. Weak capacity to pay i s accompanied by low willingness to pay, reinforced by institutional weakness o f the WUA and a lack of support from local authorities. Moreover, erosion of watershed upstream is weighing heavily on cost ofmaintenance o f irrigation schemes. 9. Extension services have failed to have a significant impact on productivity levels either, and have demonstrated to be unsustainable. Reasons for these past failures include (i) the approach was biased in favor o f technical messages, (ii)inadequate consideration o f the demand for extension services and the economic constraints that farmers face ;farmers were considered more as the objects than as the subjects of extension services, (iii) approach was too to cenntralized, with inadequate the 32 attention for regional variation, (iv) inadequate capacity o f extension agents, (v) unrealistic expectations about the volume o f public (humanand financial) resources available. Natural resources, soil developmentand role of communes 10. One o f the basic problems o f the rural and agricultural sectors i s the rapid degradation o f natural resources, particularly watersheds. The stagnation o f yields inirrigation areas and demographic growth lead to an extension o f rain-fed crops on hill slopes (tanety/tavy), often by removing the forest cover and by replacing it with inappropriate farming practices. Unproductive pastures are degraded by frequent passage o f bushfires. As a result, soils are increasingly degraded and fragilized, and even low levels o f runoff lead to high levels o f erosion that cause damage to downstream assets, reduce the lowland area under irrigation through sedimentation, wet season flooding and dry season droughts. In addition, there are important implications in terms o f biodiversity loss and declining buffering and regulatory ecological services. More sustainable land management practices (ugro-&dogie) have demonstrated that i t i s possible to achieve the dual objective o f higher productivity and reduced soil degradation and erosion. 11. Communes and Regions are responsible for land use planning and play an important role in providinglandtenure security: the communes should therefore be at the centre o f all natural resources management and watershed development initiatives. The Communes have been established to provide a number o f basic services to the populations (role o f public service provider) and to act as the engine o f development on its territory. To that end, the capacities o f the Communes should be strengthened in the following areas: (i) initiating development within the Commune, including: (a) support for the elaboration and monitoring o f Communal Development Plans (CDP), (b) financing o f investments; (ii) implementation o f their specific mandate, including: (a) implementation o f responsibilities in the area o f education, health, water, sanitation, andmaintenance o f infrastructures that have been transferred to them by the central Government, (b) technical assistance in the area o f economic development and management o f natural resources, (c) land tenure policy (landtenure counters), and (d) the integration of intercommunal priorities inthe development policies o f the Commune'. 12. Tenure security through delivery o f formal documents is important because it can lead to better use o f land and it facilitates improved fiscal resources. Traditional leasing arrangements, currently outlawed in Madagascar, provide an environment that is non-conducive for investments in productivity. 13. Given the importance o f the responsibilities entrusted to communes and the low level o f human and financial resources at their disposal to meet these challenges, it is indispensable to put in place a support mechanism. The Ministryo f Decentralization and LandUse Planning(MDAT) has put in place a program for strengthening the capacities of Communes in administrative and financial management. To that end, District Support Centres (DSCs) will be established in the regions. These DSCs will be responsible for: (i)training elected officers and staff o f the Communes in budgetlfinancial management and administrative procedures associated with project implementation (procurement, etc.); (ii) establishing the necessary budgevfinancial management and administrative tools; and (iii) technical assistance for management andmonitoring o f the activities o f the communes. Land tenure security 14. Madagascar has a highdemand for landtenure security, as evidenced by the many requests for land title deeds (which the present system is incapable o f meeting), i d the development o f an informal local system o f `petits papiers' that is highlysolicited to record transactions. MDAT, July 2005: Review of local development programs inMadagascar, Document n02-Towards a national decentralizationsupport policy. 33 15. Specifically, situations o f high tenure insecurity exist concerning those farmers cultivating land in former AMVR, ZAF, colonization areas or indigenous reserves that are often the subject o f competitive claims, and farmers who cultivate as sharecroppers or tenants. Either category is widespread in the irrigation schemes in the intervention areas o f the project, as evidenced by the diagnostic studies. The unofficial nature o f these rights weakens particularly the fimctioning o f WUAs andO&M o f irrigation schemes. 16. To meet the high demand for land tenure security, the Government recently adopted a Land Policy Letter, which is organized around 4 strategic orientations: (i) restructuring / modernization o f land services; (ii) decentralization o f land management; (iii) revision o f land regulations and (iv) capacity strengthening. This policy is being implemented under the National Land Tenure Program that is already supporting, on pilot basis, several decentralized landmanagement experiences with support from several donor agencies. B. Lessons learned 17. Previous attempts to boost agricultural production through investments in irrigation infrastructure have been unsuccessful, in particular with respect to the sustainability o f the investments. Despite modest increases in yield levels on those schemes that have benefited from investments, a weak institutional environment and high O&M costs have undermined capacity and willingness to pay O&M charges. In addition, only 10 percent o f irrigation schemes have benefited from investment, and modest yield increases have not been visible interms o f national averages. The reasonsfor low yields and weak sustainability are notably: (i) o f market opportunities (isolation, lack unattractive prices); (ii)lack o f access to advice and inputs; (iii) to take into account watersheds failure upstream; (iv) lack o f clarity in responsibilities and capacities o f the different public, associative and private partners; (v) non-respect o f commitment by both users and the State; and (vi) indiscipline and impunity. 18. The majority o f Malagasy farmers only benefited marginally from the technological options proposed, and average yields are well below the actual potential. Tradition and risk aversion only partially explain the failure o f agricultural intensification. Other factors can be mentioned, such as: (i) weak capacity o f agricultural research to respond to request o f farmers, as well as their low level o f organization and participation in the development process; (ii) poor extension services (in terms o f access and quality); (iii)land tenure insecurity and inequitable sharing o f profits, particularly by sharecroppers; and (iv) low tolerance o f potential technologies to climate shocks. At the level o f extension services, lessons from failures (Le. PNVA) include, among others: (i)an approach excessively focused on technical solutions, (ii) poor consideration o f demand and economic concerns, (iii)excessively centralized, with low regional identity, (iv) capacity constraints o f extension workers, (v) interventionisthigid approaches and low level o f partnerships and empowerment o f beneficiaries, and (vi) unrealistic expectations o f public support in terms o f human resources and financial sustainability. 19. The conditions o f success include: (i)an integrated approach to irrigated agriculture and surrounding watersheds; (ii)conducive economic environment; (iii)clear responsibilities, in conformity with Government polices and strategies (poverty reduction, decentralization, agricultural, environmental and landpolicy, etc.); (iv) fully responsible partners with adequate capacities; (v) clear and unambiguous commitments corresponding to the capacities o f each o f the parties, contracted freely and knowingly; and (vi) mechanisms to ensure respect o f commitments made that are applied systematically. 20. The BV-PI integrated approach is a "win-win" approach, which at the same time helps to increase productivity and incomes in irrigation schemes and surroundings watersheds, conserve natural resources in watersheds, limit erosion o f slopes and sedimentation in irrigation schemes, thereby reducing the need for maintenance andrehabilitation o f the latter. 34 21. An attractive economic environment implies: (i)a policy on prices o f agricultural products and inputs; (ii) to markets interms through roads, information, promotiono f private sector and access producers' organizations for marketing (including storage) and supply o f inputs; (iii) to access efficient extension services well adapted to local needs; and (iv) access to finance. 22. Clear institutional framework: clear institutional responsibilities in line with Government policies and regulations for producerdusers and their associations, communes, inter-communes and regions, decentralized public services, specialized agencies and authorities (ANDEA, etc.), and private operators. 23. Participatory approach, concerted decisions and respect of commitments made: actors with clear and acknowledged rights and obligations, and adequate resources and capacities, participating fully indecision-making; incentives and mechanisms ensuring responsible ownership and respect o f commitments made; interfaces for dialogue and communication; and equitable access to resources, especially for the most vulnerable population groups. 24. The improvement o f irrigation infiastructure and the establishment o f sustainable mechanisms for funding O&M will not be enough to increase rice production beyond about 3.5 h a , which is still low compared to the technical potential. Promotion o f intensification o f rice production systems inIPS ( S W S R I ) , including inareas with poor control over water, will need to be undertaken. Moreover, the agro-ecological techniques o f seeding and plantingon permanent plant cover (SCV) developed by the GroupementSemis Direct Madagascar (GSDM), supported by CIRAD, are opening new prospects for sustainable and profitable agriculture on slopes. The environmental advantages o f SCV techniques include: (i)erosion control, soil conservation and regeneration o f soil fertility at reduced cost; (ii) improvement o f infiltration, efficient management o f water in the upper watersheds; (iii)sustainable improvement o f soil fertility and productivity inthe upper watersheds; and (iv) indirect contribution to sequestration o f carbon and reduction o f the greenhouse effect. Finally, agricultural diversification, including off-season production o f higher value-added crops will help improve incomes and living conditions o f farmers, and facilitate their greater participation in the financing o f O&M o f irrigation schemes. 25. Addressing local or regional diversity in terms o f natural, social, economic and physical resources is essential for ensuring sustainable and appropriate agricultural development. Success inthe duration o f a program largely depends on its level o f ownership by target groups: consequently, strengthening dialogue and decision-making capacity o f the peasant community constitute the cornerstones o f sustainability. C. National Irrigation and Watershed Management Program 26. The National Irrigation and Watershed Management Program (PN/BV-PI) i s part o f a program under the PRSF/PRSP that aims at reducing rural poverty through sustainable improvement in the living conditions and incomes of rural populations in irrigated perimeters and surrounding watersheds, and efficient management o f naturalresources. 27. The Government has clearly defined its new medium-term vision o f the management o f BV- PI, based on national policies on rural and agricultural development and the decentralization policy, which is at the centre o f its development and poverty reduction strategy. This approach requires: (i) clear responsibilities for each o f the actors in the management o f irrigation schemes and surrounding watersheds (farmers, water users, professional associations, districts and inter-communities, regions, central Government); (ii)effective participation o f rural populations in diagnosis o f problems and identification o f options; (iii)co-management o f P I and BV by all the actors concerned; and (iv) incentives and efficient mechanisms to ensure that all stakeholders respect their commitments. 28. One o f the key objectives o f the first phase o f the PN/BV-PI, o f which the IDNGEF funded project constitutes a major part, is to put in place a clear and attractive institutional environment as 35 well as adequate capacities at all levels, with a view to attaining the Government's vision and objectives. For its implementation, the project will adopt a flexible approach adapted to the reality in the field and evolution of capacities o f the institutions, which will be gradually strengthened with a view to their empowerment. D. Project Zones Marovoay 29. The Marovoay plains i s a rice production zone o f prime national importance, situated in the Boeny Region, about 80 km South-East o f Mahajanga. The Marovoay river is a tributary on the right bank of the Basse Betsiboka, inthe upper delta of the river. Subjected to quasi-complete submersion during the annual flooding of the Betsiboka, the development of the valley started in the early 20th Century for off-season rice production (once the water-level has dropped). Later extensions to the gravity systems included schemes supplied through pumping from the Betsiboka. The scheme i s divided into 13 completely independent irrigation sectors, fed from a great number of different sources. The system faces serious O&M challenges. The submersion o f schemes by waters from the river requires annual rehabilitation of the irrigation infrastructure, thus making O&M expensive and the overall economic profitability uncertain. For a total area of about 20,000 hectares, an estimated area o f 12,000 hectares was cultivated in 2004. Beneficiaries o f all plots developed during the successive programs were mainly immigrant populations from other regions o f the country. The percentageo f sharecroppers is today very high. 30. Until recently, the central Government was responsible for O&M of the irrigation schemes and pumps. Presently, public funds for maintenance o f structures considered as `non transferable' are unreliable. Restructuring into WUAs and federations o f WUAs has not resulted inthe establishment of an adequately O&M. The Performance Contract signed with the federation for the period 2001-2003 was not renewed and funds earmarked for 2004 were reallocated. 31. The main watershed serving the Marovoay irrigated perimeters i s that o f River Betsiboka, whose hydrology i s determined by phenomena occurring some hundreds o f km upstream. Sub- watersheds o f River Marovoay and its tributaries supply a major part o f the system: their sources are mainly inthe zone of Ankarafantsika National Park, where humanactivities are controlled. Finally, all around the plain, small lateral watersheds with mainly intermittent flows do not constitute a source o f irrigation water supply but have a major impact interms o f erosion, sedimentation and destruction of protectionand distribution structures alongside irrigatedperimeters. Itasy 32. Itasy Region, located around Lac Itasy, is situated about 100 kmto the West o f Antananarivo. All irrigation schemes in Itasy (Gruppedu Lac Itusy 1,980 hectares, Ifanja 1,900 hectares, Mangabe 270 hectares, Analavory 140 hectares, Ampary 90 hectares, Antanimenakely 80 hectares- or a total o f 4,460 hectares) are presently classified as autonomous perimeters, as complex `non-transferable' infrastructure i s absent. The region offers great potential for agricultural production, given the natural fertility ofvolcanic, basal andalluvial soils and its favorable climate for agricultural diversification. 33. The highconcentration of population in the zone (107 inhabitantshZon average) has caused problems o f gradual over-exploitation o f tanety located upstream o f the irrigation schemes. The deforestation o f watersheds caused by annual bushfires, uncontrolled exploitation o f the tanety for rain-fed crops and grazing o f zebus, causes problems o f erosion and silting-up o f the rivers and irrigation systems. 34. Although most o f these schemes benefited from projects implemented from 1998 to 2000 (project PPI 2), they are currently facing serious problems due to a combination o f erosion o f the upper watersheds and lack of maintenance of the systems. In addition, storage infrastructure has been 36 silted up and is no longer adequate, also given the change inthe flow regime o f the rivers (increase in flood flow and reduction o f dry-weather flow). Hence, 30-50 percent o f the perimeters are no longer adequately irrigated. Given (or as the origin of) these problems, W A Shave stopped collecting maintenance fees for several years, since a greater part o f the users have refused to pay as they are no longer benefiting from water control. The actions o f the W A Sare limited to maintenance works carried out by interested users, i.e., in most cases, those o f the downstream sectors o f the irrigated perimeters. Andapa 35. The Lokoho watershed at Andapa, situated in the Sava Region at about 100 kms South West o f Sambava, i s formed by three concentric landscapes: (i)the first covers a vast plain o f crops, 18,000 hectares, drained by 4 main rivers whose confluences form River Lokoho at the exit o f the basin; (ii) the second i s constituted by tanety, at the periphery o f rice farms, marked by a diversity o f annual crops (mainly rain-fed rice) on cleared forest (tavy) or planted fallow lands, as well as coffee and vanilla crops; (iii) third, at an altitude o f over 900 m is distinguished by a denser tree cover. the The basin i s bordered inthe North-East by Marojejy National Park, inthe South-East by Anjananaribe South Special Natural Reserve, which is the only forest zone o f the basin where tree cutting is still authorized, though regulated. 36. From 1962-1997, the Andapa basin has benefited from a development program funded by EDF.The project comprised aninfrastructure component, which included the roadlinkingAndapa and Sambava, drainage o f the basin, internal network o f access roads, development o f the main waste water outfall o f the basin and construction o f a pumping station. The agricultural component focussed on development o f rice farms on a total area o f 4,400 hectares, introduction o f double season rice cultivation, measures aimed at improving collection and marketing, and an extension and diversification program. In 1979, the public company "Andapa Mamokatra" took over as the organization in charge o f the Andapa basin development project. The impact evaluation o f the project in 1998 was severe, particularly: (i)failure of pumping irrigation on the Ankai'be perimeter (2,100 hectares); (ii) lack o f maintenance o f structures on all perimeters developed by the project; (iii) total disorganization o f the AWUs; (iv) failure o f intensification attempts. Lac Alaotra Sahamaloto Irrigation Scheme 37. The Lac Alaotra watershed forms a vast depression o f around 1,750 kmz,with an average altitude o f between 750 and 770 m, surrounded by eroded hills. The lake (a Ramsar site) is shallow and surrounded by swampy marshes. It covers an area o f about 220 - 250 kmz(free water surface) and around 550 km2with surrounding marshes. The watershed serves about 80,000 hectares o f rice farms, o f which 30,000 hectares are developed. The watersheds are subjected to strong man-made pressure. Deforestation, overgrazing (with bushfires) and increasing pressure from rain-fed crops have seriously degraded the fragile soils on the slopes, already marked by numerous lavaka. The effects are silting-up o f beds o f rivers and dams, degradation o f derivation and protection o f facilities. 38. The history o f the zone i s marked by interventions o f the public company SOMALAC (1962- 1981) which constructed the irrigation facilities, and was responsible for extension, processing and marketing activities. Morphed into a socialist enterprise from 1982 to 1991, SOMALAC ensured the maintenance o f the irrigation system, supervised rehabilitation works carried out between 1984 and 1989, with notably the creation o f water users associations (1989-1991). These efforts were accompanied by the implementation o f projects aiming to intensify agriculture. 39. The watershed supplying Sahamaloto irrigation scheme stretches over an area o f 356 km2.The irrigation scheme has a developed area o f 6,400 hectares, o f which 80 percent is cultivated when the rainfall conditions are favorable. The area is supplied by a storage dam constructed in 1957. The initial storage capacity o f 26 million m3,was gradually reduced to about 13-14 million m3.The scheme was fully rehabilitated in 1988-1989, including the construction o f a new intake tower, an increase in the 37 volume of storage water to 18 million m3.Emergency repair and rehabilitation works were initiated in 1998-1999. 40. The 12 federated WUAs o f the irrigated perimeter, with a total o f 1,800 members, are physically participating in the construction o f secondary canals, thus contributing to the maintenance costs o f the primary system and operational costs o f the office o f the federation. Contribution in cash for the maintenance costs at the charge o f the WUAs (secondary systems) varies from one WUA to the other, but remains generally weak, with recovery rate rarely exceeding 60 percent o f amounts voted. E. Rehabilitationof hydro-agriculturalInfrastructuresin the Project Zones 41. The definition o f a priority investment program demands that rankingcriteria be defined for determiningpriority interventions. The following three levels are defined.Level 1interventions consist o f those works that would resolve problems that are o f capital importance to the entire area. The rehabilitation of infrastructures in this category helps to ensure: (i)access to water resources by protecting the headwork and primary structures that are indispensable for supplying the second system; (ii) to cultivated land by rehabilitating cultivated schemes during raining season lost access through dysfunctional drainage; and (iii) protection o f property, by protecting the structures against floods or a strategic structure. The non-intervention o f Level 1blocks the hnctioning o f the system. Hence, in most cases the interventions concern primary infrastructure: control dam and diversion offtakes, supply channels, main canals, main drainage systems, or flood protection dyke. 42. Level 2 interventions consist in structures that block access to water or access to land or protection o f assets o f part o f the network: secondary or upstream/downstream links. The non- intervention of Level 2 makes it impossible for part o f the users to cultivate or harvest. It concerns mainly secondary systems, sections o f the main canals or additional structures on the main canal (floodgates, control structures), secondary canals and secondarydrainage systems. 43. Level 3 intervention consists in structures that would boost agricultural production either by improving water control (irrigation and drainage), or increasing the cultivable area. It involves earth roads whose state hampers the marketing o f agricultural production, works on secondary canals, and eventually tertiary canals. 44. Table 1presents the estimated costs of rehabilitation works, including the Sahamaloto scheme o f Lac Alaotra. The costs are those borne by the contractor; the manual contribution o f the user i s not includedinthe estimates. 45. It is important not to focus solely on total amounts. Hence, the major budgetary allocations presented inthis table are as follows: (i) adding the Sahamalotoperimeter at Lac Alaotra, the total by budgetis tripled, from US$5.8 million to US$17.6 million; (ii) the three priority intervention zones for (Marovoay,. Itasy and Andapa), 65 percent concerns priority 1 works, 27 percent priority 2, and 8 percent priority 3; (iii) the other hand, for the Sahamaloto perimeter at Lac Alaotra, 71 percent on concerns priority 3 works; 29 percent priority 2, and 0 percent priority 1; (iv) for all possible intervention zones, 50 percent concerns priority 1 works, 28 percent priority 2, and 21 percent priority 3. 46. It should also be noted that the pumping stations in some o f the blocks in the Marovoay scheme, and their primary system, whose rehabilitation falls under priority 1, accounts for 50 percent o f the total rehabilitation budget for the Marovoay zone. 47. The project will not totally finance the rehabilitation o f works that the users should cater for in the future. The contribution of users will be equal to what they should pay infuture for O&M of these structures. In that regard, the envelope that the project will allocate to rehabilitation works will be calculated by deducting the annual amounts users should pay for management and maintenance in the future. 38 Table 1: Cost ofrehabilitation works onhydro-agricultural irrigation schemes Site Number of Surrounding Level 1 Level 2 Level 3 Total Works Perimeters area inha Worksin Works in Works in inmillion million million million MGA MGA MGA MGA Marovoay 13 21 290 2 755 1 640 549 4 944 Itasy 6 3 590 2 468 371 91 2 930 AndaDa 3 1650 200 281 53 534 39 Annex 2: Major RelatedProjectsFinancedby the Bankand/ or Other Agencies Projet de Mise en Valeur et de AFD Active Protectiondes BassinsVersants du Lac Alaotra(BV-Lac) Projet d'Appui a la FCdCrationd'AUE I AFD IActive AmCnagement des basinsVersants dans le Lac Alaotra DCveioppernent du Sud (ACCORDS) Projet de DCveloppement RuralSAHA Swiss Cooperation Active Eco-Regional Initiative (ERI) USAID Business and MarketExpansion 1USAID (BAMEX) ParticipatoryCommunity-based UNDP CEO approved Conservationin the Anjozorobe Forest Corridor Wind and Hydro Energy Market UNEP Proposed Development Projet d'Appui a la Valorisation des FA0 Active Bassins Versantset des PCrimetres IrriguCs 40 go m = 0 o m o o o m m o m o o o o o o o o o o o o o o o ~ m~ mo oo ~ m m ~ m or o~ ~v ~ o o o o o~ om o o o o) ~m ~ o ~ ~ ~ ) t t ~ o v o o ~ ~ e r ~ m m N v )~m ~ m) r - h ~ r v )mu a~~ o h tmm r n o ~m~ m ~ v m ~ r m ~ ~ -hr m am r r r m ~ I I 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 o o o o o o o o o o o o m o o o o o o o o o P o o o o o o o o o o ~ m ~ o o o ~ o0o ~ o v) In 0.0. 0.0.0.0. -. Y Y 0.-.30. 0.0.0.0.0.0- 9". 0" 0 m ~ m r m r ~0. m r t v ) ~ - o ~ ~ rno o p ~ ~ m 0 r r rr r N N N m r&9 r 0 'a - m 0 0 3 0 21 5 ~ m o ~ o m m o m o o o o o o o o o o o ooo o m o o o m m o m o o o m o o o ooov) O m m O O m m O m O O h t o O O 0 0 0 0 m v) m ~ t o N m ~ o h v ) m o ~ t o c v m o o N o m Ir r r N r r N r r N N P r r rrr r e I h s -.0. 0. '9 N O m N N r r r 0 0 O h N m - m N W I r S600363 S313IdX3dflS (44)S'33NVNIdSL3fOlId Annex 3: Results Framework and Monitoringand Evaluation 1. Performance indicatorsare linkeddirectly to the CAS goal ofpromoting broad-based social and economic growth, and inparticular (i) to reach an economic growth rate of 8 - 10 percent per annum; (ii) increase the level of investment to 20 percent; (iii) to promote the vitality of the private sector so that it participates in an investmentrate of 12 14 percent; (iv) to open up Madagascar's - economy to greater competitionwith aview to reducingcosts and improving quality; and (v) foster the willingness ofthe populationto participate. 2. Overall monitoring of the project's implementation, as well as assessing the development impact ofthe project would be the responsibility of the Departmentof Statisticsand Information (DSI) under MAEP. It will be supported by technical assistance. A specializedproject M&E / management information system will be prepared to the satisfaction of IDA, as well as procedures for data collection and reporting. M&E will be based on direct reporting by institutions involved in project implementation(MFI, MEF, ASC farmers and WAS),relevant data collected on a systematic basis for other purposes, participatory assessments, user satisfaction surveys (e.g., in irrigation schemes), income surveys, and targeted data collection (among others through satellitephotos), as establishedin the project implementationmanual. DSIwill commissiontwo evaluations ofproject output and impact indicators, at mid-term and at completion. The project will establish a baseline before or soon after project effectiveness. The output of the M&E would provide sufficient evidence in linking periodic and annual monitoring with subsequent annual project planning activities so that M&E data are interpretedand used as an instrumentfor projectplanning. 3. The project will establish Regional Monitoring Committees in each of the four project areas that will be chaired by the Head of the Region and made up of members of GTDR. The Regional Monitoring Committee will be supported by the GTDR's Technical Secretariat, and will be responsible for (i)ensuring consistency of project actions with project objectives and work plan, national strategy and policy, and regional development priorities and programs; (ii)preparing and validating detailed work plans and budgets at the regional level; (iii) reviewing project progress and performance, and the implementation of corrective measures if necessary. The Regional Monitoring Committeewill meet twice ayear. 4. MAEP will be responsible for submittingto IDA semi-annualprogress reports on the project. Progressreportswill focus on (i) keyperformance outcome, output and input indicators as indicatedin the project logframe; (ii)progress in procurement; (iii)progress in implementation works; (iv) progress on technical assistance and training; (v) status of disbursementsfrom the credit; (vi) progress on community sensitizationand mobilization, inparticular with respect to the Performance Contracts; (vii) work plan for the next six months. 5. An internal mid-term review will be conducted jointly by MAEP and IDA during the third year of project implementation. To facilitate this review, MAEP will prepare a mid-term evaluation and will summarize the findings in a detailedreport that will be submitted to IDA no later March 30, 2009. The mid-term evaluation and review would take stock of project implementation progress, constraints and recommendations for improvement, and would assess logframe indicators in the light of actual achievements on the groundandpropose improvements. 6. No later than 6 months after the credit closing date, MAEiP will provide to IDA a project ImplementationCompletion Report (ICR). The completionreport would include: original and revised project targets and actual achievements; project impact assessments focusing on results; and performanceofproject managementand IDA infulfilling their respectiveobligationsunder the credit. 42 A. ResultsFramework To sustainably improvethe living Increasedaverageproductivity of :onditions andincomesof rural irrigatedriceinthe project Year 1 : establishbaseline )opulationsinsix main irrigation areas (MTha): ;ites and their surrounding -- Baseline Endof Year 4 : confirm progress Watersheds, andthe managementof after implementation of iatural resources. Andapa 2.0 3.5 project activities, and Marovay 2.0 3.5 adjust intervention strategy Lac Alaotra 3.5 5.0 ifrequired Itasy 3.O 4.5 Year 12 :measureproject Increasedaverageproductivityof impact rainfedrice inproject areas (MTha): -- Baseline Endof Woiect Andapa 1.5 2.25 Marovay 1.5 2.25 Lac Alaotra 1.5 2.25 Itasy 1.5 2.25 nonrice areain irrigatedschemes as a percentage of overall cultivatedarea over two seasons increasedby25 percent increaseinareaunder production inirrigatedschemes duringthe dry To establishthe basisfor viable Disseminationof innovative irrigatedagricultureandnatural technologiesandequipment to Year 1: establishbaseline resourcesmanagementinfour main 30,000 beneficiariesthrough irrigationsites andtheir surrounding extension, capacitystrengthening Annually: confirm progress watersheds: (i) Andapa (Sava andtargeted cost sharing, after implementation of Region),(ii) Marovoay(Boeny Improvedmanagementof about project activities, and Region),(iii) ItasyRegion,and(iv) 21,780 ha of irrigationinfrastructure adjust intervention strategy Lac Alaotra (AlaotraMangoro throughinvestmentsin ifrequired Region). rehabilitation, trainingand institutionalreforms Improvedmanagement of about 8 sub-watershedsthroughcapacity strengtheningandinvestmentin watershed infrastructure Increasedgovernment support for agriculturalintensificationin irrigatedandrainfedareas through Improvethe environmental Increaseinlandareaunder Year 1: establishbaseline sustainability of landmanagement sustainablemanagementas a practicesinfour targeted watersheds percentageof baseline, intargeted Annually: confirm progress projectinterventionareas after implementation of Increaseinvegetationcover as a project activities, and percentageof baseline adjust intervention strategy ifreouired 43 Result 1: Developmentof Result 1: CommercialAgriculture FiveASC established Results 1-3 : :ntensification,marketing, and 50 OPs, unions and iiversification of selected agricultural federations of active lralue chains inprojecttarget areas. producershavingregistered APL 1:monitorprogressindicators with ASC on an annualbasis MatchingGrantsfully disbursed Endof project: 5,000 HHtrainedinagro- assess and adjust component ecological cropping strategy ifrequired. practices assess lessons for extending Result 2: Irrigation Development Result2: programat nationallevel Bettermanagementoftargeted 21,780 hairrigation area irrigatedschemes through rehabilitated infrastructurerehabilitation, 30 WUAs trained improvedinstitutionalframework, 100 percent of operation and capacitybuilding of Water User andmaintenancefunds Associations. coveredby irrigation service fees collected Four Performance Contracts satisfactory executed FERHA established Result3: WatershedDevelopment Result3: Enhancedcapacity of stakeholdersin FourWDP and eight the four watersheds to manage participatorysub-watershed naturalresources in sustainable managementplans manner. preparedandadopted 60 communitySLM groups trained and supported 145 hotspoterosioncontrol interventionsrealized Five guichets fonciers operational Result4: ProgramManagement Result4: Result4 : Use of Projectresources in 100percentunqualified Review financial audits on an annual compliancewith agreedobjectives financial andtechnical basis and procedures, and settingup a audits Years4 :TechnicalAudit and policy framework that is favorable to Nationalfertilizer strategy adjustments extendingthe programto the national and andlegalguidelinesfor level. implementationof seed policy implemented Effectiveoversight,monitoringof NIWMP incorporatedinto projectactivities, policy guidance and MAEP's mediumterm lessons learned expenditureframework Timeliness andadequacy of annualwork plansand reports (including M&E reports, expenditureand accountingreports) 44 4gricultural Value chains 5ASCs 4,000 private sector Development supported by established inall households investments in project identified in project areas trained in agro- agriculture all four sites 3,000 ecological increased as Training curriculum households technologies evidencedby inago-ecological trained inagro- Matching Grant disbursements technologies ecological disbursed60% underthe prepared technologies matching grant Regional partners Matching Grant mechanism; recruited disbursed 30% p ASCs TOR and business established and plans for ASCs operational in prepared in all sites the four project Matching Grant sites. operational. Irrigation TA for WUA 10 WUAs Performance e Scheme Development mobilization established and contracts signed Development recruited trained in all inall four sites Plans and Scheme four sites Recruitment of Performance Development Plans Recruitment TA contractor for Contracts (as part of WMP) technical studies rehabilitation executed prepared in all four Technical O&Mfee satisfactorily. sites studies recovery in Acceptable Maintenance costs completed in all accordance with institutional study conducted in four sites PC mechanism for all four sites Inventory FERHA the funding of FERHA study transferable established non- completed infrastructure 20WUAs transferable completed established and irrigation Legal trained in all infrastructure framework four sites (FERHA) revised established and operational; Watershed SLMgroups Watershed Participatory guichets Development established Development sub-watershed fonciers Watershed Plan (as part o f management established and Development Plan WMP) adopted plans adopted operational in (as part o f WMP) inall four sites SLMgroups the four project study launched in Participatory trained in all sites. all four sites sub-watershed four sites Watershed Regional partners management according to Development recruited plans developed curriculum Plans executed inall four sites erosion control satisfactorily Training interventions curriculum for realized in all SLM groups four sites in developed accordance with 3 guichets Watershed fonciers Master Plan established. 4guichets fonc iers established 45 6t,2 Q 8 .-8 4Y 'E I 5L 3P 3 s 3 4 IF- 2 I N 4 s3 0 m d s 0 0 s0 N g 2 s g 3 0 + 7 2 0 o o v \o 2 - 0 0 3 N 3 d W s O N d c o r n '", 3 2 - 0 3 0N s a v - - 0 10 0 o w - N 4- -- 0 7 7 0 0 o o c s 0 Annex 4: Detailed Project Description ProjectObjective, Outcomes and Components 1. Theproject development objective is to establish the basisfor viable irrigated agriculture and natural resources management in four main irrigation sites and their surrounding watersheds: (i) Andapa (Sava Region), (ii) Marovoay (Boeny Region), (iii) Itasy Region, and (iv) Lac Alaotra (Alaotra Mangoro Region) . 2. The expected project results include (i)dissemination o f innovative technologies and equipment to 30,000 beneficiaries through extension, capacity strengthening and targeted cost sharing, (ii)improvedmanagement ofabout 21,780 ha of irrigation infrastructure through investments in rehabilitation, training and institutionalreforms, (iii) improved management o f about 8 sub-watersheds through capacity strengthening and investment in watershed infrastructure, and (iv) increased government support for agricultural intensification in irrigated and rainfed areas through increased public expenditures. 3. The global environmental objective o f the project i s to improve the environmental sustainability o f landmanagement practices in four targeted watersheds. The interim results are (i) 20 percent increase in area o f land under sustainable management in targeted project intervention areas (as a percentage o f baseline), and (ii)15 percent increase in vegetation cover (as a percentage o f baseline) 4. The project concept is based on the following principles: (i) responsibilities for each o f clear the actors in the management o f irrigation schemes and surrounding watersheds (farmers, water users, populations and their associations, communes and inter-communities, regions, central government); (ii)effective participation of the population (male and female) and all stakeholders (including vulnerable groups) in the diagnosis o f problems and identification o f options; (iii) co-management of irrigation schemes and watersheds by all the actors concerned; and (iv) adequate incentive systems and efficient mechanisms to ensure that all respect their commitment. 5. The proposed project includes three technical components covering three strategic orientations: (i)development o f commercial agriculture, (ii)irrigation development, and (iii) watershed development. A fourth component i s related to program management. Information o f GEF funded activities within the components can be found inthe Incremental Cost Analysis (Annex 15). In accordance with the "integrated rural poles" approach, the project proposes four similar subprojects in the four regions involved: Andapa, Marovoay, RCgion Itasy, and Lac Alaotra - the Sahamaloto irrigation scheme (Annex 1). 6. The four sites have been selected based on their accessibility, availability o f agricultural services and potential for increased productivity through improved water management. At the same time, public irrigation schemes are characterized by serious institutional weaknesses, lack o f clarity with respect to roles andresponsibilities o f stakeholders, andwatershed degradation. Component1: Developmentof CommercialAgriculture (US$12.68 million, including an IDA contribution of US$7.45million, a GEFcontribution of US$2.72 million, and a beneficiaries' contribution of ~ $ 2 . 5 million) 1 7. The objective for this component is to lay thefoundations for improved market access and sustainable intens@cation and diversification of irrigated and rainfed agriculture in the project S watersheds. 48 8. The `Development o f Commercial Agriculture' component involves the project area as a whole: irrigated schemes and uplandor tanety areas. Inuplandareas, it is part o f a coherent framework which is `Watershed Development' proposed in subcomponent 3.2. Its specific objective will be achieved through an approach focusing on market-driven demand, agricultural technology development and dissemination, initiative by private operators and vertical integration and coordination o f selected supply chains by promoting partnerships among actors, including public private partnerships (PPP). 9. The component aims at improving, all along the targeted supply chains: Access to market and marketing systems inorder to reduce costs and increase farm gateprices Addedvalue through diversification into higheraddedvalue products andagro-processing Capacities o f farmers, farmers groups andprofessional organizations 0 Agricultural productivity through better access to extension, improved technology, inputs, and credit 10. The estimates o f targeted areas in terms o f rice intensification and sustainable diversification inrainfed production(agro-ecological, etc.) anddry season (including private irrigation) are presented intable 1: Table 1: targeted areas in terms of rice intensification and sustainable diversification in rain fed production (agro-ecological, etc.) and dry season (including private irrigation) I PI (ha I Rain f e d Agro- I Counter-season I Targeted 21,500 I /a: Rehabilitatedphysical areas (see RDC-IRAMstudy) x use intensity.RBME= rice with goodwater control; WIVE= rice with poor water control. /b: Irrigation infrastructure rehabilitationis focusing on Sahamaloto, but activitiesto promote agriculturalproductiontarget a larger area (8 communities), including Anony andapart of Amparafaravola. 11. Direct beneficiaries from the agricultural development component are presentedintable 2: 12. Intermediate results are (i) established ineach o f the four sites, (ii) ASCs increase by 50 o f the number of POs, unions, and federations of active producers who have registered with an ASC, (iii) 5,000 households trained in agro-ecological cropping practices, and (iv) increase in private sector 49 investments in agriculture as evidenced by full disbursement o f the matching grant. GEF will contribute to training and support o f households engaging in agro-ecological cropping practices and participate inthe matching grants for agroecological and agroforestry activities. 13. Critical risks include: (i)capacity among producers and their organizations to meet technologies supply and to and to manage the support-guidance scheme; (ii) will among private the operators to invest directly in long term contractual relations with agricultural producers; and (iii) maintaining and strengtheningincentive policies from the State infavor o f agricultural private sector. Sub-component 1.1:Support to agricultural services 14. The aim in this subcomponent i s to lay the foundations for the development o f commercially oriented agricultural production by implementing innovative technologies for production, storage and processing o f agricultural products, by improving access to markets, and by supporting the development o f commercial agricultural supply chains. Investment under this subcomponent will be targeted at improving the enabling environment and providing incentives (in addition to on-demand support to investment projects by private initiative to be funded under subcomponent 2). This includes the promotion of sustainable and profitable agriculture on hillsides (for example, ago-ecological and agroforestry techniques). The project will take a gender sensitive approach and will also specifically support vulnerable groups in their demands The project will finance the services, work, equipment, training and operational costs o f suchpublic investmentand o f the activities corresponding to the core public responsibilities. Activities will be adjusted to specific needs on each site, and may includethe following: (0 Support to the development of commercial agricultural supply chains. The project will recruit for each site one or several professional service providers for promoting market-driven supply chains. The project will as much as possible use the existing schemes for supporting the private sector and agribusiness which are already operating in Madagascar, such as the network o f "business centers" set up by the BAMEXproject and/or interprofessional technical support centers, such as CTHT and CTHA. Such service providers will be responsible for the following activities : (a) market researchand surveys for national and export markets, as well as thematic studies in storing, processing, packaging, post-harvest treatment and quality management, (b) FUD on improving technical itineraries for production, conservation, and valorization, (c) helping eligible operators prepare documents for submission o f sub-projects to the matching grant mechanism and to the banking system, and (d) developing partnership contracts between producers and operators for the marketing and processing o f targeted products. (ii) Strengthening the capacities of farmers and professional organizations, as well as the establishment o f agricultural service centers (ASC). The project will aim to buildprofessional and institutional capacity among farmer organizations (OPA, GIE, TT, etc.), and their federations. The project will finance the establishment o f ASCs as an interface, at district level, between supply and demand to support the provision o f on demand advisory and extension services. Each ASC will include a small technical team and platform (decision making unit) grouping farmer organizations, the private sector, the government, the local authorities and the regional partners at district level. The project will finance civil works for office rehabilitation, equipment and travel costs, training and ASC operating costs (staff and operating expenses), for the 5 ASCs in the BVPI area, as well as operating costs for the platform. (iii) Strengthening the supply of technologyfor production and valorization of agricultural products, in particular technologies geared at promoting intensification o f rice cultivation on irrigation schemes, promoting the adaptation o f ago-ecological cultivation techniques to sustainable rainfed production systems and diversification o f production systems for targeted and priority supply chains, including livestock production. The project will support: (a) service providers for adaptive research and dissemination o f improved technologies identified as 50 priorities by the partners, and (b) the strengthening o f capacities o f regionalpublic services for seed quality and phyto-and zoo-sanitary control. A distinction will be made between (a) the more productive land at the bottom of the hillsides that lends itself more easily to intensification compared to some o f (b) the traditional agricultural upland systems that depend on slash-and-bum practices (tavy). These uplandsystems, found inmarginal and remote areas of the upper watersheds, are often based on deforestation, thus threaten biodiversity, degrade soil productivity quickly due to burning practices and short fallow periods, and contribute to erosion. Often these farming practices do not allow farmers to achieve satisfactory incomes. However, it i s possible to develop sustainable agricultural production systems that can be productive and profitable (e.g. through ago-forestry, ago-ecological and horticultural techniques). The improvement o f these systems will need more time and effort than for the systems downstream, and needs intensive on-farm technology development work in order to develop sustainableand profitable farming practices. 15. The project main implementingbody will be DRDR. Detailed implementation modalities for each activity group insubcomponent 1are specified inthe table 3. Table 3: Detailedimplementationmodalities Subcomponent Implementation Development of sustainableandmarket-driven Recruitmentofregionalpartnersby DRDR supplychains Capacity-buildingof producersand support to Recruitmentof service providersbyASC producers organizations Appliedresearchandtechnology dissemination Recruitmentof one or several serviceproviders (FOFIFA,TAFA, ONG, etc.), ina competitiveway and under contractwith DRDR Sub-component 1.2: Support to PrivateInvestment 16. The specific objective in this subcomponent is to link, extend and upscale the incentive and promotional activities financed under subcomponent 1. This will be achieved through support o f on demand private investmentsby operators, farmers and farmers organizations at all levels o f the supply chains. To this end, the project will finance, through a matching grant, individual or collective initiatives and sub-projects as presentedintable 4. Table 4: individualor collectiveinitiatives andsub-projects Support to marketingchains Market surveys, supply chain analysis, development of quality and certification managementsystems; commercial/markettrials Infrastructurefor grouping, storageand post-harvesttreatment Integratedprojects for setting up contract-based agriculture systems to the benefit of smallscale producers Support to input, credit and Establishing/extending networks for distributing inputs and equipment providers equipment ; Technical and management advisory services (for example, technical andmanagerialcapacitybuildingfor seedproducers). Technical support andextension of microfinancenetworks Technical support for the development and implementation of new products (e.g.,weather insurance) Support to productive Adaptive, agricultural, and ago-industrial research (varieties, investment technologies andproductionandprocessingequipment) ; Introduction, dissemination and on-farm development of new agricultural production techniques (agroforestry and agro-ecological techniques, etc.); Awareness raising and demonstration campaign (inputs, equipment, etc.) Rehabilitationidevelopmentof quality seedproduction; Reforestationand improvementof degraded soils. 51 private -The 17. implementation modalities o f the cost sharing mechanism for financial assistance to individual or collective - investments corresponding to the broad objectives o f the BVPI project will be outlined in the implementation manual. The manual will include a list o f eligiblehon eligible activities, selected on the basis o f their potential contribution to project/government objectives. Eligible activities will clearly relate to agricultural production and management o f natural resources sub-projects that are presented by beneficiaries, and co-financed exclusively incash, either under own capital, or under micro credit. Project contribution will range from 20 to 80 percent o f total cost, depending on the public good nature o f the investment and to the degree o f poverty o f the beneficiaries. Proposals will be selected by a decision making body at regional level (Comitd de Sdection, to be set up within GTDR). This committee will be in charge o f approving requests for subsidy(see Annex 6, andthe Project's ImplementationManual). Component 2: IrrigationDevelopment (US$17.47 million total, IDAfunding US$15.67 million, and a beneficiaries' contribution of US$1.80 million) 18. The objective o f this component is to lay the foundations for improved management, maintenance and sustainability of irrigation servicesprovision infour large-scale irrigation schemes through rehabilitation of irrigation infrastructure, capacity strengthening of stakeholders and clarification of roles and responsibilities, and establishment of an appropriate incentiveframework. 19. The component will contribute to the overall project objectives by improving irrigation services thus creating a more favorable environment for use o f inputs, agricultural technologies and marketing. In so doing, the project will initiate a virtuous circle o f increased productivity, improved capacity to pay for O&M, and improved irrigation service provision. The project will adopt a contractual approach that empowers stakeholders and clarifies their respective roles, and that will be based on the principle that investments in infrastructures enhance and at the same are conditioned by stakeholder performance. The instrument for clarifying and formalizing commitments and responsibilities is the annual Performance Contract (PC) that will be signed between (F)WUAS, the Communes, Regions and MAEP. Contracts signed with service providers will systematically include a payment provision that will partly depend on achievement o f agreed performance indicators. Irrigated areas targeted for rehabilitation and beneficiaries are presented intable 5. Itasv 5.660 5.000 Sahamaloto I 6,400 5,000 Total 21,780 19,500 of average irrigation landholding 20. Critical assumptions include that stakeholders are willing to pay for better irrigation service provision, and that a more reliable access to water leads to higher agricultural productivity which in turnleads to an improved capacity to pay. The main risks are that stakeholders are not willing or able to respect terms and conditions o f the PC and that the project will not sign subsequent PCs. Positive environmental externalities include reduced population pressure on fragile lands and therefore reduced land degradation, erosion and sedimentation through intensification. Negative environmental and social impacts include increased use o f agro-chemicals and limited displacement following construction o f irrigation infrastructure. 52 21. Intermediate results are (i) rehabilitation o f 21,780 ha o f irrigation scheme, (ii) capacity strengthening o f 30 Water Users Associations, (iii) four Performance Contracts signed in each o f the sites, and (iv) fully satisfactory execution o f Scheme Development Plans and Performance Contracts, including payment o f O&M fees as agreed. Sub-component 2.1:Support toIrrigation Development 22. The project will finance technical assistancefor the mobilization and capacity strengthening o f stakeholdersinfour large-scale irrigation schemes. This will include: (i) Participatory preparation o f a Scheme Development Plan (SDP, as part o f the WMP) that identifies a long-term vision on the performance o f the scheme, proposes specific measures to achieve this vision, and agrees on associated investment, maintenance and management costs. The SDP will also identify non-transferable infrastructure, outline a planning for implementation o f proposed investments and rehabilitation works (if any), and will identify roles andresponsibilities for each o f the stakeholders. (ii) Participatory preparation o f a PC that will be signed between (F)WUA, Commune, Region and DRDR. The PC will identify commitments o f each of the stakeholders, and provide an annual work planfor their implementation, as well as performance indicators. Communes and (F)WUAs will prepare a seasonal list of all water users in the scheme. Commitments may include, (a) for (F)WUAs: agreement on irrigation management transfer o f transferable infrastructure, full recovery o f O&M charges sufficient to finance sustainable O&M o f the scheme; recruitment o f an executive director and o f a fee recovery agency; establishment o f a relief committee to decide on requests for nonpayment o f O&M charges; and increased land tax for irrigation schemes; (b) for Communes and Regions: active support for the compliance with rules and regulations of (F)WUAs; rehabilitation of critical inter-communal roads; and payment o f shortfall in O&M fee collection; and (c) for DRDR: implementation o f rehabilitation works agreedinthe SDP; maintenance o fnon-transferable infrastructure through FERHA; and support for the compliance with rules andregulations of (F)WUAs. PCs will also include provisions for the signing o f an Irrigation Management Transfer Agreement between DRDRand (F)WUAsthat will specify roles andresponsibilities. (iii) Monitoring o f agreed performance indicators, annual evaluation o f the PC and design and implementation o f an annual user satisfaction survey. (iv) Mobilization and capacity strengthening of stakeholders in the implementation o f the PC, including, (i)for (F)WUAs: preparation o f an annual budget and work plan; contract management; evaluation o f the preceding year's results; decision on bonus o f executive director; and development and implementationo f a strategy for the recovery o f O&M fees; (ii) for Communes and Regions: preparation and implementation o f SDP and PC; and support for the mobilizationo fwater users. (v) Studies, includingone to determine and update O&M charges in each o f the four sites, and the nature and scope o f cyclone damage to irrigation schemes. The study would provide guidance on what farmers can reasonably be askedto pay interms o f O&M and the annual contribution to the CP. (vi) Support for the harmonizationo f irrigation related legislation (including arrkttd 4292-97,4293- 97, 1366-98, 1708-97, decree 90-642, and the law 90-016) in view o f the Irrigation and Watershed Management Policy Letter. Specific issues that need to be addressed include the management o f autonomous schemes and partner schemes, the process for irrigation management transfer andthe membership o f W A S . 23. The project will also finance office equipment and office running costs, vehicles and their running costs, workshops, training and study tours. Additional support will be provided in a 53 competitive way, Le., based on how ambitious the targets are that have been agreed in the PC (e.g., relatedto O&M cost recovery), and on their achievement. 24. Specific outputs of the mobilization sub-component include an SDP and four annual PCs, annual evaluations of the PCs, capacitystrengthening and anindication on capacityto payby farmers. 25. The mobilization and capacity strengthening of stakeholders will be outsourced to a national consultant with support from an international consultant. One contract will be signed for the preparation of an SDP and a PC, and capacity strengthening of and support for each of the stakeholders inthe implementationof SDP and PC. Annual evaluation of the PC will be done on the basis of agreed indicatorsby a PC evaluation committee. The SDP will be prepared together with the Watershed DevelopmentPlanas part ofthe overallpreparationofWatershedMaster Plans. 26. Procurement packages will include recruitment of international consultants' services for the preparationof SDP and PC, mobilization and capacity strengthening of stakeholders, and studies. Sub-component 2.2:Irrigation Investments 27. This component will finance the implementationof those rehabilitation works that have been agreedinthe CP. This will includethe followingactivities: (i) technical detailed design studies of the works that have been agreed in the PC, including the preparation of bidding documents. Works will be designed so that they will resist future cyclonedamage. (ii) rehabilitation of irrigation and appurtenant infrastructure(inter-scheme roads) as agreed inthe PC on 21,780ha ofirrigation schemes. The work will be outsourced to national contractors (iii) supervision of rehabilitation works. The work will be outsourced to the consultancy firm responsible for the preparationofthe detaileddesign studies. 28. Specific outputs of the irrigation development component include detailed design studies and bidding documents, rehabilitation works, and capacity strengthening. 29. The project will also promote individual low-cost irrigation technologies. These may include treadle pumps, drip irrigation kits, motorized pumps and other individual irrigation technologies. Support will include product development, mass-marketingcampaigns, and capacity strengthening of farmers and local manufacturing workshops. No support will be provided for the purchase of the equipment. It is expectedthat farmers request credit throughmicro-finance organizations. 30. Preparation of detailed design studies and supervision of rehabilitation works will be outsourced to a national consultancyfirm for the full duration of the project. The consultant will also be responsible for works supervision capacity strengthening of the DRDR. Rehabilitation works will be outsourced to a national contractor. In order to simplify procurementprocedures, the project will sign a multi-year framework agreement with a contractor in each of the four sites that will determine unit prices and aprocess for annualprice amendments. 31. Procurement packages will include (i)recruitment of national consultants' services for the preparation of detailed design studies and supervision of rehabilitation works, (ii) recruitment of national contractors for the implementation of rehabilitation works, and (iii) promotion of low-cost affordable irrigation technologies. 54 Component 3: Watershed Development (US$4.33 million, including IDA funding of US$1.82 million; GEF contribution of US$2.42 million, and beneficiaries contribution of US$O.09 million) 32. The objective of the component i s to lay the foundations for sustainable management of watersheds including irrigated and rainfed agriculture, the conservation of the natural heritage, and improvedproductivity of the natural resources. 33. A participatory and integrated approach to sustainable land management should encourage local population (male and female) to take responsibility and engage inthe sustainablemanagement of their natural resources. The component aims to contribute to (i) protection of watersheds by the reducing erosion and sedimentation; (ii) increased productivity and sustainability of upland systems (including agro-ecological, agroforestry, forestry, andpastoralsystems), (iii) improvedmanagement of naturalresourcesto generate environmentalbenefits, (iv) improvedaccess to land and user rights. 34. Critical risks include (i)farmers may be hesitant to participate in activities outside their own fields, as they fear not to directly benefit from environmental improvements.Where possible, on-site improvements that produce upland and lowland benefits are promoted (which are expected to be numerous due to advanceddegradation status ofthe land). Inaddition, other incentives suchas support to land tenure security will be favored. Only in cases with a distinct disconnect between upland and lowland activities, the project may seek to pilot other available and innovative incentive systems (e.g. payments for environmentalservices). The project will remainflexible with the responsedepending on the analysis and the feasibility of implementing the various solutions; (ii) handing over of land the rights to local community groups could be perceived by some as threat to free access to natural resources. The project will establish and strengthen communication and negotiation platforms. By forming networks of community groups, local communities will be ina stronger position to withstand outside interference. 35. Intermediate outcomes are (i) watershed development plans (as part of Wh4Ps) and eight four participatory sub-watershed management plans developed and adopted, (ii) SLM groups trained 60 and supported (including the support for 32 contracts of delegated land use rights (GELOSE) provided, (iii) number of hotspot erosion control interventions realized (100 small, 40 medium, 5 large), and (iv) five guichetsfonciers operational. GEF will finance the participatory sub-watershed managementplans, andthe training of and support to the 60 SLMgroups. Subcomponent 3.1: Support to WatershedManagement 36. The watersheds in the four project zones are very different in terms of geography, climate, biodiversity, population density, land use, productive potential, ongoing development programs, availability of potential partners, etc. The following description of the component and the various activities is an overall description.The project will adopt a flexible approach that will allow modifying activities according to needs, on-going programs and collaboration potentials with partners who are already working inthe project areas. 37. Planningofwatershed managementis done inthree steps: (i) The first step i s preparing a watershed management plan for the watershed areas adjacent to the irrigation schemes in the in the four project zones (about 400km2 for Sahamaloto/Lac Alaotra, 500 km2for Itasy, 1,000 km2for Andapa, and 500 km2for Marovoy). The "large" irrigation schemes consist of groups, clusters or sectors of schemes, each associated with a 55 sub-watershed.The WSM planwill cover all the sub-watersheds that are directly associatedto the irrigation schemes'. This will be donewith anadditional 0.77 million US$through aPPF. (ii) ThesecondstepinvolvesthedevelopmentofparticipatoryWSMplansfortheapproximately eight sub-watersheds associated with the irrigation schemes covering an area of between 10km2to about 500 h2. (iii) Thethirdstepreferstotheparticipatoryplanningofsub-basindevelopmentandmanagement within larger watersheds, which will beundertakenbyuser associations of local communities. 38. The project will finance the following activities: (9 At the level of watersheds - the project will finance technical assistanceto prepare one WSM plan for each o f the four project zones, which will include: 0 Zoning and description o f land use systems, ecosystems, settlements, institutions and partners; Strategic analysis o f erosion problems (as the main source o f downstream sedimentation) and o f natural resource degradation; A specific and detailed analysis to identify responsibility for the implementationofproject activities, while takinginto account existing partnersinthe area; 0 Establishing a baseline for monitoring and evaluation o f component results. (ii) Atthesub-basinlevel-theprojectwillfinancetechnicalassistancetofacilitatepreparation Of: 0 a participatory zoning o f sub-watersheds to determine the optimal land use according to (a) topography along a gradient from downstream to upstream, (b) current land use and land rights, (c) diagnosis o f soil fertility and soil production potential, (d) location and characteristics o fwater sources and streams, and (e) origin andpathways o f erosion, and 0 Participatoryplans for sustainablesub-watersheddevelopment andmanagement. (iii) Supporttoexistingcommunicationandnegotiationplatforms withtheaimto Involve stakeholders and partners (communes, farmer organizations, NGOs, etc.) in information exchange and communication; Discuss, negotiate, andvalidate participatory WSM plans; 0 Negotiate conflict settlement; Support o f environmental platforms inthe project areas. Training and capacity strengthening o f SLM groups, and o f local and regional staff in, among others: Environmental awarenessraising campaigns for local communities; 0 Training andor strengthening o f farmer organizations innatural resource management by providing technical assistance for instance for example, for cattle herders or charcoal makersand their associations; 0 Specific training to local and regional staff (NGOs, technical government services) in techniques that are required for the implementation of the component, such as participatory planningmethods or ago-ecological techniques. Improvement of land tenure security: The project will contribute to the implementation of the National Land Tenure Program (PNF) and finance the setup o f (inter)communal land tenure windows in charge o f the following activities: (i)recording the acknowledgement o f `non titled property rights' and land tenure transactions (inheritance, sale, transfers, etc.); (ii)regularizing landrights; (iii) securing secondary rights(sharecropping andtenant farming) in particular on PI and negotiated agreements (GELOSE) for sustainable management of resources on some key watershed space. The project will subcontract the implementation o f four land tenure windows (one in each intervention area) in close consultation with the PNF. The exception i s Lac Alaotra area inwhich the projecttargets one single scheme, so one single sub watershed ina group of irrigatedschemes andawatershed of about 1,800 km2. 56 The project will also support communities in obtaining community-based land rights (e.g. GELOSE) and will provide for technical assistance to support the preparation o f natural resourcesmanagementplanswithin the framework o f GELOSE. Subcomponent 3.2:Investment in watersheds 39. Depending on the WSM plans that have been prepared, a menu o f investments eligible for project support will be prepared, and specific conditions (positive and negative list) will be prepared, from which local populations may select investments they consider appropriate for their specific needs. Inprinciple, investments with long-term environmental impacts, and community based groups or associations will be eligible. Specific eligibility conditions will include co-financing (in kind or in cash), institutional capacity among groups, and the confirmation o f social and technical validity o f the proposals. Additional support will be provided ina competitive way, Le., depending on the targets that stakeholders agree to set for themselves, and the level of their achievement. 40. The project will finance the following activities: (i) Strategic erosion control. Erosion "hot spots" will be identified through strategic and participatory analysesconducted under subcomponent 1. Through negotiations, local strategies will be developed for controlling erosion, arresting gullies and reducing the quantity of sediments transported to downstream irrigation areas. The project will finance the establishment o f strategic anti-erosion works including though works, and biological methods and techniques. Works will be built favoring use of local manpower. In principle, W A Sin irrigated schemes should participate in planningof erosion control measures and should pay part o f costs. Many o f these strategic anti-erosion works will actually be part o f the irrigation investments. Examples are: construction o f retention structures (fascines) in combination with vegetative interventions for halting gully and lavaka erosion; and revegetation and protecting river banks and planting o f anti-erosion hedges (vetiver, fodder crops, andmultipurpose shrubs). (ii) The project will finance, through matching grants, all aspects o f reestablishing vegetation cover to reduce erosion to improve the land use productivity of the upper watersheds and to support the communities in an improved management o f land under secured land tenure arrangements(GELOSE): (a) Improved pasture management, including the cessation o f fire use, planting o f fodder grasses and fodder banks, establishment o f drinking points for cattle, rotational grazing, andkeepingcattle instablesfor manurecollection; (b) Awareness raising campaigns that address destructive traditional practices such as fire use for pasture and agriculture, and providing support in developing technical alternatives with a participatory approach. (this will be complementary to activities conducted underEnvironment Program (EP3)); (c) Reforestation and revegetation o f degraded land, including the restoration o f natural vegetation, support to community or private reforestation; (d) Provision o f support to protect natural forests and its biodiversity, and natural habitats such as marshesand lakes. Component 4: Program Management (US$4.31 million, including IDAfunding of US$3.45million; GEFcontribution US$0.86million) 41. The objective o f this component i s to manage and use resources in accordance with the project's objectives and procedures, and to put in place a policy framework that is favorable to upscaling of theproject at the national level. 42. Intermediate results include (i) financial and technical audit reports are unqualified, (ii) all national strategy on fertilizer supply and legal guidelines for the application of new seed legislation 57 adopted and implemented, (iii) Program BV/PI incorporated into MAEP's medium term expenditure framework. GEF will support project management and monitoring and evaluation to assure that the environmental global objective i s well mainstreamed inthe project. Sub-Component 4.1: Project Management 43. This sub-component would manage the project, by providing technical assistance, training, ofice equipment and vehicles, minor ofice upgrading works, auditing and evaluation studies, and incremental operating costs insupport o fproject management. 44. The sub-component will carry out overall project planning, quality oversight, procurement, financial management, and monitoring o f project activities. It will also contract out quality oversight through independent financial and technical audits, and evaluation o f project activities. The sub- component will also design and implement a communication strategy to disseminate core project messagesto beneficiaries andpartners o f the project. 45. Project management will encompass all four target watersheds as well as national level coordination. Sub-Component 4.2: Policy Support 46. This sub-component would provide technical assistance, studies, training, information campaigns, cross visits and workshops for the development of major nationalpolicies, regulations, and plans considered critical to the Government's National Irrigation and Watershed Management Program. These are expectedto include, among others: National strategy on fertilizer supply adopted and implemented; 0 Legislation andpolicy for privatization o f seed centers, and support to seed certification; Norms and standardsfor key export markets (particularly rice); 0 Sustainablefinancingo f watershedmanagement and irrigation maintenance; Feasibility studies to expand the national programto new watersheds. 47. The sub-component would also provide initial technical assistance support to emerging professional groups, inparticular the Platjiorme Consultative de Riz and the Association Malgache de Producteurs de Semences. 48. The scope o f this sub-component would be national. The improved policies are expected to benefit all key distributors andproducers involvedinthe sub-sector. 58 Sub-component 4.3: Monitoring and Evaluation 49. This sub-component would provide technical assistance and capacity strengthening to the Department o f Statistics and Information (DSI) under MAEP that will be responsible for project M&E and assessment of development impact. This will be done on the basis of a specialized project M&E / managementinformation systemthat will be preparedto the satisfaction o f IDA, as well as procedures for data collection and reporting. Inits collection o f relevant data, DSI will depend on direct reporting by institutions involved inproject implementation (MFI, MEF, ASC farmers and WAS),systematic data collection for other purposes, participatory assessments, regular user satisfaction surveys (e.g., in irrigation schemes), income surveys, and targeted data collection (among others through satellite photos), as established in the project implementation manual. The data that will be collected and monitored include those presentedinAnnex 3, 50. DSI will commission two evaluations o f project output and impact indicators, at mid-termand at completion. This sub-component will establish a baseline before or soon after project effectiveness. The output of M&E would provide sufficient evidence in linkingperiodic and annual monitoring with subsequent annual project planning activities so that M&E data are interpreted and used as an instrument for project planning. In addition, the outcome of user satisfaction surveys will form an input into the determination o f any merit payments to consultants providingtechnical assistanceto the (F)W A S , 51. Regional Monitoring Committees will be established ineach o f the four project areas that will be chaired by the Head o f the Region and made up of members of GTDR. The Regional Monitoring Committee will be supported by the GTDR's Technical Secretariat, and will be responsible for (i) ensuring consistency of project actions with project objectives and work plan, national strategy and policy, and regional development priorities and programs; (ii) preparing and validating detailed work plans and budgets at the regional level; (ii)reviewing project progress and performance, and the implementation o f corrective measures if necessary. The Regional Monitoring Committee will meet twice a year. 59 Annex 5: Project Costs ProjectCost By Component and/or Activity Local Foreign Total US$million US $million US$million Component 1: Developmentof CommercialAgriculture - 9.90 1.69 11.59 Component2: IrrigationDevelopment 11.78 2.08 13.86 Component3: Watershed Development 3.41 1.17 4.58 Component4: ProgramManagement 2.58 1.15 3.73 PPF 0.59 0.59 Total BaselineCost 27.67 6.09 34.35 PhysicalContingencies 1.38 0.37 1.75 PriceContingencies 4.00 0.30 4.30 Total ProjectCostsa 33.64 6.76 40.40 Interestduring construction -Front-endFee Total FinancingRequired 33.64 6.76 40.40 ProjectCost By Componentand Financier IDA GEF Communities Borrower Total Component 1: Developmentof 7.45 2.72 2.51 12.68 CommercialAgriculture Component2: IrrigationDevelopment 15.67 1.80 17.47 Component3:Watershed Development 1.82 2.42 0.09 4.33 Component4: ProgramManagement 3.45 0.86 4.31 PPF 1.61 1.61 Total FinancingRequired 30.00 6.00 4.40 40.40 'Identifiable taxes and duties are US$4.68 million, andthe total project cost, net of taxes is US$35.72 million. Therefore, the share of project cost net of taxes is 88 percent 60 Annex 6: Institutional and Implementation Arrangements A. Project Implementation 1. The project concept i s based on the following principles: (i) responsibilities for each o f clear the actors inthe management of irrigation schemes and surrounding watersheds (farmers, water users, populations and their associations, communes and inter-communes, regions, central government); (ii) effective participation of the population inthe diagnosis o fproblems and identification o f options; (iii) co-management of irrigation schemes and watersheds by all the actors concerned; and (iv) adequate incentivesystemsandefficient mechanismsto ensurethat all respecttheir commitment. 2. The project will be implemented at four levels: national, regional, Intercommune/district and local. 0 National. MAEP will be responsible for the overall implementation of the project, in full consultation with the other Ministries at the national level that are involvedin order to ensure that project activities are consistent with national policies. 0 Regional. The DRDR are responsible for the implementation o f a large part o f project activities. The Region i s the operational level that ensures (i) coherence and planning o f the project activities, and (ii) implementation o f certain support or investment activities (e.g., rehabilitation o f large irrigation schemes) at the level o f the four project sites. 0 Inter-commune/District.This is the level responsible for the implementation o f those activities that require collaboration at the intercommunal level (e.g., management o f watersheds and largeirrigation schemes, ASC, guichetsfonciers). 0 Local: Main level for the implementation o f the project at the level o f grassroot communities and economic operators. 3. Detailed Project Manuals, including a Project Implementation Manual, a Project Monitoring and Evaluation Manual and a Project Administrative, Financial and Accounting Manual, have been preparedduringthepreparation oftheproject. B. ImplementationArrangements SteeringCommittee and Guidance 4. A National Steering Committee (NSC) will be established comprising of representatives, at the SG level, of the other ministries concerned (Ministryof Decentralization and Land Use Planning, Ministry o f Environment, Water and Forest Resources, Ministry of Economy, Finance and Budget), the Chairperson ofthe Permanent Steering Team of the Rural Development Action Plan, andthe main professional organizations (Chamber o f Agriculture, and associations/ forums involved in the main value chains such as the "Rice Platform"). The NSC will ensure coherence o f the project activities with national policies under the National Irrigation and Watershed Management Program. The NSC will be chaired by the SG of the Ministry of MAEP. The NSC will be responsible for (i) review, approval and consolidation o f regional annual programs, work plans and budgets, (ii)monitoring implementation and results, including in particular the analysis and approval o f activity reports and financial and operational audits, and (iii) recommending corrective measures that may be necessary. The NSC will meet twice a year. The NSC will be assistedby a technical secretariat managed by the DGDRof MAEP. 5. Regional Monitoring Committeeswill be established in each o f the four project areas. It will be chaired by the Head of the Region and made up o f members of GTDR. The Regional Monitoring 61 Committee will be supported by the GTDR's Technical Secretariat, and will be responsible for (i) ensuring consistency o f project actions with project objectives and work plan, national strategy and policy, and regional development priorities and programs; (ii)preparing and validating detailed regional annual programs, work plans and budgets, at the regional level; (iii)reviewing project progress and performance, and the implementation o f corrective measures if necessary. The Regional Monitoring Committee will meet twice a year. Implementation of ProjectActivities 6. The overall coordination o f the project will be ensured by the General Department for Rural Development (DGDR) at MAEP. The Director o f DGDR will report to the NSC, responsible for oversight and approval o f annual reports and work plans. The overall coordination involves: 0 DGDRwill ensure project ownership at national level; 0 Regional Director for Rural Development (DRDR) will be responsible for project ownership o f project investments intheir respective areas; 0 To help them in these tasks, the project will finance recruitment (i) national level, o f an at international technical assistant (operations), advisor to DGDR, and (ii) regional level o f at four national technical assistants (operations), advisors to DRDR for implementing project investments; 0 Finally, DGDR and DRDR will select in their respective units one staff member who will provide support for coordination and project monitoring; 0 The DGDR will also be responsible for the implementation o f project activities at the national level, including capacity building at Ministrylevel, support to national policies and strategies, etc. The DGDR will do so by outsourcing to private sector service providers. 7. The project financial management will be ensured at national level by the Department for Administration and Finance at MAEP and, at regional level, by the DRDR finance director. The project will recruit a national financial management and procurement agency that will provide technical financial management assistance to MAFiP's Finance Director. The project will also recruit, at each DRDR, a national financial manager, who will be under contract with the DRDR and who will be in full time charge o f financial management o f the project. This person will work closely with MAEP DAFand will benefit from support fromproject financial TA at the national level. 8. Procurement will be ensured, at central level, by PRMP and, at regional level, by relevant units of the DRDR. The project will recruit (i) national financial management and procurement a agency (same as the above mentioned financial TA) that will provide technical assistance to the PRPM, and (ii) at the level o f each region, an additional staff, under contract, who will be full time in charge o f project procurement. This staff will work closely with PRMP and will benefit from the project support inprocurement TA. 9. Technical assistance. Recruitment o f TA - international (1) and national (7) - will be done under two separate contracts (one for financial and procurement management, and one for operational assistance) with specialized firms. The International "Operations" TA will be in charge o f (i) advising DGDR and their assistant and DRDRsl their assistants regarding operational strategy, project implementation and monitoring o f the project; (ii)training and providing operational support to MAER staff involved inproject implementation. The National "Operations" TAs who are recruited at the level o f DRDRs will be in charge o f advising and supporting DRDRs inproject implementation in their respective areas and o f ensuring coordination o f all project components at regional level. National TA in financial management and in procurement will be responsible for financial management and procurement and for providing technical support to DRDR staff. The four financial and four 62 procurement consultants at region level will be responsible for financial management andprocurement at the regional level. They will be recruited under one contract with the national level financial management andprocurement specialist, and will report to the national specialist. Implementation of ProjectComponents Component 1: Developmentof CommercialAgriculture Sub-component 1.1:Support to agricultural services 10. The DRDR will be responsible for the implementation o f this component. The project activities will be implementedas follows: Support to the development of commercial agricultural supply chains. This support includes identification and mobilization o f operators, strategic review o f market and value chains opportunities and constraints, identification and analysis o f productive sub-projects and will be provided by regionalpartners recruited ineach zone by the DRDR.The priorities and work planof these partners will be defined in consultation with the ASC and local platforms, and approved by the GTDR. They will be responsible for (i)administration o f the matching grants; (ii)support to eligible operators in the preparation o f sub-project proposals; and (iii) strengthen capacities and provide technical assistance to the ASC. Remuneration o f the partners will be partlybased on performance. 0 Building the capacities of farmers and strengthening of professional organizations, as well as the establishment o f agricultural service centers (ASC). These activities will be implemented under the responsibility o f the ASC that will be established at the beginning o f the project in each o f the four sites. The contractual staff o f the ASC will be recruited by the DRDR. Strengthening the supply of technology for production and valorization of agricultural products will be defined by the afore-mentioned platforms, with assistance from the ASC. They will be approved by the GTDR and implemented by one or more service providers (private sector, FOFIFA, ONG, others) that will be recruited competitively on the basis o f a multi-year contract with the DRDR. Sub-component1.2:Support to Private Investment 11. Support to private investment will be done through matching grants that will be provided on a demand-driven basis to individuals or groups. Ineach zone, matching grants will operate as follows: 0 A list with eligible (positive and/or negative) activities will be prepared, based on the contribution that these activities will make towards achieving the project's objectives. The GTDR will appoint a Selection Committee at the regional level. The GTDR will approve the request for matching grants after analysis and following a recommendation from the Selection Committee. An external review will be conducted twice a year. A regional partner will be recruited by the DRDR and will have the following responsibilities (i) identify and analyze market andvalue chains opportunities; (ii) awareness raising and mobilization o f private operators and potential investors; (iii)facilitate the preparation o f sub-project proposals by individuals or groups; (iv) facilitate their access to a financier; and (v) conduct a technical and financial analysis o f the sub-projects that request a matching grant. 63 Specialized service providers will be recruited by the DRDR on an as-needed basis to conduct strategic market and value chain studies. These studies can be conducted either by the demander or by a service provider following competitive bidding. A network of regional partners at the regional level will be compiled by the ASC. The network, with the ASC, will sign multi-year contracts that specify the modalities and the expectedresults. 12. Matching grants will be provided to activities that have been identified as priority by the Government: investments, technologies and advice. Inputs and technologies will only, andtemporarily (one or two year for the same beneficiary), be supported if they are necessary for the dissemination o f innovative technologies (e.g. conservation, ago-ecological technologies). The project will under no circumstances finance inputs that are already widely available and used by the producers and financed bymicro finance institutions. 13. Financial public support can be justified by the proportion o f "public good" o f the investment (roads, information, etc) and therefore by the assumption that leaving these investments to the private sector would lead to under-investment from a public resource allocation point o f view. 14. Financial public support can also bejustified for those beneficiaries that don't have the means to invest themselves -- vulnerable groups, etc. -- but where public support can help to lift these groups out o f poverty and to take care o f themselves: small productive investments (e.g., rice mills, oil press, etc) for women's groups. This is what is understood by the "merit good" o f the intervention, which i s related more to the beneficiary thanto the type o f investment. 15. In order to reduce the number o f subsidy levels for activities supported by the project, the following table is proposed: %public good 16. The beneficiary contribution will be paid fully in cash, either from own means or through credit, except for environmental protection activities (forestation, revegetating farmers' fields or reclamation o f degraded soils) by clearly defined beneficiaries (see component 3), where the contribution can be inkind. 17. Eligible operators and activities. The project will partially subsidize the following private operatorsg : Professional agricultural and ago-industrial organizations; Producers' organizations (crop, livestock, forestry); Rural communities; Commercial agricultural operators and ago-processors; agro-industrial companies; These operators need to prove their existence during at least two years before being eligible. 64 0 seed producers (associations andindividuals); 0 Distributors o f inputs and agricultural equipment; 0 Micro finance networks. 18. Eligible activities are clearly associated with agricultural production and with management o f natural resources (a specific positive andnegative list) as presented intable 1. Table 1: Eligible activities latching grant Support to marketing Market studies, value chain studies, development o f quality management and certification, testing o f samples. Support to innovation 0 Adaptative agricultural and ago-industrial research (varieties, technologies andproduction andprocessing equipment); 0 Introductionhest o f new agricultural production techniques (ex. agro-ecological); 0 Awareness raising and demonstration (inputs ,equipments); 0 Development o f new micro-finance products (e.g., weather insurance). Support to management 0 Technical and management advice (e.g., strengthening o f technical and substantive capacities of seed farmers); 0 Technical advice and extension of micro-finance networks. Support to investments 0 Rehabilitationof seed production; 0 Storage and harvest infrastructure; 0 Establishment/extension o f input and equipment distribution networks; 0 Integrated projects for the implementation of contract farming between private investors andsmallholder producers; 0 Forestation reclaiming o f degraded soils. 19. Funds under the MatchingGrant will be disbursed as indicated infigure 1. 65 Figure 1: MatchingGrant disbursements 1 3 Information Submission of Reviewof dissemination activity and proposals and andrequest for I> selectionby proposals proposals Technical Secretariat I I 13 Completionand Disbursement of deliverybythe I-> remainder of the grant A 11 5 Analysis of Implementationof External detailedproposals sub-projetand periodic byTechnical monitoringby Secretariat Technical Secretariat 6 10 Transmissionto Disbursement Selection of first part of Committee for the grant decision 9 Openinget replenishment Signingof Approvalby of account by Selection beneficiary Committee DRDR Component 2: IrrigationDevelopment 20. This component will be implemented under the responsibility of the DRDR. There are two sub-components. Sub-Component 2.1: Management of Irrigation Schemes. 21. Activities in this sub-component include (i)awareness raising and mobilization o f irrigation farmers and their associations; (ii)participatory diagnostic of options for management and 66 rehabilitation of the irrigation scheme (Scheme Development Plan or SDP); (iii) selection o f the preferred option for the mobilization and utilization o f water resources; and (iv) preparation of a Performance Contract between water users, Region, communities and MAEP. The DRDR will recruit an international consultant who, with support from a national consultant, will implement the above activities inthe four project zones. 22. The rehabilitated irrigation schemes will be managed in accordance with the relevant institutional framework (see table 2): (i) will be responsible for the operation and maintenance DRDR o f non-transferable irrigation infrastructure and for the mobilization o f financial resources; (ii) (F)WUAs will be responsible for operation and maintenance of transferred irrigation infrastructure, and for the mobilization of adequate financial resources among the water users through O&M fees; (iii) Communesaretheownersoftransferredirrigationinfrastructure,andwillbeco-responsible, the with the WUA, for maintenance. They will need to provide adequate assistanceto the (F)WUAs. They will also be responsible for the maintenance o f roads within the schemes. However, the three stakeholders - region, communes, WUA - will only be able to collect adequate finds progressively. This will require: (i)increasing agricultural production and productivity, which will improve the capacity to pay and (ii) implementation o f effective mechanisms for the mobilization o f financial resources (O&M charge, land tax, FERHA). Project resources will temporarily provide financial incentives on a cost sharingbasis. The Performance Contract will clearly define the obligations of all stakeholders. Sub-component 2.2: Irrigation Investments 23. The DRDRwill be responsible for the implementationo f the irrigation rehabilitation works. In each region, specific activities will be outsourced to (i) national consultant for the technical studies a and design of the works, including supervision of the works, and (ii) contractor for the construction a works. A single contract per region will be signed with a consultant for the duration o f the project. 24. (F)WUA will sign all contracts directly related to irrigation activities, and will be co- responsible for the selection and evaluation o f consultants and contractors. They will need to sign off on the completion of the works andpayments to contractors. Component3: Watershed Development 25. The component includes, in each o f the four project sites, (i) activities that aim to combat erosion and to conserve natural resources; and (ii)activities that aim to promote marketing and sustainable intensification o f agriculture in watersheds (outside irrigation schemes) through the promotion of production systems and appropriate production technologies. Activities related to agricultural intensification and marketing will be implemented under component 1 "Agricultural Development'' as described above. The sections below only relate to appropriate management and conservation o fnaturalresources. Subcomponent 3.1: Support to Watershed Management. 26. The DGDR and the DRDR will be responsible for the implementation o f activities under this sub-component: The DGDR will recruit international technical assistance for the preparation o f the Watershed Development Plan (as part o f the Watershed Master Plan) ineach o f the four project sites (one Plan per site); Each DRDR will recruit a regional partner responsible for (i) mobilization and capacity the strengtheningof the local and regional consultation platforms; and (ii) participatory planning and implementation o f the sustainable development and management of the various catchments. 67 0 A Memorandum of Understanding, one per Project Area, will be entered into by MAEP through N I W M P and h4INENVEF through the ThirdEnvironment Support Program to ensure adequate integration, in all Project Areas, o f the N I W M P and Third Environment Program Support Project financed under Credit from the Association No. 4362; Land registration ofices will be established by the respective communes. The DRDR and the communes will receive technical assistance from the National LandTenure Program (NLTP). The communes will be responsible for the activities and the proper functioning o f their Land Tenure Offices, and inparticular o f the recruitment o f adequate staff and financing. 27. The first activities that the project will launch will be an intensive awareness raising and communication campaign to inform the populations o f the watersheds, including irrigators, o f the project objectives andto mobilize them with respect to its implementation. Subcomponent3.2: Investment in watersheds. 28. The Watershed Development Plans include a number o f investments that will be implemented as follows: (1) Strategic anti-erosion works that have been identified as priority in the Watershed Development Plans. They will be 100 percent financed by the project and implemented by private contractors contracted by the DRDR. In as far as possible, works will be implemented through local labor to promote the appropriation by the local population. The selection o f contractors and payments made under the contracts will be certified by the involved communities; (ii) Establishment ofzonesundercollectivelandmanagement(GELOSE).Theserviceprovider under contract with the DRDR will be responsible for the facilitation o f these activities. The DRDRwill be responsible for satisfying the administrative requirements andthe registration at the Land Tenure Offices. Necessary investments, as well as the running costs o f the Land Tenure Offices, will be financed through Component 1; (iii) Disseminationofagro-ecologicaltechnologiesthatrequiredistributionofspecialinputsand access to extension will be implementedthrough the regional partners that will be recruitedby DRDR. Alternatively, in the case o f adaptive research, activities will be implemented by service providers that are contracted by the DRDR under component 1; (W Appropriate productive investments (forestation, revegetation o f land) that will be implemented by beneficiaries themselves and partially financed, on demand, through component 1. 29. Specific conditions regarding the participation o f beneficiaries and the support that they will receive through the matching grant will be determined on the basis o f an analysis that will be conducted during the preparation o f the Watershed Development Plans (nature o f the interventions, capacity to pay), that will also take into account similar programs under implementation in each o f the four sites. As a general principle, beneficiaries will contribute a minimum o f 20 percent to the investment costs (in kind or cash), with the exception o f the strategic anti-erosion works (see (i) above) that will be fully paid for by the project. 68 Component 4: Program Management Sub-Component 4.1: Project Management 30. Responsibility for the implementation and management of the project will be assured by the DGDR at the national level and the DRDR at the level of each of the four project sites. The DGDR and the DRDR will in particular be responsible for (i) preparation annual work plans and detailed the budgets (at regional level, and consolidated at the national level); (ii) monitoring o f implementation progress in accordance with the operations manual o f the project; (iii) preparation o f annual progress reviews that will be presented to the National Steering Committee and to the Regional Monitoring Committees; and (iv) conducting annual financial and technical audits. Specifically, the DGDRwill be responsible for the organization o f abi-annual external technical audit o fproject operations. 31. The DGDR will sign a M O U for the provision o f strategic and technical support to the land tenure operations o fthe project. 32. Monitoring and Evaluation. Monitoring and evaluation will be conducted under the responsibility of the Director o f Information Systems (DISE) of MAEP, who will be assisted by the international technical assistancelocated within the DGDR. Inorder to better integrate monitoringand physical investments, the project will adopt the Integrated Management System (SIG) developed by the PSDR. Independent technical audits will be conducted by service providers that will be qualified annually, beginning in the second year o f the project. Two external impact evaluations will also be conducted: (i) mid-term; and (ii) the end o f the project. The analyses and recommendations o f at at these evaluations serve to extend the activities at the national level. 33. Monitoring and evaluation will consist of three separate but closely related systems: (0 a system of internal monitoring conducted by MAEP under the responsibility o f DISE in collaboration with DGDR staff at central andregional level so as to ensure harmonization and coherence in the monitoring o f the various programs implemented by MAEP. However, this function can be delegated or outsourced to other entities either for an entire component (e.g., EP3 for the Watershed component) or for all activities at the regional level (e.g., GTDR for each site); (ii) asystemofparticipatoryevaluationateachofthefoursites(whichwouldallowforabetter appropriation and internalizationby beneficiaries) by directlyinvolving the main beneficiaries (PO, (F)WUA, etc.) in the definition, collection and analysis o f progress and impact indicators, and the identification o f corrective measures inthe event project objectives are not being achieved; (iii) a system of collaborative monitoring that invitesother stakeholdersto participateinthe collection, interpretation and analysis o f progress and impact indicators defined by the project (e.g. the GTDR disposes, ineach of the four sites, o f a regionalrural development plan, and of a data base with indicators, and that has a mandate in regional monitoring and evaluation. These GTDR could be directly involved in the monitoring and evaluation systems in each o f the sites)". 34. Monitoring indicators. Overall project monitoring is based on indicators that will be defined during preparation and that will form part o f the Project Brief Document (see Annex 3), and on the implementation plan that was agreed during project negotiations. Specific achievements under each of the components will be measured more in detail with the aid of a series o f more specific indicators. These indicators are grouped in two categories: (i) performance indicators that measure the resources lo Similarly, the PE3 through anMOUcouldbe maderesponsiblefor monitoringandevaluation of the Watershed component. 69 [input indicators] that the project has allocated and the activities it has implemented [output indicators]; and (ii)impact indicators that measure the results that the project has achieved [outcome indicators] as well as its impacts. 35. Integrated Management System (SIG). The monitoring system will be integrated into an Integrated Management System (SIG) that not only allows for a close interconnection between the implementation o f activities from identification to final delivery, but also and in particular for establishing a connection between technical and physical achievements and disbursements. The SIG also includes a procurement module that integrates the project procurement planand the status o f each o f the procurement activities o f the project. Sub-Component 4.2: Policy Support 36. The DGDR o f MAEP will be responsible for the implementation o f activities that aim to define national policies relevant for the agriculturalha1 sector. That will inparticular be the case for the definition o f the operational modalities o f management and replenishment o f FERHA. The G o M has committed to finalizing this before June 30, 2007. To achieve this, the DGDR will competitively recruit the technical assistance that it needs, and will organize necessary consultations with stakeholders at the national level (e.g. Consultative Platform for Rice, Fertilizer Producers' Association, Malagasy Association o f Seed Producers). 70 ..... . ... . . 0 . 0 . . e . . . 0 . . . . . . . . . 0 0 . 0 0 0 . I o 0 . 0 0 0 . e, r b) w c m p ... - c` ... m Annex 7: FinancialManagementandDisbursementArrangements Countryissues 1. The World Bank's CFAAKPAR, completed in 2003, and some diagnostic works carried out over the last three years by the Bank and other donors, identified a range o f weaknesses and issues hampering the performance o f Madagascar's budget and expenditure management system. T o address these issues, the government has developed in2004 and 2005, in conjunction with all key development partners, a priority action planfor public finance reform. 2. The main achievements encountered so far include: i)adoption o f a new organic public finance law; ii)introduction o f program budgets to improve the presentation o f the budget and its alignment with policy priorities o f the government as specified inthe PRSP; iii)reinforcement o f the Treasury internal control system by recruiting additional staff and improvingthe operational efficiency o f the "Brigade du TrCsor"; iv) simplification o f the expenditure management process by integrating the functions o f "sous- ordonnateurs" and credit managers; v) creation o f an internal control mechanism (IGF: General Finance Inspection) at the level o f the Ministryo f Finance; vi) review o f the legal framework for the control institutions (IGF, Brigade du TrCsor, Auditor general) and implementation o f capacity building measures to improve their efficiency; vii) development and implementation o f a computerized integrated financial management system in six main treasuries, capturing the different phases o f the expenditure process. The reinforcement o f the Treasury internal control and the implementationo f the integrated financial management system (IFMS) insix treasuries have improved timeliness and quality o f financial information. The Auditor General has also cleared the backlog o f accounts and completed the examination o f the draft budget execution laws for the years 1998 until 2003. The oversight function o f Parliament has been strengthened through capacity buildingandtraining o fthe Public Finance Committee. 3. While overall implementation progress o f the reform program is encouraging, significant efforts remain to be done, including the followings: i)improvement o f budget execution rate inpriority sectors; ii)reinforcement o f the capacity o f the line ministries in public financial management, especially in the implementation o f the new program budget structure; iii)strengthening o f cash management; iv) production o f the treasury accounts within the legal timeframe; vi) reinforcement o f control over State owned companies and national public institutions. The strategic coordination o f the multiple reforms and the monitoring & evaluation system need also to be improved. To mitigate risks in public expenditure management, the World Bank, through the Governance and Institutional Development Program (PGDI), and a number o f donors continue to support Government's public finance reforms reflected inits annual priority action planfor 2006. 4. Regarding the accounting profession, some positive developments have been noted over the last three years through assistance provided by the FIDEF (Fddiration Internationale des Experts Comptables Francophones) and INTEC (Institut National des Techniques Economiques et Comptables). However, a number o f local accounting firms continue to operate below the international standards due to the lack o f proper auditing standards, clearly defined guidelines and procedures for systematic peer reviews, quality control mechanisms to harmonize methodology. To improve the capacity and the competitiveness o f local auditing firms, the following measures have been taken while auditing BanWIDA financed projects: i)obligation for local auditors to enter into partnership with international accounting firms; ii)effective participation o f the international accounting firm in audit fieldworks and submission o f audit report jointly signed by the local and international audit firms. An accounting and auditing ROSC would be certainly helpful to identify clearly both issues and actions to be taken to strengthen the capacity o f the accounting profession inMadagascar. 74 FMRiskAssessment andMitigation 5. The following table identifies the key risks that the project management may face, and provides the measures to be taken to mitigate them: Risks Risk RiskMitigationMeasures Condition of rating Effectiveness (Yes/No) 1- Inherent Risk Country Level. Delays inthe productionof S These issues are beingaddressed NO public accounts, and audit may throughthe ongoing PFMreforms not be conductedin compliance supportedby IDA (through the with internationalauditing Governance and Institutional standards due to: weak capacity DevelopmentProject) and other ofthe accountingprofession in donors. Inthe meantime,the audit of Madagascar, and; ii)inadequate the WB-financedprojectswill be number of skilledand carriedout by internationalaccounting experiencedauditors at the firms or by internationalaccounting "Chambre des comptes" in firms associatedwith local auditing particular. firms, with effectiveparticipationofthe former (internationalaccountingfirm) inthe fieldwork. Entity Level Accounting, financial reporting Developmentpartnerswill continue to andauditingnot timely. S supportthe GoM priority actionplan NO for public finance reformsincluding trainingand capacity buildingof key line ministries(includingMAEP) inthe areaof public financial management. This trainingwill includeplanning, budgeting, accounting, reportingas well as monitoringand evaluation. Recruitmentof an international auditingfirm acceptable to IDA to carry out the annualaudit ofthe project financial statements duringthe first Effectiveness phase ofthe Project. condition Project Level Recruitmentof a Financial The MAEP Financial Management Agency (FMA) in Directorate (DFB) andthe three conformitywith Bank proceduresto: i) regionalspendingunits(RDFB) S providethe DFB and RDFBwith Effectiveness have no experiencewith qualifiedand experiencedaccountants condition implementingan IDA- financed and; ii)assist the accountingstaff of project. these DirectoratesinperformingFM tasks. 2- Control Risk Budget Budgetpreparationis clearly M Beforecredit effectiveness,the MAEP NO 75 definedbut ther is no accountingstaffwill attendspecific systematicexplanation of traininginFMfor World Bank- significantvariances. financed projects, includingplanning, budgeting, accounting, internal controls, reporting, disbursements, and auditing. Accounting Projecttransactionsmaynot be Implementationof an accounting Effectiveness properlyaccounteddue to systemacceptableto IDA, including: condition . absenceof a Chart of accounts. 1 Preparationof a Chart of accounts. Policiesand controlprocedures Elaborationof an accounting maynot be applied efficiently manual ofproceduresacceptable to due to absence of appropriate IDAto provideclear guidanceto documentationcovering staff. accounting, managementof 1 Users trainingto ensure constant finances andfinancial reporting. applicationofpolicies/procedures. InternalControls Absence of written control S [See above mitigating measuresrelated Effectiveness proceduresfor the project to "Accounting" section] condition Internalauditorsmay not be M Beforecredit effectiveness, internal NO familiar with the World Bank auditors (MAEP internalaudit FMprocedures. Department and IGF/IGE)will attend specific training infinancial managementfor World Bank- financed projects. Funds flow Absence of guidelines for M Theproceduresto be followed for NO accountingofbeneficiary accountingof beneficiarycontributions contributions,and and disbursementswill be describedin DFB/RDFBaccountingstaff the accountingmanualofprocedures. maynot be familiar with the WB disbursementsprocedures, Organization of users trainingto ensure proper applicationoftheseprocedures. Financial reporting Risk of delays in the production S Recruitment of a consultant to design Effectiveness and submission of financial and implement a computerized system condition reports requiredfor managing to allow for timely production of andmonitoringproject reliablefinancial statements and FMRs. activities. Organization of a trainingsession to ensureefficient use ofthe new computerizedsystembythe project accountingstaff. Auditing The quality of the audit maynot S Recruitment of an international Effectiveness be acceptableand the reportnot auditingfirm acceptable to IDA to condition delivered in time due to weak carry out the annualaudit ofthe project capacity of the accounting financial statements duringthe first profession. phaseof the Project. S: Substantial M:Moderate 76 Strengths, Weaknesses and Action Plan 6. The project financial management i s strengthenedby the following salient features: 0 The organizational structure within the MAEP defines the lines o f responsibilities and authority that exist and seems appropriate for planning, directing andcontrolling operations. Existence o f an Internal Audit Department within the MAEiP which can collaborate with the IGFAGE (Inspection GBnBrale des FinancesAnspection GCnCrale de 1'Etat) while carrying out internal audit. 7. The main deficiencies noted in the MAEP financial management system are summarized in the following table which also providesrelevant measuresto address them: Significantweaknesses Actions - Date Responsible due bv Lack of skilled andexperienced Recruitment of a FinancialManagement 12131/06 MAEPDFB, accountingstaff acquainted Agency (FMA) inconformitywith Bank IDA with Bankprocedures proceduresto: i)providethe DFB and RDFBwith qualifiedand experienced accountants and; ii)assist the accounting staff of these Directoratesin performing FMtasks. Before credit effectiveness, a training 1O/ 17/06 IDA session will be organized to familiarize DFB and RDFB accounting staffwith the Bank proceduresin financialmanagement and disbursements. Absence of a Chart of accounts Elaboration and implementation of an 11/30/06 Consultant reflecting project components accounting and financial manual of and activities outlinedinthe procedures by a consultant to facilitate PCNIPAD. adequate record keeping, ensure maintenanceof proper control over assets Absence of an accounting and satisfy MAEPIproject reporting manual ofprocedures. requirements. Users training providedby the consultant 01/15/07 Consultant to ensure effective application of control procedures; Absence of an integrated Design and installation of a computerized 12115/06 Consultant computerizedsystem capable financial management system to satisfy of: i)producingtimely MAEPProject requirements and ensure financial informationand other timely production of financial statements reportsrequiredfor managing and FMRs. andmonitoringProject activities; ii)ensuringtimely Organization of users training by the 01/15/07 Consultant communicationof information consultant to ensure efficient use of all from central levelto regional modules offeredby the software. levels and vice versa: 77 Actions Date Responsible due bv Absence of acceptable Recruitment of an auditing firm 12/31/06 MAEP, IDA arrangement in acceptable to IDA to carry out the PHRD, auditing. GEF/preparation, PPF, annual audit of uroiectaccountsduringthe first uhase Budgeting 8. Each DirectorateDepartmentIService within MAEP prepares its own budget and submits it to DFBfor consolidation. The MAEPbudget request is therefore presentedto the Ministryo f Finance for discussion and decision-making in conformity with the defined calendar. Since FY 2005 the Government has set up a task force to assist key sector ministries (including MAEP) inthe preparation o f their program budget inorder to improve the quality o f their submissions. The accounting software to be implemented will facilitate significantly budgetary management. Accounting 9. The DFB and RDFB (DRDR- Department o f Financial & Budget) will use an accounting system incompliance with generally accounting standards/PCOP (Plan Comptable des OpCrations Publiques) and IDA requirements. This system will operate on a decentralized basis with the four regions concerned and will use standard book accounts (journals, ledgers and trial balances) to enter and summarize transactions. Revenue will be recorded when cash is received, while expenses and related liabilities will be recorded when incurred, especially upon receipt o f goods, works and services. Each RDFB will maintain separate financial records for all transactions under its responsibility and sends, on a monthly basis, the balance sheet to the DFB for consolidation. The DFB, at the central level, is in charge o f timely production of: i)periodic financial reports with supporting documents requiredby the Ministry of Finance to ensure proper recording o f project expenditures in the government budgeting andaccounting systems; ii)quarterly FMRsand annual financial statements. 10. To strengthen the MAEP financial management system it was agreed that a consultant will be recruitedto elaborate an accounting manual o fprocedures inorder to complete the administrative manual inplace and to describe the outline o f the project accounting system, the accounting policies to be followed, the formats o fbooks andrecords, the Chart o f accounts, the financial reporting, and relevant information to facilitate record keeping andmaintenance o fproper control over assets. H e will also provide adequate training to staff to ensure better understanding andproper applicationby the staff o f all procedures described inthis manual. 11. The Watershed Management Project will use an integrated financial management system capable o f recording andproducing ina timely manner all financial reports required for managing and monitoring project activities. This computerized system would in particular facilitate: annual programming o f activities and project resources, record-keeping (general accounting and cost accounting), financial and budgetary management, fixed assets management, procurement management, follow-up on project implementation progress, preparation o f project financial statements and quarterly Financial Monitoring Reports as requiredby the Bank/IDA.The consultant in charge o f the design and implementation o f this integrated FM system will provide also users training to ensure efficient use of all modules offered by the software. The TORSfor this consultant will be reviewed by the Bank Financial Management Specialist. The new computerized system will be fully functional before project implementation begins. To avoid double data capture, this system should allow for extracting efficiently all required information from the Data Base ORACLE presently in place and used by the M A E P for recording commitments, "liquidations" and settlement orders. 78 Internal Control& InternalAuditing 12. The accounting staff in charge o f the project financial management i s not appointed yet. The Financial Management Agency (FMA) in charge o f the financial management o f this project will provide DFB and RDFB with adequate number o f qualified accountants composed of: i)a FM specialist responsible for overall supervision o f finance, accounting, budgeting and reporting; ii)five (5) qualified and trained accountants: one for DFB and the remaining for the four selected regions (RDFB). Whereas the accounting assistants will be selected among the DFB/RDFB staff based on their qualification and performance. The project accounting staff should be in place prior to credit effectiveness. 13. The internal control system actually in place needs to be strengthened to ensure timely preparation o f reliable information. Specific attention should be given to the following control procedures: use o f adequate documents and records, and physical control over assets and records. To encourage consistent application, control procedures should be spelled out in the project accounting manual that should be available prior to credit effectiveness. To ensure efficient use o f credit funds for the purposesintendedandconsistent applicationofproceduresonprocurement, financial management, disbursement, the IGF/IGE inclose collaborationwith the MAEP Internal Audit Department will play the role of internal auditors. They will report directly to the Minister of Finance and the Steering Committee. All issues identified during internal audit should be addressed quickly to improve the project performance. Funds Flow and Disbursement arrangements 14. The flow o f funds from IDA, GEF and Government i s presented as follows: - 1 MAEP Directorateof Financeand Budget (DFB) -- SpecialaccountIDA (S/A: A) SpecialaccountGEF (S/A: B) 1 I DRDRs Departmentof Financeand Budget - (RDFB) -- Sub-DesignatedAccount IDA (SS/A: A) Sub-DesignatedAccount GEF (SS/A: B) Contractors, suppliers of goods and services Disbursementfrom IDA credit, and GEF 15. For the implementation o f Watershed Management Project the following bank accounts will be opened inlocal commercial banksunder conditions satisfactory to IDA: Special Account A to be managed by DFB: Denominated in US$, disbursements from the IDA credit will be deposited on this account to: i)finance project components/activities 79 agreed with IDA in accordance with the disbursement percentage(s) indicated in the Financing Agreement; ii)replenish sub special accounts A managed by RDFB. 0 Special Account B: Denominated in US$, disbursements from the GEF grant will be deposited on this account to i)finance project components/activities agreed with IDA in accordance with the disbursement percentages indicated in the Financing Agreement; ii) replenish sub special accounts B managed by RDFB; 0 Special Accounts 30-Day Advance Procedure A and B to be managed by each RDFB: Denominated in local currency, disbursements from IDA and GEF special accounts will be deposited respectively on these accounts opened in a local commercial bank to ensure prompt payment o f contractors/suppliers operating in the regions, in conformity with the disbursement percentages indicated inthe FinancingAgreement. 16. Funds deposited in these accounts will be used to ensure timely payments of contractors, suppliers o f goods and services. The project implementation and accounting manuals will describe in details all procedural aspects regarding financial management and disbursements from the special account(s), andproject account (payments, replenishment, accounting, reporting and internal controls). Method of Disbursement: 17. During the first year o f project implementation, the DFB would follow the transaction- based disbursements procedures (traditional mode) outlined in the Bank's Disbursement Handbook. The use o f report-based disbursements could be possible if requested by the borrower and if the following criteria are met: i)the FMrating has been maintained at satisfactory level; and ii)the submission o f at least three quarterly satisfactory FMRs that could be relied upon for purposes o f disbursement. Detaileddisbursement procedures will be described inthe project accounting manual o fprocedures. Minimum of Application Size: 18. The minimum application size for direct payments, to be withdrawn directly from the Credit Account, and special commitments is 20 percent o f the amount advanced to the related special account. Use of Statements of Expenses (SOEs) 19. Disbursements would be made against Statement o f Expenses (SOEs) for contracts and goods not requiringthe Bank's prior review. Therefore disbursements for all contracts for: 0 Contracts for works o f less thanUS$500,000; Contracts for equipment and goods inan amount inferior to US$250,000; 0 Contracts for consulting services, training by firms o f less thanUS$lOO,OOO; 0 Contracts for consulting services, training by individual o f less than US$50,000; 0 Training not subject to contract and all incremental operating expenses. would be made on the basis o f SOEs and certified by the DFB. SOE statements would be audited semi annually by independent auditors acceptable to the Bank. All SOEs supporting documentation would be kept therefore by the project and made available for review by Bank supervision missions and external auditors. Designated Accounts 20. Payments from the IDA Credit and GEF would be administered by the DFB from two separate Designated Accounts which would be opened in local commercial banks o n terms and conditions 80 acceptable to IDA. The authorized allocation for the special account covering IDA's contribution would be 2.5 million covering IDA's share o f four (4) months o f estimated expenditures. The initial deposit will be limited to US$1.25 million and subsequent advances may be made as the need arises. The ceiling for the designated account under the GEF grant shall be US$500,000. The DFB would be responsible for preparing disbursementrequests. The Designated Accounts would finance all project eligible expenditures inferior to 20 percent o f the authorized allocation, andreplenishmentapplications would be submitted at least on a monthly basis. Further deposits by IDA into the Special Accounts would be made againstwithdrawal applications supported by appropriate documents. 21. The Designated Account would be replenished on the basis o f documentary evidence provided to IDAby DFB,justifying the payments o f expenditures that are eligible for financing under the credit. All supporting documentswill beretainedby the project (DFB, RDFB) andmade available for review byperiodic Banksupervision missions, internal andexternal auditors. SpecialAccount 30-Day advanceprocedure 22. The initial advance paid to each regional account would represent funds sums covering no more than 30 days estimated expenditures basedupon submission of satisfactory budgetedwork plans. Subsequent payments will be based on SOEs submitted by RDFB after appropriate authorization and approval by the DFB. The RDFB will submit at least monthly expenditure reports indicating sources and uses o f funds, physical achievements, a summary of expenditures by category in SOE format, and accompanied by reconciled bank statements. No further advances will be made from DFB to delinquent RDFB until proper justification has been received from them. All SOE supporting documents will be retained by each RDFB and made available for review by DFB, IDA, internal and external auditors. TableA. Allocation of Loan/Credit Proceeds Expenditure Category Amount inUS$ million Financing Percentage'' Credit (1) Goods, works consultants' services, training and operatingcosts 100% (a) for Parts B.2 (b) and (c) of the Project; 10.56 (b) for other Parts of the project 13.86 (2) Sub-project Matching Grant 2.34 100% (3) Refund of Project Preparation 1.61 Advance (4) Unallocated 1.63 Total 30.00 "TheFinancingPercentageisbasedontheFinancialManagementParametersfor Madagascarwhichallowsthe financing oftaxes. 81 FY 7 8 9 10 11 Annual 1.45 4.34 6.95 10.56 6.70 Cumulative 1.45 5.79 12.74 23.30 30.00 23. To monitor project implementation, the DFBwill produce the following reports that should be preparedincompliance with international accounting standards: 0 Annual financial statements comprising: i)Summary o f Sources and Uses o f Funds (by components/project activitiedcredit category and showing all sources o f funds); ii)Project Balance Sheet; iii)the Accounting Policies Adopted and Explanatory Notes; iv) a Management Assertion. 0 Quarterly FMRs: The FMRs include financial reports, physical progress reports and procurement reports to facilitate project monitoring. The FMRs should be submitted to IDA within 45 days of the end of the reporting period (quarter). 24. The form and content o f quarterly FMRs and annual financial statementshas been determined during project appraisal and was agreed at negotiations. Models o f these reports will be presented in the project accountingmanual ofprocedures. Auditing 25. The project financial statements will be audited annually by an international private accounting firmacceptable to IDA, inaccordancewith InternationalStandards o f Auditing andthe FM Practices in World Bank-financedInvestment Operations issuedby the FMSector Board inNovember 3, 2005. The auditors will be required to: i)express an opinion on the project financial statements; ii) cany out a comprehensive review o f the internal control procedures and provide a management report outlining any recommendations for their improvement. The audit report will be submittedto IDA not later than six months after the end o f each fiscal year. The auditors should be recruitedprior to credit effectiveness. Since the project is expected to be effective in January 2007, separate audit i s not required for the PPF: amounts disbursed for this purpose will be accounted for in the first reporting period of the new project. The terms o f reference o f the audit will be reviewed by the financial management specialist of the Bank/IDA to ensure the adequacy o f the audit scope, drawing special attention to particular risk areas identified duringproject preparation. 1- Project financial statementsincluding PPF Audit Report DueDate Within six months after the end o f each financial year. 26. A supervision mission will be conducted twice a year to ensure that strong financial management systems are maintained for the project throughout its life. Our input to FMrating will be indicated in the Implementation Status and Results Report (ISR). Periodic review will be also carried out when neededto ensure that expenditures incurredbythe project remaineligible for IDA funding. 82 Annex 8: Procurement Arrangements A. General 1. Procurement for the proposed project would be carried out in accordance with the World Bank's "Guidelines: Procurement under IBRD Loans and IDA Credits" dated May 2004; and "Guidelines: Selection and Employment of Consultants by World Bank Borrowers" dated May 2004, and the provisions stipulated in the Legal Agreement. The various items under different expenditure categories are described in general below. For each contract to be financedby the Credit, the different procurement methods or consultant selectionmethods, estimated costs, prior review requirements, and time frame are agreedbetweenthe Borrower and the Bank inthe Procurement Plan. The Procurement Planwill be updatedat least annually or as requiredto reflect the actual project implementationneeds andimprovementsininstitutional capacity. 2. Advertisement: A General Procurement Notice will be published in UN Development Business and Development Gateway Market (dgMarket) and will show all International Competitive Bidding (ICB) for goods and works and major consulting service requirements. Specific Procurement Notices will be issued in Development Business and dg Market and at least one newspaper with nationwidecirculation for ICB contracts andbeforepreparationof shortlists with respect to consulting contracts above US$200,000, inaccordance with the Guidelines 3. Procurement of Works: Works procuredunder this project would include: the rehabilitation of irrigation and appurtenant infrastructure, small infrastructure (storage, markets, etc) and erosion control works in watersheds. The procurement will be done using the Bank's Standard Bidding Documents (SBD) for all ICB and National SBD agreed with or satisfactory to the Bank and with any special requirements specific to the Project. To the extent practical, contracts shall be grouped into bid packages estimated to cost the equivalentof US$500,000 or more and would be procuredthroughICB procedures. For contract estimated to cost less than US$500,000 equivalent per contract, civil work procurement may be carried out through National Competitive Bidding and contractsfor small works, estimated to cost less than US$50,000 may be procured through quotations procedures. The bidding documents shall include a detailed description of the works, including basic specifications, the required completion date, basic forms of agreement acceptable to IDA and relevant drawings where applicable. Specific proceduresdetails canbe found inthe Project ImplementationManual. 4. Procurement of Goods: Goodsprocuredunder this project would include office furniture and equipment, vehicles, motorbikes, computer hardware and software, and office equipment, pumps and other water related equipment. The procurement will be done using the Bank's SBD for all ICB and National SBD agreedwith or satisfactoryto the Bank and with any special requirementsspecific to the Project. To the extent practicable, contracts shall be grouped into bid packages estimated to cost the equivalent of US$250,000 or more and would be procuredthrough International CompetitiveBidding (ICB) procedures. For contract estimated to cost less than US$250,000 equivalent per contract, procurement of goods may be carried out through National Competitive Bidding (NCB) procedures and purchase of small furniture estimated to cost less than US$30,000 will be conducted through prudent shopping procedures. 5. Direct Contractingfor works and goods may be used in exceptional cases, such as for the extension of an existing contract, standardization, proprietary items, spare parts for existing equipment, and urgent repairs and emergency situations, according to paragraphs 3.6 and 3.7 of the Guidelines. The items to be procured through Direct Contracting would be agreed on in the procurement plans. 6. Procurement of non-consulting services: Procurement from United Nations agencies for supplies carried out under their own procedures may include UNDP andor the International Agency Procurement Services Organization(IAPSO). The standard form of contract with UNagencies will be 83 used for such procurement. The items to be procured from UN agencies would be agreed on in the procurement planif andwhen to be used. 7. Selection of Consultants: The project will finance the contracting o f consultancy services for design o f most o f the civil works included in the project, construction supervision studies, technical assistance, financial and technical audits, specialized advisory services and capacity building. Firms will be recruited on the basis o f the Quality and Cost Based Selection (QCBS) method, using the Bank's Standard Request for Proposals. Selection based on consultants' qualifications (CQ) can be used for the recruitment o f training institutions and for assignments that meet criteria set out in para 3.7 o f the Consultant Guidelines. Single source selection can be used to contract firms for assignment that meet criteria set out inpara 3.9 to 3.13 o f the Consultant Guidelines. For the purpose o f very small assignments referred in para 3.10 o f the Consultant Guidelines, and for which the contract estimated costs do not exceed US$lOO,OOO. These items would include specialized advisory services to the Steering Committee andthe National Project Coordination for the program and project implementation monitoring and for the "Panel o f Experts" hired to monitor the social, environmental and cultural impact o f the project. Specialized advisory services would be procured through Individual Consultants Selection (ICs), based on the qualifications o f individual consultants for the assignment in accordance with the provisions o fparagraphs 5.1 through 5.3 o fthe Consultant Guidelines. 8. Community participationin procurement:Community participation in Procurement would be based on AFR Guidelines - Simplified Procurement and Disbursement Procedures for Community- Based Investment. This would comprise a broad spectrum o f activities related to watershed management and irrigation. Procurement will be described inthe Project Implementation Manual to be approved at Project effectiveness 9. Operating Costs financed through the Project would be procured using the implementing agency's administrative procedures, which were reviewed and found acceptable to the Bank. 10. The procurement procedures and SBDs to be used for each procurement method, as well as model contracts for works and goods procured, are presented inthe Project Implementation Manual 11, Review by the Bank of ProcurementDecisionsThe thresholds for prior review by Bank are specified in the procurement plans. Table 16 shows (i) the proposed thresholds for the different procurement methods, and (ii) the proposed initially-agreed thresholds for prior review by the Bank. The Bank will preview procurement arrangements proposed by the Borrower for the items specified in the procurement plans for their conformity with the Development Credit Agreement and the applicable Guidelines. Any procurement item not specified for prior review may be subjected to a post-review o f the procurement process. 84 ExpenditureCategory ContractValue Procurement Contracts Subject to Prior Threshold (US$) Method Review (US$) Works 500,000 or more ICB All (US$21.2 Million 50,000 or more and less NCB than 500,000 Less than 50,000 Quotation Goods 250,000 or more ICB All (US$1.OMillion) 50,000 or more and less NCB than 250,000 Less than 50,000 Shopping ConsultantServices - 100,000 or more QCBS All (US$2.6Million) Firms Less than 100,000 CQS Less than 100,000 sss All ConsultantServices - 50,000 or more ICs All (US$5.5Million) Individuals Less than 50,000 Less than 50.000 sss All B. Assessment of the agency's capacityto implement procurement 12. Procurement activities will be carried out by the Direction G6nkrale du D6veloppement Rural (DGDR) at national level and Directions R6gionales du Dkveloppement Rural (DRDR) at regional level. These units are MAEP departments and properly staffed; the procurement function under Unit6 de Gestion des marches Publics (UGMP) is staffed by Procurement Officer anddiverse civil servants. 13. An assessmento f the capacity o fthe Implementing Agency to implement procurement actions for the project has been carried out by Daniel Thirion, Consultant on August, 2005. The assessment reviewed the organizational structure for implementing the project and the interaction between the project's staff responsible for procurement Officer and the Management's relevant central unit for administration and finance. 14. The key issues and risks concerning procurement for implementation o f the project have been identified and include the phasing o f activities to be undertaken and possible emerging o f emergency cases. The corrective measures which have been agreed are the close follow-up o f the agreed procurement plan and activity scheduling. A procurement action plan will be fine tuned quarterly and the main procurement planwill be up-dated accordingly. 15. The overall project risk for procurement is Average Designation Concerns Risk mitigation Due date Planning and Lack of budget -Capacity building on budgeting Project effectiveness budgeting planning -Recruitment of ProcurementTA -- At At Project effectiveness Execution and Lack of internal Developmentof cost and - At Project effectiveness monitoring Audit contract management control Staffing Competentbut -Recruitment of ProcurementTA - At Project effectiveness insufficient Competition among Lack of Use of GPN at national level - At Project effectiveness private sector advertisement Use of Bank procedure for -. 85 advertising Project management Lack definition o f Development o f project - AtProject effectiveness responsibilities implementation manual C. Procurement Plan 16. The Borrower, at appraisal, developed a procurement plan for project implementation which providesthe basis for the procurementmethods. This planhas been agreedbetweenthe Borrower and the Project Team on June 30,2006 and is availableat the DGDRoffice. It will also be available inthe project's database and in the Bank's external website. The Procurement Plan will be updated in agreement with the Project Team annually or as required to reflect the actual project implementation needsand improvementsininstitutional capacity. D.Frequency of Procurement Supervision 17. In addition to the prior review supervisionto be carried out from Bank offices, the capacity assessment of the Implementing Agency has recommended annual supervision missions to visit the field to carry out post review ofprocurementactions. E.Detailsof the ProcurementArrangements InvolvingInternational Competition Goods, Works, and Non Consulting Services (a) List of contract packagesto beprocuredfollowingICB: Table3: List ofcontract packages 4 5 6 7 8 9 Estimated Procurement P- Domestic Review Expected Comments Method Q Preference by Bank Bid-Opening (yesho) (Prior / Date Post) 2.1-11 rehabilitation 985,700 March 2008 2.1-12 works Lac 3,112,500 March 2009 2.1-13 IAlaotra 1,087,600 ICB N o No Prior March 2010 2.1-21 Irehabilitation II 386,400 March 2008 2.1-22 works 1,2 19;8QO March 2009 2.1-23 Marovoay 426,200 ICB N o No Prior March 2010 2.1-32 rehabilitation 252,500 March 2008 2.1-33 works Andapa 797,400 ICB No No Prior March 2009 2.1-34 278,600 March 2010 2.1-43 rehabilitation 442,600 March 2008 2.1-44 Itasy 1,397,500 ICB N o N o Prior March 2009 2.1-45 Iworks 488,300 March 2010 Prior March 2007 IAPSO Agencies 1.1-C IIComputers UnitedNations No No motorbikes and I 212,000 ICB No No Prior April 2007 office equipment (b) ICB contracts estimated to cost above US$500,000 for works and US$250,000 for goods per contract will be subject to prior review by the Bank. Consulting Services (a) List of consultingassignments with short-listofinternationalfirms, 86 Table4: List of consultingassignments TAFA are Gov (b) Consultancy services estimated to cost above US$lOO,OOO per contract and single source selection o f consultants (firms) and o f individual consultants assignments estimated to cost above US$50,000 will be subject to prior reviewby the Bank. (c) Short lists composed entirely o f national consultants: Short lists o f consultants for services estimated to cost less than US$lOO,OOO equivalent per contract may be composed entirely of national consultants inaccordance with the provisions o fparagraph 2.7 o f the Consultant Guidelines. 87 Annex 9: Economicand FinancialAnalysis The Irrigation and Watershed ManagementProject 1. The Malagasy government is preparing a new irrigation program based on a watershed approach inorder to increase agricultural productivity and farmer income inselected rural areas. These two objectives should be reached through a mix o f software and hardware investment in agricultural service improvement, irrigation scheme development andupper watershed protection. 2. The World Bank will support this program by financing, together with the Global Environment Facility (GEF) and the private sector, direct investments in three areas: (i) commercial amicultural develoument to better link farmers to markets for inputs and credits and introduce agricultural technology inirrigated areas andtanety, for a total o f US$12.8 million (baseline costs plus taxes); (ii) irrigation development in the lower watershed to rehabilitate physical infrastructure (such as water intakes, dikes, canals, and drainage systems) and financial infrastructure'*, for a total o f US$17.47 million; and (iii)watershed management in the w u e r watershed to promote more sustainable land management for a total o f US$4.33 million. From the point o f view o f the economic analysis, these three components represent one integrated package and cannot be treated separately. A fourth component will support the project management for a total of US$4.31 million, including the Project Preparation Fund. 3. Overall, the project will cost WS$40.40 million (including physical and price contingencies), o f which 74 percent will come from IDA (US$30.0 million), 15 percent will come from GEF (US$6.0 million), and 11 percent will come from the private sector (US$4.4 million). GEF will support the watershed development (component 3), and the private sector will support the development o f commercial agriculture (component 1). 4. The project will target 30,500 hectares o f cultivated area in four watersheds with potential for agricultural development: Marovoay, in the Northwest, and Lac Alaotra (the Sahamalato scheme), in the Middle East, the two rice granaries in the country, and Itasy in the mid-West and Andupa in the North. The objective will be to sustainably and significantly increase agricultural production in these areas. 5. The total irrigated area o f these four watersheds represents 71,800 hectares, a third o f which will be concerned by the project (21,800 hectares), also constituting eight percent o f the irrigated schemes area (300,000 hectares), and 2.25 percent o f the country's total irrigated areal3 (1 million hectares). Some o f the irrigated perimeters, like Lac Alaotra and Marovay, are over 1,000 hectares; others, like Itasy, are between 100 and 1,000 hectares; the rest, like Andapa, are under 100 hectares. All 21,800 hectares are already equipped with concrete irrigation infrastructure, but only 60 percent are considered irrigated areas; the other 40 percent, mostly located downstream from the irrigated areas, are not irrigated year-round. Mainly the Fondsd'Entretienet de RehabilitationHydro-Agricole (FERHA) for the part of the irrigationinfrastructure that benefits the nationalpublic, andWater User Associations ( W A S )for the others. l3InMadagascar,indeed, morethan onemillionhectares, or 40 percentof all cultivatedlands, are irrigated,andproduceless that 3 million tons per year (300 grams a day per capita), nearly half of the daily energy intake. The bulk of these irrigated lands, 800,000 hectares or 70 percent of the total irrigated area, are very small in terms of average superficies (a few hectares), and are not equipped with irrigation infrastructures such as concrete dams, water intakes or line-canals. Consequently,water control is low and so i s the paddy yield. These 20,000 or so perimeters built by farmers inalmost every lowland area of the countryare called traditional irrigatedareas or micro/villageperimeters; they are not irrigatedperimeters in the sense given by this study. The remainingportion covers 300,000 hectares, or around30 percent of the irrigatedareas, and is equipped with infrastructure meant to improve water management and thus intensify production (i.e. improve productivity). The infrastructurewas built by the State during the colonial era at the very beginningof the twentiethcentury andwas maintainedby it untilthe liberalizationof the sector in 1984. 88 6. The project will finance the rehabilitation o f irrigation infrastructure in the 60 percent o f the schemes that are well-irrigated, and will finance the introduction o f agricultural technologies (such as improved varieties) in the other 40 percent, which should help raise productivity without improving water control over the year. On the irrigated area, the project will develop off-season cropping (mostly tomatoes and potatoes) on 4,700 o f the 21,800 hectareso f irrigated perimeters targeted by the project, mostly inItasy and Andapa. 7. The project will also develop rainfed agricultural production in the lower part o f the four watersheds (the first hillsides or tanety surrounding the irrigation systems) on around 7,700 hectares, less than 10 percent o f the rainfed cultivated area (45,000 hectares), through the introduction of agroecological techniques such as tillage and zero plowing. The objective o f developing agro-ecology i s to increase on-site productivity and decrease off-site siltation. Indeed, as the intervention will target the first hillsides around the systems, it is expectedthat erosion control will leadto decreasedsiltation. Table 1: Areas and Beneficiaries ofthe Project Partially- Off-season Watershed Irrigated irrigated Agro-ecology Beneficiaries Areas (ha) Areas (ha) IrrigatedTanety(ha) (ha) on (households) Marovay 3,670 2,400 200 2,000 18,120 Itasy 2,060 3,600 3,000 2,400 25,915 Andapa 2,150 1,500 1250 1,500 14,851 Lac Alaotra 6,000 400 250 1,800 30,676 Total 13,880 7,900 4,700 7,700 89,562 8. The project investment cost will be around US$1,330 per hectare for the four-year period o f project implementation, or about US$330 per hectare per year. This represents an increase of almost 4.5 times the investment o f the last 20 years, which was about US$1,500 for that time period, or about US$75 per hectare per year (see section 2 of this annex). 9. The project will take place inrural areas where the vast majority of households live under the threshold of absolute poverty (less than US$1 a day). The project will benefitabout 90,000 households or, with an average o f 5.5 people per household, around half a million people in a country that has a total population o f about 20 million people, 10. The improvement in agricultural productivity will likely contribute to poverty alleviation for these people by yielding some benefits to farmers, laborers and consumers, including the poorer marginalized people that suffer from regular famines during "soudure" periods. Indeed, in Madaga~car'~,increased irrigated paddy productivity is believed to be positively linked with an increase in real agricultural wages, a reduction in the number o f the food insecure and a reduction o f the paddy price to consumers. Public Spendingand Irrigation ProductivityinMadagascar 11. A World Bank ESW (P096045) entitled Madagascar: The impact of public spending on perimeters productivity, 1985-2004,has looked carefully at the impact o f public spending on irrigation system productivity during the last twenty years, a period that started with a major regulatory shift - the liberalization o f the rice sector - accompanied by an abrupt transfer o f the operation and maintenance (O&M) o f the medium-sized and large irrigation systems (irrigated areas equipped with l4A doubling of rice yields leads to a reduction inthe price of rice by 45 percent inthe harvest period and by 20 percent in the lean season (Minten and Barret, 2005). A one-percent increase in rice yields leads to a reduction of the number of the food insecure by 5 percent and reduces the length o f the leanperiod by 2 months (Minten and Barret, 2005). 89 infrastructure meant to improve water management and thus intensify production) from the State and State-ownedenterprises to Water User Associations (WUAs). 12. A sample of 108 irrigation schemes that were given help to organize O&M, representingan area of 123,500 hectares and400 WAS,receivedUS$190million, or halfofpublic spending, through 25 irrigation projects (including 2 extension projects) financed by 11 donors". These irrigation schemes represent around 60 percent ofthe irrigated areas of more than 50 hectares that were endowed with concrete infrastructure at the beginning of the period and represent around 11 percent of Madagascar's total irrigated lands (believed to cover around 1 million hectares). The beneficiaries were the about 100,000 water users, or around 600,000 people, and about 60 percent of the $190 million was investedinhardware for irrigation infrastructure, while the other 40 percent was invested in software, mostly for capacity building of WUAs and, less importantly, for promoting agricultural technologies, such as chemical fertilizers, improved seeds, off-season crops, early rice transplanting, and the system ofrice intensification (SFU) that was inventedinMadagascar. 13. Overall, the investments resultedinincreased system productivity'6, from 1.5 at the beginning of the period studied to 2.4 tons of paddy equivalent per hectare, a 60 percent increase and a significant achievement in view of Madagascar's reputation for stagnant productivity (the FA0 figures) and irrigation project failure. Inthese systems, paddy yield" increased from 2.2 to 2.7 tons of paddy equivalent per hectare, contributing to 50 percent of the productivity increase, while the other 50 percent came from the increase incroppingintensity", which grew from 0.6 to 0.8. 14. The latter improvement is a direct consequence of investment in the hardware (better infrastructure improves areas under irrigation) while the former is believed to be the indirect and combined result of investment in the hardware and in promoting agricultural technologies (better infrastructureallows better water control, which encourages the adoption of new methods that have a direct effect on yield). 15. This overall improvement in paddy productivity, when compared with the counterfactual situation-the situationwithout the investments of the past twenty years, including the ones devotedto repair cyclone damage- shows that without investmentsinirrigation infrastructureand inbuildingthe capacityof WAS,the 123,500 hectaresof irrigated perimeters would haveproduced, at the end of the studiedperiod, 140,000 tons ofpaddy less than inthe alternative situation. The differencebetweenthe two scenarios translates into economic benefits for the country that represent a Net Present Value (NPV) of US$200 million at a 10 percent discount rate or an Economic Rate ofReturn (ERR) of 18.5 percent. 16. Therefore, overall, the donor investments in the irrigation sector during the past twenty years, while seen as a failure, in reality have significantly improved country welfare, and, given what is known of the relationship between paddy productivity increases and poverty alleviation they have contributedto poverty alleviation inrural areas as well. 17. However, two factors helped limit the impact of these twenty years of investment in the irrigation sector to the lower bound of what was possible: cyclone damages and the somewhat poor functioning of the WAS.Both kept cropping intensity improvement at a lower level than what was expected. Indeed, 20 percent of the hardware investment during the last twenty years was diverted The projects totaledUS375 millibn but only US$200 million were allocatedto the surveyedperimeters; the other funding went to other perimeters or to irrigatedareas that are not consideredperimeters. 16 The productivity of an irrigated perimeter, for a given year, is the ratio between its annual paddy production and the equipped area (the area that could be irrigatedby the irrigationinfrastructureif it was in perfect condition and sufficiently water-supplied). For a given rice harvest (wet / dry season), the yield is the mean paddy weight per hectare. For nonrice harvest, the mean paddy yield is givenintons of paddyper hectare, by cost-equivalency. For a given year, the cropping intensity is the proportion of irrigated area. The cultural intensity has thus no unit and is between0 and 2 inMadagascar, becauseof the wetldry seasons. 90 from initial objectives to repair cyclone damages. The financing could have been used to expand the irrigated area and thus the cropping intensity. Moreover, WUAs have been losing, on average, 5 percent o f the irrigated area each year because o f a low O&M fee recovery rate. Therefore, an importantpart o f the investment was used, infact, to rehabilitate the former investment. 18. In addition, paddy yield improvement could also have beenmore important if compare with green revolution technologies achievement in other country such as Indonesia that share the same natural conditions than Madagascar and reached 4 tons per hectare. In the Madagascar case, tiny extension projects combined with relatively low producer prices (compared to fertilizer prices) could give areasonableexplanation ofthat relatively low improvement. Basis of the WatershedManagement Project EconomicAnalysis 19. The economic analysis i s carried out separately for each o f the four watersheds selected by the project because the initial conditions of the irrigation systems and upper watersheds are different along with the amount and balance between the three components that will be applied to each o f the watersheds. The results are addedto provide the economic analysis for the whole project. 20. The type and magnitude o f the expected incremental economic benefits o f the project depend on what would have beenthe situation inthe absence oftheproject and on what the project will affect. Thus, the counterfactual situation is described and defended below before the different categories of expectedbenefits from the project are presented. a) Counterfactual Situation (baseline) 21. The baseline describes the evolution o f paddy productivity in irrigated areas and uplands (tanety) inthe four watersheds inthe absence o f the project. As far as paddy productivity i s concerned, the economic analysis builds on the database assembledduring the ESW, extracts cropping intensity, the paddy yield and the un-irrigated production, and simulates their progression over the next 25 years if no investment in the rehabilitation of infrastructure is realized (even if cyclones and floods occur during the period), no capacity building o f local institutions is done, and no disaster risk financing mechanisms are put in place. The evolution o f tanety production is estimated from data collected by the task team duringproject preparation. Table2: Productionat the beginningofthe project 22. In the four areas targeted by the project, paddy production in irrigated areas is currently at around 53,000 tons per year, or 2 percent o f the estimated current national paddy production (around 2.8 million tons). More than two-thirds o f the paddy production comes from well-irrigated areas, located upstream from the irrigation systems, as mentioned in the introduction, and one-third o f the paddy production comes from partially-irrigated areas, located downstream from the irrigation systems. Almost 38 percent o f the paddy production comes from the Sahamaloto perimeter located in the Aalotra watershed, which covers only around 28 percent of the irrigated areas concerned by the 91 project. In addition to this paddy production, which comes from the 21,800 hectares of irrigation systems that will be targeted by the project, around 44,000 tons per year are produced on the 7,700 hectares of tanety located in the lowland parts of the watersheds and the 1,075 hectares of irrigated perimeters that are cultivated off-season and that will targeted by the project. These so-called tanety and off-season irrigated productions are composed of 45 percent cassava, 13 percent sugarcane, 12 percent tomatoes, 10 percent maize and 6 percent rainfed rice. The remaining 14 percent are sweat potatoes, potatoes, bananas, peasand groundnuts. Watershed Irrigated Partially-irrigated Off-season Tanety (tons) Areas (tons) Areas (tons) Irrigated (tons) Marovay 2.5 1.5 5 5.9 Itasy 2.9 1.9 10 3.9 Andapa 2.3 1.3 5 5.2 Lac Alaotra 3.3 2.3 5 3.7 WeightedAverage 3 2 8 5 24. Inthe partially-irrigated areas, productivity hasbeen estimated, by hypothesis, at minus 1ton per hectare compared to well-irrigated areas in the same watershed. This result is derived from the ESW: Theimpact ofpublic spending onperimetersproductivity, 1985-2004that shows arough gain of 1ton whenwater managementis improvedwith investmentincivil works. 25. Off-season irrigated productivity is at a weighted average of 8 tons per hectare, 10 tons per hectare for Itasy and 5 tons per hectare inthe 3 other watersheds. 26. Inthe absenceoftheproject,paddyproductivity inpartially-irrigated areas is likely to decline as well as the productivity of uplands, the former because of declining water control on well-irrigated upstream areas and the latter because of soil erosion and the related loss of nutrients. Off-season irrigatedproductivity and areas are likely to remainthe same. 27. The analysis undertakenin: The impact of public spending on perimeters productivity, 1985- 2004 shows that, during the last twenty years, when any irrigation scheme does not receiveinvestment inhardware and does not face any external shock, it on averageloses around 5 percent of its irrigated area (cropping intensity) per year. The analysis also shows that this loss of 5 percent of irrigated area per year is quite homogeneous all over the country over the studiedperiod. Therefore, if there is no investment in the irrigation systems, 5 percent of well-irrigated areas of these schemes will be transformed each year into partially-irrigated areas and will therefore show a productivity loss of around 1ton per hectare per year. As far as tanety are concerned, inMadagascar, other analyses show that soil erosion is important and results in a productivity loss that can be estimated roughly at 5 percentper year. 28. The first analysis shows also that around 50 percent of the 123,500 hectares of irrigation systems studiedwere found to be sensitive to cyclonic floods during the last twenty years. The capital cost of these floods was US$23 million for an area of 65,000 hectares, or US$350per hectare for the whole period and US$17.5 per hectare per year. The amount was thus not used to directly improve productivity but to ensure that these 65,000 hectares would continue to be irrigated. As a result, they would have amuchlower productivity than current levels". By hypothesis, it is possible that, without any work, the downstream part of the watershed would have become non- irrigatedandthat the productivity loss would havebeenanother 1ton per hectareofpartially-irrigatedarea after acyclone. 92 29. With all of this information about the irrigated areas and the diminution of uplands productivity without investment in the irrigation and watershed management project, the progression of production inthe counterfactualsituation can be reconstructedfor the project's period itself as well as for the next 21 years so that the analysis would cover atotal of 25 years (2007-2031). The evolution i s shownbelow infigure 2. Figure 1: "No Investment" Model ofthe Irrigation Systems andTanety's ProductionEvolution 100 ,000 Tonnes per year 80 - 60 - 0 1 , , , I , , , , I , I I , I , 8 8 8 , , , , , , I 2007 2013 2019 2025 2031 .......IrrigatedAreas ----PartiallyIrrigated Areas Tanety -Total -- -Off-season irrigated 30. Without the irrigation and watershed management project, the paddy production of the irrigated areas that would havebeenconcernedby the project would decrease from around 54,000 tons per year at the beginning of the period to 43,000 tons by the end of the period (without taking into account the impact of cyclonic damageson productivity), or a loss of 21percent. At the same time, the tanety production would decrease from 35,000 tons to 11,000 tons, or a loss of 69 percent. The loss in paddy production alone is equivalentto the milledrice consumptionof 12,000 households for an entire year. b) Origins and Order of Magnitude of Economic BeneJits Associated with theProject 31. The approachto estimating the economic benefits of the irrigation and watershed management project is mostly traditional. Most of the investment aims to provide the software and hardware infrastructurenecessary to generate increased agricultural development, enabling the four watersheds to support greater productivity both in irrigated areas (well, bad and off-season) and in lowlands (tanety) around irrigation systems. 32. Thus, the economic benefits of the project are based on projected agricultural production increases in the four watersheds, compared with agricultural production in the situation without the project. This additional agricultural production will come from additional paddy from well- and partially-irrigated areas, potatoes and tomatoes that are cultivated off-season in irrigated areas, and from uplands where crops such as cassavaandmaizeare cultivated. 33. Paddy productivity will increase in well-irrigated areas as a result of investment in irrigation infrastructure(software and hardware). The productivity increase will be mainly driven by cropping intensity improvement, and marginally by paddy yield improvement. Measured for a 20-year period and an area of 100,000 hectares (see section 2 of this analysis), the productivity gain was around 0.9 ton per hectare. With this project, one can expect an additional 1ton per hectare (therefore an average productivity of 4 tons per hectare in the areas concerned by the project; see table 3) one year after investment in irrigation infrastructure because the project will focus on the rehabilitation of the upstreampartsofthe irrigation systems. 93 34. Inpartially-irrigated areas, paddy productivity will increase as a result of the introduction of agricultural technologies adapted to low water control situations (mainly using the rice variety called Sebota) combined with extension services and better access to credit. As a result, paddy yield will increase significantly without improving water management inthese areas. Current experiments inthe Lac Alaotra watershed show very promising results. With this project, the expected gain will be 1ton per hectare one year after the introduction of the technology, meaning an averageof 3 tons per hectare compared to 2 tons par hectare (see table 3). 35. Off-season, the area cultivated will increaseby a factor o f more than4, from 1,075 hectaresup to 4,700 hectares, andproductivity will reach 10 tons by the end o f the project inAndapa and Alaotra, andwill reach 15 tons perhectare inItasy. 36. Concerning the tanety surrounding the irrigation systems, the increased productivity of the other crops will come from better soil management, which will reduce nutrient depletion. The introduction of agro-ecological techniques will likely improve the average productivity o f crops that are grown around the systems concerned by the project by 50 percent by the end o f the project, meaning an average gain o f 2 tons per hectare. 37. Additional benefits will come from the reduction o f erosion in tanety, where agro-ecological techniques will be introduced. Soil erosion and their translation into scheme siltation vary greatly form one region to another; therefore, the magnitude o f the benefit will only be estimated through order of magnitude. As a hypothesis, soil erosion will be reduced by 5 tons per hectare 3 years after techniques have been introduced, which means that 0.45 tons of sediment will not have to be removed form the irrigatedareas, at US$2.50 per ton, or around US$l per hectare. 38. Additional benefits will come from soil erosion control and natural resource management (pasture and forests) in the upper parts o f the watershed concerned by the project. This may help, in the mediumterm, to mitigate cyclone damage to irrigation infrastructure. For the last 20 years, the capital costs have been estimated at US$17.50 per hectare o f irrigation schemes. By hypothesis, with this project, one can expect to reduce capital cost in the irrigation schemes that are concerned by the project by an amount o f US$5 per hectare the first year after project completion, increasing by an additional US$l each year. 39. Consequently, the major categories o f incremental economic benefit from the project will be (i)additionalpaddyproductioncomingfromimprovementincroppingintensityinwell-irrigatedareas andinyield linkedto introduction ofnewvarieties inpartially-irrigated areas, (ii) additional crop other production coming from a reduction in nutrient depletion, and (iii) lower O&M costs coming from a reductioninsiltation and incyclone damages. Table 4. Categories and order of magnitude of expected incremental benefits Category Location Origin Increment Paddy Productivity well-irrigated schemes Mainly cropping 1ton/hectare/year after rehabilitation intensity Paddyproductivity partially-irrigated yield 1ton/hectare/year after introduction schemes Off-seasonirrigated Well-irrigated schemes Area andyield 5 tons per hectareafter 3 years Other Crops productivity tanety yield 2 tons/hectare/yearafter 3 years O&M reduction all irrigatedareas Avoided tanety US$l/hectare/yearstarting after 3 erosion years after technology introduction Avoidedinfrastructure all areas Avoided cyclone avoidedUS$5 per hectareafter 7 years damages repair damages 94 40. Ifthere is a substantial production increase due to the project, there mightbe some foreign exchange gains related to a decrease inrice imports. However, it i s very unlikely, as the areas that are I targeted represent 3.5 percent o fthe nationalpaddyproduction. 41. There might also be some positive externalities, such as reduced deforestation associatedwith improved paddy productivity in the irrigation systems or improved agricultural productivity in the upper watershed, butthey will not be quantified(too difficult for results that are too meager). EconomicCosts 42. The watershed project's economic costs are composed of: (i) fill base costs o f the public the investment2' without taxes2' (from COSTAE3 - see Annex 5 and project files for more details) in agricultural development (component l), the full cost o f investment in irrigation development (ii) without taxes (component 2); (iii) the full base costs of investment in the watershed development without taxes (component 3); (iv) the full base-costso f project management without taxes (component 4); (v) physical contingencies that represent real costs and, unlikeprice contingencies, are included in project economic costsz2,and (vi) incremental recurrent costs. To the extent that the economic costs include policy and capacity building investments which will benfit also future phases o f the APL that have not been accounted for, this will result in underestimation of the net economic benefits from the project. 43. Adverse environmental effects may represent major economic costs. As mentioned in Annex 10 of this PAD, the project has been classified as Category A. Negative externalities associatedwith irrigation infrastructure investmentsare believed to be fixed because they consist o f rehabilitation o f civil works. However, as stated in AnnexlO, the success o f the project in the watersheds that are concerned by the project might present a major environmental risk. Poor migrants from other parts of Madagascar might flock to the watersheds to demand their share of increased agricultural productivity, the expected outcome of this project, and therefore amplify soil degradation and deforestation. The costs o f the project's environmental and social safeguard measures needed to address this migration have been integrated into the various project components, as stated in Annex 10. Moreover, from a country perspective, this pressure might be assimilated by a transfer from the departingwatersheds to the watersheds that are going to be targetedby the project andtherefore becomingneutral. Therefore, these prevention costs will be usedas aproxy for the negative social and environmental externalities. 44. Incremental recurrent costs are recurrent costs specifically generated by the project at completion. Inthis project, incremental recurrent costs are: (i) additional maintenance costs associated with investment inpublic infrastructure to support marketing chains in subcomuonent 1.2, support to private investment; (ii)additional maintenance costs associated with irrigation infrastructure rehabilitation in subcomponent 2.2(b), rehabilitation of irrigation infrastructure; (iii)additional maintenance costs associated with erosion control (mainly retention structure and hedge works) in subcomponent 3.2(a), strategic erosion control; and additional natural resource management costs associated with pasture management and reforestation in subcomuonent 3.2(b), reestablishment of vegetation cover. 45. Inall four situations, the incremental recurrent costs have beenestimated, by hypothesis, at 5 percent o f the total cost o f investmentfor four years without taxes. In table 5 below, they appear as a 21-year sum, inthe last column. *'The part o f investment in agricultural development (around 4.5 million, counted in the project's total costs) that is supported by the private sector i s taken into account inthat component's economic analysis. Benefits will indeed stay in the country. *' Taxes as well as subsidies are transfer payments, not economic costs. When looking at the project form a society's viewpoint, a tax for the project entity is an income for the government. Inthis case, however, taxes will be paid by the project and will be considered then as benefits for the government. *'Physical contingencies represent an amount o f US$1.55 million, or 4 percent o f the total investment cost of the project. 95 46. The economic costs o f project objective achievement are summarized for the four watersheds in table 4 below, which also shows the contribution o f each o f the major cost categories to the calculated aggregate present value o f the project economic cost. The detailed calculations for each of the 4 watersheds are presented in Appendix 9.1. The calculations assume a real discount rate o f 10 percent, a total life o f public investment o f 25 years, and use o f foreign currency (US$) at the border price level. Table 5: Project investmentandrecurrentcosts (US$millions), all watersheds 47. In total, the present value of the irrigation and watershed management project's economic costs will be US$32 million o f which US$27.6 million represent investment costs and US$4.4 million represent recurrent costs. The bulk o f the cost resides in commercial agricultural development and irrigation systems investments, which account for 72 percent o f the project economic cost's present value. Investments in watershed management will represent only 13 percent o f the project economic cost's present value, while costs o f project management and physical contingencies will represent 15 percent. 48. Alaotra and Marovay will represent respectively 32 and 25 percent, while Andapa and Itasy each represent 21 percent o f the total economic cost o f the project (see table 5 below). The difference lies in component 2. This component's economic cost will be 2 times more important for Alaotra than for Itasy. In Alaotra, the high cost i s explained by the importance o f the civil works involved in the rehabilitation o fthe dam. Table 6: Economic Costs per Watershed Type of Costs, PV ($thousand) Marovay Itasy Andapa Alaotra Total Commercial Agricultural Development 2,308 2,312 2,3 12 2,312 9,245 Irrigation Systems 3,390 2,815 2,138 5,799 14,142 WatershedDeveloDment 1.171 1.018 980 903 4.072- - 7 - Project Management 836 882 882 882 3,482 Physicalcontingencies 323 313 250 488 1,374 Total 8,027 7,340 6,563 10,385 31,979 Recurrent Costs 1,119 II 703 II 692 II 1,887II 4,401 49. Recurrent costs will be a significant part o f the project economic cost's present value: 14 percent, representing a yearly flow o f US$0.75 million, immediately after project completion. The bulk o f the recurrent costs will be generated by the incremental maintenance o f the rehabilitated irrigation infrastructures (8 1percent), followed by the incremental maintenance o f erosion works and management o f forest andpasture (12 percent). 50. The Lac Alaotra and Marovay watersheds will account for 68 percent o f the recurrent cost o f the project. The discussion about who will finance the recurrent costs and how is crucial given what has been learned from past irrigation projects, and will therefore focus on these two watersheds and will take place inthe financial analysis section o fthis annex. 96 EconomicBenefits 51. As stated in section 3, the incremental benefits of investment in the irrigation and watershed management project are likely to be threefold: (i)additional paddy production coming from improvement of cropping intensity in well-irrigated areas and improvement in yields linked to the introduction of new varieties in partially-irrigated areas; (ii)additional production o f other crops coming from the development o f off-season irrigation production and from a reduction in nutrient depletion on tanety; and (iii) O&M costs coming from a reduction in siltation and in cyclone lower damages. 52. For each o f these three categories, the amount o f economic benefits that will bebrought by the project depends on the unit value or cost reduction associatedwith each category o f benefits and their importance, as well as their pace o f appearance inconjunction with project investment. a) Incremental Agricultural Production 53. Incremental agricultural value associated with additional crop production i s the difference between the producer price and the cost o f production under the new conditions. For paddy, the unit incremental values are taken at US$82 per ton inwell-irrigated areas and at US$81 per ton inpartially- irrigated areas, with a paddy producer price at 20.7 cents (US$) per kg23.For the other crops, the unit incremental value is a combination of agricultural value associatedwith cassava, tomatoes, beans and rainfedrice; it i s taken at US$70 per ton in irrigated areas and US$71 per tons intanety, reflecting the proportionof crops that are going to be grown with the project ineach o f the four watersheds. 54. The pace o f production growth i s given by the pace o f investment during the project. For example, the pace o f paddy production!growth in well-irrigated areas is given by the distribution of investment of component 2 in each watershed: The production increase is the proportion o f irrigation systems that have their infrastructure rehabilitated times the productivity increase that is associated with infrastructure rehabilitation (+1 tonper hectare). Figure 2: PaddyProductionWith and Without Project ThousandTons 80000 / 70000 gg.--.-....-.__l._.__.l_l-l._. 40000 ------ ------ -.-...-...__... 30000 , , l l l I l I , , , , 1 , I , , , , 2007 2013 2019 2025 2031 -Paddy Productionwith Project - - - - PaddyProduction without Project OtherCrops Production with Project ---------- OtherCrops Production without project 23The producer price is a good proxy of the economic price in this economic analysis, because: (i)the producer price used (20.7 cents per kg) is close to the import parity price derived from the FOB price for rice from Pakistant (US250 per MT or 25 cents per kg) which is at the present time the main source of rice imports in Madagascar, even if rice importsrepresent only 10 percent of the local production but 30 percent of the traded rice. 97 55. Under the conservative assumption that there will not be any productivity gain in paddy production in well- and partially-irrigated areas and in other crop production on tanety after project completion, the irrigated areas that are concerned by the project will produce 25,000 additional tons o f paddy per year at project completion, 50 percentmore than the current situation. Inthe tanety and off- season irrigated areas that are affected by the project, 27,000 additional tons o f other crops will be produced, 60 percent more than the current situation. 56. The cumulated difference o f production in the situation with the project compared to the situation without the project is 105,000 tons duringthe 4 years of project implementation, 55,000 tons of which are paddy, the equivalent o f the annual consumption o f 355,000 people or around 65,000 households. b) Reduction in O Mcosts associated with reduced siltation and avoided cyclone damages 57. The hypotheses regarding the benefits linked to reduced erosion and reduced flood in irrigation systems are very conservative. The first benefits appear proportionally to tanety areas that benefit from the introduction of agro-ecology techniques and stay stable over time. The second, as mentioned earlier, appear beginning in year 8. The first benefit is US$12.5 per hectare o f tanety cultivated with agro-ecology techniques, one year after their introduction, beingstable over time. The second i s US$5 per hectare for irrigation systems after 7 years, then growing at US$1 per year to reflect the growing vegetation inthe upper watershed and their better ability to absorb water and avoid floods inthe downstream part o fthe watersheds. Gross Benefits (US$ thousands, PV) Percentage Paddy productioninwell-irrigatedareas $15,977 39% Paddyproductioninpartially-irrigatedareas $3,973 10% Off-seasonirrigatedproduction $10,061 24% Productionof other crops on uplands $9,712 23% Reductionof siltationin irrigatedperimeters $653 2% Avoided cyclone damagesinirrigatedsystems $1,013 2% Total $41,388 100% 59. Benefits might also be regrouped more or less by component: the increase inpaddy production and off-season cropping inwell-irrigated areas coming mainlyfrom water management, andtherefore component 2 (63 percent); the increase in paddy production in partially-irrigated areas as well as the growth o f the production of other crops ontanety and the related reducedsiltation inirrigation systems from agriculture services, and therefore component 1 (35 percent); and then the reduction of cyclone damagescoming from watershedmanagement (2 percent). Resultsof the Cost-BenefitAnalysis 60. Because of the weakness o f the available data, the cost-benefit estimates presentedbelow are necessarily imprecise and should be considered only in terms o f order of magnitude, especially for 98 recurrent costs, but also for benefits derived from agricultural production on tanety along with reduced cyclones damages. For the latter, the estimates are really conservative given the absence o f data. 61. Using conservative estimates for the unit incremental values (in this analysis, the current producer price o f rice is assumed to remain constant for the next 25 years while international forecasters are predicting an increase over the next 10 years), the pace, and the quantity o f benefits (not all environmental benefits were captured in this economic analysis), as table 3 shows, the o f lo%), corresponding to an economic rate o f return (ERR) o f 14 percent, and is, therefore, justified investment is likely to increase the welfare o f the country by about US$9.5 million (at a discount rate from this point o f view. Table 8: Summary of Costs and Benefits, PresentValues as of 2006 (US$thousand) Avoided cyclone damages 1,013 0 0 0 0 5,293 Project Cost (investmentand recurrent) 31,979 6,009 8,088 12,492 8,852 15,644 Net benefits (ERR=14%) 9,409 -6,009 -7,802 -11,235 -5,940 131,920 62. At watershed level, these calculations show differences among watersheds (see table 9 below and detailed tables inappendix 9.1 at the end o f this annex). Itasy shows greater ERRthan the project as a whole (20 percent), while Marovoay and Andapa show rates o f return o f 13 percent, respectively, and Lac Alaotra a 8 percent ERRwhich is lower than 10 percent that is regarded as the opportunity cost o f capital inMadagascar. 63. In the example of Marovoay, the modest ERR is attributed to investment in agricultural development that concerns a relatively small area o f off-season irrigation compared to the other watersheds (as shown in table 8 by the small NPV o f the third benefit) and by a relatively important investment inthe irrigation development. 64. In the example o f Lac Alaotra, the low ERR is explained by the relative importance o f investment in irrigation infrastructure compared to the irrigated area that is concerned by the investment. It is also the result o f (i) conservative hypQtheses regarding the improvement o f paddy productivity; and (ii) exclusion o f the positive economic and environmental benefits associated with water storage in the dams, including benefits from reduced flooding o f the Lake Alaotra watershed, a designated Ramsar site under the Ramsar Convention on Wetlands. Moreover, Lake Alaotra is the first rice granary o f the island supplying 50 percent of Antananarivo annual consumption. Therefore, the investment in the rehabilitation o f the lake Alaotra perimeter is considered high priority by the Government, even if the economic analysis shows an ERR that is below the opportunity cost o f capital. Table 9: Comparisonof costs and benefitsbetweenthe four watersheds Watershed Benefits/Costs(PV, US$ throusand) Marovay Itasy Andapa Lac Alaotra PaddyProduction inWell-irrigated Areas $3,968 $2,391 $2,235 $7,382 PaddyProduction inPartially-irrigated Areas $1,207 $1,810 $754 $201 Off-seasonirrigated crops $428 $6,422 $2,676 $535 Other Crops Production on Tanety $3,546 $2,828 $2,246 $499 Siltation Reductionin Irrigation systems $170 $204 $127 $153 99 Avoided Cyclone DamagesinIrrigation systems $282 $263 $170 $298 Project Cost (investmentandrecurrent) $8,027 $7,340 $6,563 $10,385 Net Present Value ( ERR) 1,574 (13%) 6,578 (20%) 1,645 (13%) -1,4317 (8%) 65. The variables that influence the flow of project benefits the most are likely to be (i)the evolution o f the paddy producer price, (ii) efficiency o f the W A Sinmaintainingthe rehabilitated the canals, especially in Lac Alaotra, and (iii)the likelihood o f cyclone damages combined with the existence or non-existence of a disaster-risk financing mechanism that will immediately repair cyclone damages. 66. The benefits have been calculated with a paddy producer price at 20.7 cents per kg. During most o f the last twenty years, the producer price stayed relatively constant at 10 cents per kg. During the last two years, however, the producer price increased drastically, to more than 20 cents, inducing changes in farmer behavior that have startedto regain interest inrice production. There is a very small risk that the price will shrink again because the forecasts (UNCNED, FA0 and the World Bank Commodity Group) predict that rice prices will remainhighon the internationalmarket for the next 10 years. If producer price declines by 18 percent and falls below 17 cents per kgnext year (compared the to about 21 cents per kgnow) and stays at this level, the project will not be economically viable. This i s unlikely, however, given trends in the international rice market where a long term increase is forecasted. 67. The inability o f W A Sto maintain the irrigation infrastructure was one o f the reasons why irrigation projects were not able to maintain productivity gains long after irrigation infrastructure was rehabilitated: W A Swere losing on average 5 percent o f the irrigated area o f systems every year after the investment was made. As far as cyclone damages are concerned, the relatively important physical contingencies should be sufficient during project implementation to repair cyclone damages to infrastructure. After the project, repairs will depend on the FERHA or any disaster risk financingthat could be set up duringproject implementation. 68. IfW A Sdo not maintainproductivity on well-irrigated areas for more than 7 years after the project, or if a cyclone hits Marovay's 6,000 hectares o f well-irrigated perimeters after four years o f project implementationwithout beingrepaired, the project's NPV is zero. 69. The bottom line of this analysis is that, except from the evolution of the producer price, which i s clearly outside the project's scope and can vary, the future flow o f project benefits will be conditioned by the sustainability o f investment inirrigation and, to a lesser extent, by the sustainability o f investment in agro-ecology techniques in the lower parts of the watersheds and in natural resource managementand erosion control inthe upperparts o f the watersheds. 70. The project's outcome will greatly depend on the sustainable financing mechanisms that will be put in place to increase the recovery rate of the O&M fee, to cope with cyclone damages and, to some extent, to pay the lower and upper watershed farmers for the environmental services they will provide ifthey maintain natural resources, and control erosion. Financial Analysis/FiscalImpact 71. The main winners in the irrigation and watershed management investment will be the local ruralpopulations (around 90,000 households mostly living under the absolute poverty line) o f the four watersheds (Marovay, Itasy, Andapa and Lac Alaotra), especially farmers producing paddy with good water control. Net benefits for farmers will be around US$22.6 million (the sum o f the four benefits accruing from additional agricultural production, less what is given to communes and to traders), or 51 percent o fthe present value o f the project's benefits. The other net gainers o f the project are the traders in agricultural products that will receive US$11.8 million, or 26 percent of the present value of the project's benefits. 100 72. Under the conservative hypothesis that there will not be any additional fiscal revenues generated from the project other than the US$8.5 million intaxes generatedby the project investment plus the taxes along the value chain between paddy producers and consumers, the government is the main loser inthis simple analysis becauseit does invest the bulk o f the $28 million o f the project cost (part of which is supported by the beneficiaries) and receive a fiscal compensation o f only $8.5 million, or 19 percent o f the total benefits, in return for its investment (11 percent for communes through additional production, and 8 percent to the central government through taxes associated with the project's investment). 73. WUAs will gainUS$1.7 million from the reduced costs of siltation and from someprevention o f cyclone damages, but will also bear most o f the project recurrent costs. An additional US$4.4 million o f recurrent costs will come both from irrigation rehabilitation maintenance and natural resource management in the upper watershed. Therefore, they will lose US$2.7 million from the project. This cost, which will happen mostly after project completion, will represent a flow o f around US$0.75 million a year, or an additional US$33 per hectare o f well- andpartially-irrigated areas where the project will intervene. Therefore, inorder for the project to be sustainable, there will be the needto transfer US33 per hectare from irrigated agriculture farmers to WUAs so that the WUAs can pay for the additional maintenance cost. Given that the farmer will gain around $50 per additional ton of paddy and around $35 per additional ton of other crops produced, this option sounds feasible. 74. The existing low rates o f O&M recovery and the relative failure o f past projects to improve it mean that this project should work on institutional arrangement, involving, for example, the communes, and putting inplace enforcement mechanisms to make sure that this transfer will happen. Taxes at the commune level are one possibility. 75. Cyclone damages on irrigation infrastructures are another type o f recurrent cost that needs to be financed by project for sustainability reasons. Part of the project's benefitwill be a certain amount o f cyclone damage prevention through better management o f the upstream part o f the four watersheds; this is especially true for Marovay and Andapa. However, the improvements won't be enough to prevent major damages from particularly violent cyclones that might occur randomly during the next twenty years. Therefore, part of the farmers' net benefits should be transferred to finance insurance mechanisms at the WUA level (for transferable infrastructure) and to fund the FERHA (for non- transferable infrastructure). Conclusions 76. In sum, the analysis shows that the expected incremental economic rents, based on several assumptions about the counterfactual pace o f irrigation infrastructure degradation, soil erosion, and cyclonic damages, on the one hand, and the additional agricultural production, on the other hand, are robust enough to justify the proposed investments by the Malagasy government, even if the numbers themselves are not very high. However, assumptions on additional agricultural production are conservative, as well as the incremental benefit associated with them. Moreover, some benefits o f new production, such as rent, which will be generated along the value chains between producers and consumers, have not beentaken into account inthis analysis. 77. The main beneficiaries will be the about 500,000 people that will see their income grow through gains in agricultural productivity. However, in order to make these gains sustainable, the project will have to put inplace transfer mechanisms from these farmers (particularly those who work inthe irrigation systems) to WUAs andto the FERHA inorder to fund infrastructure maintenance and to insure them against cyclone damages. Inaddition to this first transfer, it is also possible to envisage payment for environmental services from WUAs to farmers to help prevent soil erosion in the lower and upper parts o f the watersheds. 101 References 78. World Bank. 2005. Madagascar: The impact o f public spending on irrigated perimeters productivity (1985-2004).The World Bank: Washington, D.C. Brand (J.) March2006. Synthkse des ktudes de prkparationpar site d'intervention. Processed. World Bank Institute. 2001. Economic Analysis of Investment Operations. Analytical tools andpracticalapplications. WBI development studies. The World Bank: Washington, D.C. Minten B. and Barret C. 2005. Agricultural technology, productivity, poverty and food security inMadagascar.Processed. IN.Janvier2006.Renforcementdescapacitksdespartiesprenantesdansleprojetbasins versants perimeters irrigues. Processed. Horus. Decembre 2005. Etude de prkparation du projet basins versants-pkrimbtres irriguks. Etudedes filikres. Processed. 102 Appendix 9.1: ProjectEconomic Costs,EconomicBenefits andNPV per Watershed Table 10: Marovoay Watershed, Project Economic Costs (US$thousands) Table 11: Itasy watershed, ProjectEconomic Costs (US$ thousands) Table 13:Alaotra Watershed,Project Economic Costs (US$thousands) 103 Table 14: Marovay, Summary of costs andbenefits, NPV as of 2006 (US$ thousands) Type of Benefit(US%Thousand/ year) PV 2007 2008 2009 2010 2011-2031 Paddyinwell-irrigatedareas 3,968 0 77 183 348 13,914 Paddyinpartially-irrigatedareas 1,207 0 0 27 80 4,101 Off-seasonirrigation 428 0 0 10 28 1.455 Other Crops on tanety 3,546 I 0 1 0 1 77 I 208 I 12,988 Siltationreduction II 170 I 0 1 3 1 10 I 17 I 525 Avoided cyclone damages 282 0 0 0 0 1,475 ProjectCost (investmentandrecurrent) 8,027 1,623 2,09 1 2,956 2,17 1 3,978 Net benefits (ERR=13%) 1,574 -1,623 -2,011 -2,649 -1,491 30,479 Table 16: Andapa, Summary of costs andbenefits, NPV as of 2006 (US$ thousands) Table 17: Alaotra, Summary of costs andbenefits,NPV as of 2006 (US$thousands) 104 Annex 10: Safeguard Policies Issues EnvironmentalAssessment Category and Safeguard Policies triggered 1. The Madagascar Irrigation and Watershed Management Project has been classified as a "Category A" operation under the World Bank environmental screening procedures specified in OP 4.01. The package of safeguard documentspreparedfor the project comprises three primaryreports: (i) the Regional Environmental and Social Assessment (RESA) containing and Environment and Social Management Plan (ESMP); (ii)the Pest and Pesticide Management Plan (PPMP), and; (iii)the Resettlement Policy Framework (RPF). The RESA, PPMP and RPF address the World Bank Safeguard Policies that are triggered by the project. The proposed activities for management and mitigation of the Project impacts are incompliance with the following World Bank Safeguard Policies: Environmental Assessment Policy OP/BP 4.01, Natural Habitat Policy OP/BP 4.04, Forests Policy OP/BP 4.36, Involuntary Resettlement OP/BP 4.12, and Pest Management OP/BP 4.09. Analysis of alternatives 2. Landdegradation in Madagascar has been extensive and dramatic. It has led to a significant reduction in agricultural productivity, exacerbation of rampant natural erosion by human caused erosion and widespread poverty o f the rural population. The no-project alternative will lead to a deterioration of the existing situation, expansion of the area o f low agricultural productivity leading to the destruction of globally important biodiversity resources (e.g. Marojejy National Park, the South Anjanaharibe Special Reserve, and the Makira Conservation Site all located in the upper watersheds around the Andapa irrigation scheme; the Ankarafantsika National Park located inthe upper Marovoay watershed; and the Lac Alaotra Ramsar site) and will lead over time to abandonment of many rural areas. 3. The only feasible project alternative i s the presently chosenproject design. The present project will address in an integrated manner the land degradation in four major irrigation schemes and their associated watersheds and reduce the pressure on globally important biodiversity resources. The present project design has as objective to increase agricultural production in an environmentally and socially sustainable manner, stop the expansion of the agricultural area in the project sites through intensification and to reduce rural poverty, which i s expected to lead to a reduced rural to urban migration. Environmental and Social Impacts 4. The environmental and social impacts o f the project are mostly positive. Environmental and social management measures are almost fully integrated into the design o f the various project components. The promotion o f agro-ecological production techniques are expected to increase agricultural productivity and increase farmer's incomes, and to stabilize or reduce erosion and land degradation, and over-time reduce sediment loading in the irrigation schemes. It i s also expected that agricultural intensification in the watersheds will lead to reduced pressure on the high biodiversity sites in the upper and lower watersheds. 5. A major environmental riskwill be the success ofthe project inthe watersheds. Poor migrants from other parts of Madagascar might flock to the watersheds to demand their share o f increased agricultural production. This might increase the land pressure to former unsustainable levels and exacerbate human induced erosion and it might also increase deforestation in the globally important biodiversity sites in the upper watersheds and increase the clearance o f reed marshes for rice production in the Lac Alaotra Ramsar site. Transfer of the management o f these sub-watersheds to local farmer organizations will needto provide a social fencingsystem to prevent the entry of migrants from elsewhere. 105 6. Intensification o f agricultural production normally goes hand in hand with increased use o f chemical fertilizers and pesticides. To manage the health and environmental impacts o f increased pesticide use, the borrower has prepared a Pest and Pesticide Management Plan (PPMP). This PPMP envisages strengthening the capacity o f the PlantProtection Service on the Regional level (DRDRs) to increase the oversight and control o f pesticide use and improve awareness among farmers and pesticide distributors. The PPMP also envisages strengthening the development and implementation o f Integrated Pest Management (IPM) practices. Agro-ecological practices require more inputs: herbicides and fertilizers. The question is can farmers afford this? These agro-ecological practices reduce the risks for farmers during droughts. This makes the farmers less vulnerable to climate variability. Environmentaland SocialManagement Plan(ESMP) 7. Environmental and social management measures and their costs have been integrated into the various project components. An overview o f these environmental and social management measures is presentedinthe table below. Contractor EMP 8. The contractors who will be awarded the contracts for the rehabilitation o f the irrigation schemes need to prepare their own Environmental Management Plans (Contractor EMPs). These EMPs need to specify how the contractors will handle occupational health and safety issues, in compliance with IFC Occupational Health and Safety Guidelines, during construction and how hydrocarbons (waste oils), solid and liquidwastes will be handled, where their workers will be housed, training and means to prevent HIV/Aids infections o f their workers and local communities. The contractors should have a license to establish and operate the quarries and after use should rehabilitate these quarries to acceptable international standards. The establishment, operation and rehabilitation o f the quarries should benegotiatedwith the local communities. Agro-industries 9. The project will promote the use o f ago-industries, such as rice mills and related processes, biodiesel production from Jatropha seeds, oil palm and groundnuts industries (crushing, oil refining, soap and meal production), cashew nut processing, fruit juice and pulp processing plants (citrus, mangoes and litchis). These ago-industries are essential for economic growth, but also could be very polluting. They therefore need to comply with applicable Madagascar pollution control standards or with applicable World Bank Group pollution guidelines as described in the Pollution and Prevention and Abatement Handbook (PPAH) and the IFC's Environmental, Health and Safety Guidelines. These guidelines are: Food and Beverage Processing Guidelines, Fruit and Vegetable Processing Guidelines, General Environmental Guidelines and Vegetable Oil Processing Guidelines. The standards which are the most stringent, would apply. 106 U 0 z L B (3 0 .I Y .E!ed EU Y a E -- Y m 2 B Y I ed .I 8V w 8 I 8 Y ed e E .I * W (3 I ed .I Y B Y 3 3 3 3 3 3 3 n 3 IA 33 a a %1 2 2 k 0 9 3 cd U 0 3 a w rcr 0 % as H lA $8 0 Y $ a3 .-iz rcr c&I 0 Ym U ResettlementIssues 10. In order to protect the rights o f vulnerable groups and farmers who might lose land or income or lose access to other natural resources a Resettlement Policy Framework (RPF) has been prepared by the borrower. If certain project activities require resettlement, land acquisition or certain people lose income or access to natural resources a Resettlement Action Plan (RAP) will be prepared in compliance with the World Bank Policy on Involuntary Resettlement (OP 4.12) to ensure that these people don't become poorer then they were before the project intervened. A Resettlement Action Plan (RAP) or a small Environmental Assessment (EA) might be needed in case check dams, anti-erosion structures, mini dams, markets or other infrastructure will be built. The Technical Secretariat o f the Matching Grant Mechanism, to be financed under the project, will screen sub-projects and identify if a RAP andor a small EA study as part of the feasibility will be needed. 11. The project will look carefully into the position o f share croppers in the irrigation schemes, where share cropping i s more common and in the watersheds where share cropping i s less common. The project will take care that the capacity o f the private operators i s not strengthened at the expense o f the smallholders (marginalization o f vulnerable groups). ESMPImplementationandMonitoring 12. The implementation and the monitoring of the ESMP will need to be carried out per region. One o f the Technical Assistance attached to the DRDR and to be financed under the project, needs to be qualified in environmental and social management issues and will be responsible for the implementation and monitoring o f the implementation o f the ESMP. CommunicationPlan 13. Communication between the different project components i s fundamental for an adequate implementation o f the project and to build synergies. One o f the Technical Assistance in the DRDR financed by the project needs to be responsible for the communication between the components, but also for communication with other regions and the national level and the media. 111 Annex 11:Project Preparation and Supervision Planned Actual PCN review June 28,2004 June 28,2004 Initial PID to PIC October 22,2004 October 22,2004 Initial ISDS to PIC August 25,2004 August 25,2004 Appraisal June 6,2006 June 6,2006 Negotiations September 13,2006 September 21,2006 BoardRVP approval November 14,2006 Planned date o f effectiveness February28,2007 Planned date o f mid-termreview December 31,2008 Planned closing date March 1,2011 Key institutions andpersons responsible for preparation o f the Project: Ministry of Agriculture, Livestock and Fisheries, Bruno Rakotomahefa, Rad0 Rakotondralambo, MariusRatolojanahary (General Secretary) andPhilibert Rakotoson(General Director). Bankstaffandconsultants who worked onthe Project included: Name Title Unit IJsbrand de Jong Team Task Leader, Sr. Water Resources Spec. AFTs2 GervaisRakotoarimanana Sr. Financial Management Specialist AFTFM Sylvain Rambeloson Sr. Procurement Specialist AFTPC LovaNiainaRavaoarimino Procurement Analyst AFTPC Wolfgang Chadab Finance Officer LOAG2 GillesVeuillot Counsel LEGAF Erika Styger Natural ResourceManagement Specialist (cons) AFTS1 Robert Robelus Sr. Environmental Assessment Specialist AFTS1 Arbi Ben-Achour Sr. Social Scientist AFTS1 JeanPaul Chausse Lead Specialist AFTS 1 Sofia U.Bettencourt Lead Operations Officer AFTS 1 Ziva Razafintsalama Sr. Rural Development Specialist AFTS1 FritsOhler Watershed Specialist FA0 HermannPfeiffer Agricultural ExtensionSpecialist consultant Patrick Labaste LeadAgricultureEconomist AFTS4 Jean-Christophe Carret Natural ResourcesEconomist AFTS4 Franco Russo Senior Program Assistant AFTS1 Caroline Guazzo ProgramAssistant A F T S 1 Bank funds expendedto dateon Project preparation: 1. Bankresources: US$1,250,567 2. Trust funds: 0 3. Total: US$1,250,567 Estimated Approval and Supervision costs: 1. Remainingcosts to approval: US$44,000 2. Estimated annual supervision cost: US$90,000 112 Annex 12: Documentsin the Project File BankReports Aide-memoire - April 2004 mission Aide-memoire - July2004 mission Aide-memoire - June 2005 mission 0 Aide-memoire - March 2005 mission 0 &de-memoire - November 2005 mission Aide-memoire - June2006 mission PreparationStudies WorkingPapers - Irrigation andWatershed Management Policy Letter Document de travail sur l a Securisation Fonciere, version provisoire, December 2005 Renforcement des capacites des parties prenantes dans le projet Bassins Versants Pkrimbtres Irrigues, July 2005 Etude des Filibres, December 2005 Land Titles, Investment, and Agricultural Productivity inMadagascar, October 2005 Land and Property k g h t s Review, Draft Synthesis of the Preparatory Studies on Intervention Sites - Lac Aloatra Synthesis of the Preparatory Studies by Intervention Site -Andapa site Synthesis of the Preparatory Studies by Intervention Site - Itasy site Synthesis o f the Preparatory Study by Intervention Site- Marovoay Site Analyse Institutionelle et Juridique du Programme Bassins Versants Pkrimbtres Irrigub, October 2005 Cadre de Politique de Rkinstallation-March 2006 Evaluation Environnementale et Sociale Rkgionale - February 2006 Plan de Gestion des Pestes et des Pesticides-March 2006 113 Annex 13: Statement of Loansand Credits Difference betweenexpected I andactual Original Amount inUS$Millions IBRO I IDA I SF I GEF I Cancel.I Undisb. I disbursements Ong. I I PO74448 2004 MG-Gov & lnst DevTAL 0.00 30.00 0.00 0.00 0.00 16.24 4.69 0.00 (FY04) PO82806 2004 MG-Transp lnfrastr Invest Prj 0.00 150.00 0.00 0.00 0.00 103.14 59.86 29.53 (FY04) PO76245 2003 MG-Mineral Res Gov SIL 0.00 32.00 0.00 0.00 0.00 11.39 -4.93 0.00 (FY03) PO73689 2003 MG-RuralTransp APL 2 0.00 80.00 0.00 0.00 0.00 38.00 3.49 -4.35 (FY03) PrevAPL (FY02) PO72160 2002 MG-Priv Sec Dev 2 (FY02) 0.00 23.80 0.00 0.00 0.00 8.60 4.37 -3.14 PO55166 2001 I MG-Corn Dev Fund SIL 0.00 178.00 0.00 0.00 0.00 34.18 - 3.25 (FYOI) I56.75 I I (FYOI) PO51741 I 2000 I MG-HealthSecPram Sud 2 I - 0.00 I 58.00 I 0.00 I 0.00 I 0.00 I 13.18 I -5.74 I 3.74 (FYOO) PO52186 1999 MG-Microfinance(FY99) 0.00 16.40 0.00 0.00 0.00 1.43 0.76 0.41 PO01568 1998 MG-Communitv Nutrition2 0.00 37.60 0.00 0.00 0.00 1.80 -8.83 0.00 STATEMENT OF IFC's HeldandDisbursedPortfolio InMillions ofUSDollars Committed Disbursed IFC IFC FY Company Loan Equity Quasi Partic. Loan Equity Quasi Partic. Approval 1997 AEF GHM 0.46 0.00 0.00 0.00 0.46 0.00 0.00 0.00 1995 AEF Karibotel 0.19 0.00 0.00 0.00 0.19 0.00 0.00 0.00 BFV-SocGen 6.37 0.00 0.00 0.00 0.00 0.00 0.00 0.00 1991 BNI 0.00 2.09 0.00 0.00 0.00 2.09 0.00 0.00 2005 BNI 6.37 0.00 0.00 0.00 0.00 0.00 0.00 0.00 2000 BOA-M 0.00 0.82 0.72 0.00 0.00 0.82 0.72 0.00 2004 BP Madagascar 0.00 3.51 0.00 0.00 0.00 0.00 0.00 0.00 CREDIT LYONNAIS1 6.37 0.00 0.00 0.00 0.00 0.00 0.00 0.00 114 Approvals PendingCommitment FY Approval Company Loan Equity Quasi Partic. 2001 Besalampy 15.23 0.00 0.00 0.00 2006 IDA-IFC PCG 12.20 0.00 0.00 0.00 Total pendingcommitment. #IO ~~~ 115 Annex 14: Countryat a Glance Sub- POVERTYand SOCIAL Saharan LOW. Madagascar Afrlca Income IDevelopmentdiamond' 2005 Population,mid-year(millions) 16.6 741 2,353 GNIper capita (Atlasmethod, US$) 290 745 580 Lifeexpectancy GNI (Atlasmethod, US$billionsj 5.4 552 1,364 Averageannual growth, 199945 T Population(%) 2.8 2.3 1.9 Laborforce (%J 3.1 2.3 2.3 Gross per primary Most recent estlmate (latest year avallable, 199945) capita Poverty(% ofpopulationbelownationalpovefly line) 71 Urbanpopulation(% oftofalpopulationj 27 35 30 Lifeexpectancyat birth(yearsJ 56 46 59 Infantmortality(per 1.0001ivebirlhsJ 76 100 80 Chiidmalnutrition(% of childrenunder 5) 42 29 39 Access to improvedwater source Access to an improvedwater source(% ofpopulation) 46 56 75 Literacy(% ofpopulationage 15+) 71 62 Grossprimaryenrollment (% ofschool-agepopulation) 134 93 104 -Madagascar Male 136 99 110 Low-incomegroup ~ Female 131 87 99 KEY ECONOMIC RATIOSand LONG-TERMTRENDS i985 1995 2004 2005 Economic ratios'0 GDP (US$ billions) 2.9 3.2 4.4 5.0 Gross capitalformation/GDP 8.5 10.9 24.3 22.4 Exporhof goods andservices/GDP 12.2 24.1 32.6 25.6 Trade Gross domesticsavings/GDP 0.3 3.4 7.6 7.7 Gross nationalsavings/GDP -0.2 0.9 13.5 10.9 Currentaccountbalance/GDP -6.2 -8.4 -10.8 -11.4 Interestpaments/GDP 1.4 0.5 0.7 Capital Total debVGDP 86.2 136.1 79.3 savingsDomestic formation Total debt serviceiexports 38.1 7.1 5.6 Presentvalueof debVGDP 40.6 Presentvalue of debthxports 123.9 Indebtedness 1985-95 199545 2004 2005 200549 (average annualQDWthJ GDP 1.0 2.6 5.3 4.6 6.6 -Madagascar GDP per capita -1.8 -0.1 2.4 1.8 4.6 Low-incomegroup Exportsof goodsand services 4.4 0.6 1.5 8.1 12.4 STRUCTURE of the ECONOMY :::: :E (% ofGDPJ 1985 1995 2004 2005 Agriculture Industry t::: 2::; Manufacturing 11.3 7.9 14.2 14.1 Services 51.5 64.1 55.2 56.0 Householdfinalconsumptionexpenditure 90.0 89.9 82.7 64.2 Generalgov'tfinal consumptionexpenditure 9.8 6.7 9.6 8.1 lmpOflS Of goods and services 20.5 31.7 49.2 40.3 1985-95 199545 I (average annualQrOWthJ IGrowthof exports and imports('A) 1 Agriculture 2.2 1.9 3.1 industry 1.0 2.6 6.6 Manufacturing 0.4 3.5 6.1 Services 1.0 2.6 6.0 Householdfinal consumptionexpenditure 0.6 3.3 3.2 Generalgov'tfinal consumptionexpenditure -1.2 3.1 -6.6 Grosscapitalformation 1.7 10.4 57.6 2.3 -Imports ImportsOf goods and services -Expork 0.7 8.6 24.8 18.5 Note:2005 data are preliminaryestimates. This tablewas producedfrom the DevelopmentEconomicsLDB database. * The diamondsshowfour keyindicatorsin the country(inbold)comparedwith its income-groupaverage.If data are missing,the diamondwill be incomplete. 116 Madagascar ~ PRICESand GOVERNMENTFINANCE 1985 1995 2004 2005 Domesticprices lnflatlon ('4 i (% change) Consumerprices 10.6 49.1 13.8 18.5 implicitGDP deflator 10.4 45.1 14.3 18.3 Government flnance (% of GDP, includescurrentgrants) Currentrevenue 13.7 9.0 16.0 12.3 Current budgetbalance 7.0 -1.6 3.4 1.4 -GDP deflator +CPi Overallsurplus/deficit 1.o -7.9 -9.1 -8.7 TRADE 1985 1995 2004 2005 (US$ mi//ions) Export and Import levels (US$ mlll.) Total exports (fob) 291 566 864 996 I 2.0wT Coffee 22 59 62 79 Vanilla 44 41 108 102 Manufactures 42 296 653 736 Total imports (ci0 466 739 1,634 1,733 Food 50 67 83 79 Fueland energy 85 81 204 253 Capitalgoods 98 141 300 293 Exportprice index (2000=100J 85 97 85 71 Importprice index (2000=100J 65 97 147 154 Tens of trade (2000=100) 130 100 56 46 BALANCEof PAYMENTS 1985 1995 2004 2005 (US$ mi//ions) Current accountbalanceto GDP (YO) Exportsof goods and services 350 803 1,423 1,291 Importsof goodsand services 569 992 2,143 2,030 Resourcebalance -219 -169 -720 -739 Net income -112 -158 -79 -84 Net current transfers 98 80 330 249 Currentaccount balance -233 -267 -470 -674 Financingitems(net) 217 328 488 576 Changes in net reserves 16 -61 -18 -1 Memo: Reservesincludinggold (US$ millions) 49 110 489 495 Conversionrate (DEC,/ocaWS$) 132.5 853.1 1,868.9 2,003.0 EXTERNALDEBT and RESOURCEFLOWS 1985 1995 2004 2005 (US$ mii/ionsJ Total debtoutstanding and disbuned 2,520 4,302 3,462 IBRD 28 12 0 0 IDA 316 1,110 2,269 2,298 E: 380 F:e G:4 Total debtservice 150 56 81 IBRD 3 5 0 0 IDA 3 15 27 48 Compositionof net resourceflows Officialgrants 40 164 859 Officialcreditors 133 70 431 Privatecreditors 5 -4 -2 Foreigndirect investment(net inflows) 0 10 45 Portfolio equity (netinflows) 0 0 0 World Bank program Commitments 73 65 87 A IBRD - E Bilateral - Disbursements 58 76 227 221 B. IDA D.Other multilateral F Private . Principalrepayments 2 11 11 29 C. IMF G Short-tern ~ Netflows 56 65 217 191 Interestpaments 4 9 16 18 Nettransfers 52 56 200 173 Note:Thistablewas producedfrom the DevelopmentEconomicsLDB database. 6113/06 117 Annex 15: Incremental Cost Analysis 1. This section discusses the incremental costs eligible for GEF funding for the "Irrigation and Watershed Management Project", defined as the difference between the GEF alternative scenario and the IDAbaseline. For each o f the four components o f the project, the section will: 0 Identify the baseline, Describe what would happen ifthe baseline is implemented, 0 Indicatethe costs o f the baseline, 0 Describe the alternative scenario, e Describe the expected benefits under the alternative scenario, 0 Report the cost o f the alternative andthe incremental cost. 2. The relationship between the activities o f each component and the environmental benefits generated i s synthesized inthe below tables. The Incremental Cost Matrix is reported at the end of the section. As most o f the decisions, practices and technologies that the beneficiaries o f the project will adopt cannot yet be determined, the analysis favors a qualitative approach. Component1: Developmentof Commercial Agriculture The objective for this component is to lay thefoundations for improved market access and sustainable intensijkation and diversiJication of irrigated and rainfed agricultural systems in theproject's watersheds. 3. Baseline This component will promote agricultural development inlowland and upland areas. The aim will be to improve (i) to market and marketing systems in order to reduce costs and increase access farm gate prices, (ii) value through diversification into higher added value products and agro- added processing, (iii)capacities o f farmers (male and female), farmers groups and professional organizations, and (iv) agricultural productivity through better access to extension, improved technology, inputs, and credit. The component includes two sub components: one involving activities that largely depend on public/collective initiative; the other one depending essentially on demand from stakeholders. 4. Expectedresultsunder the baselinescenario The results expected under this component will be the increase in number o f producer organizations, unions, and federations o f active producers, the increase in the volume o f credit allocated to agricultural investments, an increase in the proportion o f products marketed by local households, an increase in the quantity o f seed and fertilizer sold to producers, and an increase in the number o f contracts signed and executed between producers and the private sector, and an increase in the volume o fproducts marketed inthis way. 5. Baseline cost: US$9,960,000 (7,450,000 US$ IDAand2,510,000 US$ Beneficiaries) 6. GEF alternative scenario(OPE): GEF funding will contribute to assuring that intensificationand diversification of agricultural production will be based on ago-ecological principles. These are based on improved organic matter management through improved rotations, cover crops, improved fallows, agroforestry technologies and diversified and locally adapted varieties and crops. This will lead to improved above-ground and below-ground carbon sequestration, increase o f agrobiodiversity within the cropping systems and reduce pressure on natural habitats, and thus secure important global environmental benefits. Funding will be used to assure that high quality technical assistance is provided and adjusted to the specific environmental conditions o f the four project zones. Furthermore, the GEF grant will be used for training o f technicians and farmers in the agroecological techniques and principles, and for the testing and adaptation o f these techniques in farmers' fields. Special attention will be paid to upland systems that are based on slash-and-bum agriculture (tavy), which causes deforestation, threats to biodiversity, 118 carbon loss and soil fertility loss. The improvement will need more time and effort then the systems downstream area that lend themselves better to agricultural intensification. Intensive on-farm technology adaptation will assure that farmers' constraints and needs are adequately taken into account and addressed. Expectedloc and globalbenefitsunder the GEF alter itive (OP15) 4ctivities Direct imDact and local environmental Global environmental benefits benefits Technical assistance, Improvedlocal capacity (technicians, Increase in carbon sequestration, trainingo f extension agents and farmers) in soil carbon, above-ground carbon technicians and implementingagro-ecological farming (through cover cropping, relay farmers, and on-farm techniques cropping, agroforestry) research of agro- Improvedagricultural production based on Increase inagrobiodiversity ecological o Technical improvement through agro- (through diversification) and production ecological and agroforestry techniques. below-groundbiodiversity techniques o Improvedsoil fertility management and (through improved soil organic nutrient recyclingthrough organic matter matter status) management, Reducedenvironmental o Improvedprotection of soils through soil degradation and pressure on coverage and erosion control with natural habitats for agricultural vegetative measures fields (avoided deforestation) due o Increasedagro-biodiversity through to satisfactory and increased increase o f locally adapted varieties, agricultural productionon crop diversification (annual and existing fields; resultingin perennial) o Reducedcarbon emissions o Improvedcrop rotationand integrated o Protection ofecosystems and pest management possible restoration o f Diversification of agricultural production ecosystem integrity system Improved ecological resilience of agricultural system, with improved resistance to climate variability Available alternative farming techniques to slash-and-burn practices, through agro- ecological techniques, improvednutrient cycling and targeted inputs, agroforestry and horticulture Reduction o f pressure on forests, and Drotects biodiversitv 9. GEF Alternativecosts: US$12,680,000 (IDA,Beneficiaries andGEF) 10. Incremental cost: US$2,720,000 GEF. The incremental cost will finance the technical assistance to the project, training of technicians and farmers, and adaptation o f new techniques through on-farm research. Component2: IrrigationDevelopment The objective o f this component i s to lay thefoundations for improved management, maintenance and sustainability of irrigation services provision in four large-scale irrigation schemes through rehabilitation of irrigation infrastructure, capacity strengthening of stakeholders and clarification of roles and responsibilities, and establishment of an appropriate incentivepamework. 11. Baseline The component will contribute to improving the quality o f irrigation services and operation and maintenance (O&M) o f the irrigation schemes. The project will finance the rehabilitation o f irrigation and appurtenant infrastructure, including technical design studies, implementation o f works and their supervision. In addition, the project will fund the participatory preparation o f a Scheme 119 Development Plan (SDP) and an annual Performance Contract (PC), negotiated between (F) W A S , the Communes and Regions, and MAEP. The project will also provide support to stakeholders during implementation o f the PC, including capacity strengthening, development o f a strategy for mobilization o f water users, annual evaluation o fperformance indicators anduser satisfaction surveys. 12. Expected results under the baseline scenario: Expected results concern the rehabilitation of the irrigation infrastructure and improved capacity o f water users association to operate and maintain the infrastructure. This will lead to increased surface o f fields under irrigation for the rainy and dry season. In addition, a number of second phase Performance Contracts will be signed, and the O&M costs will be recovered as percentageo f overall O&M needs at 100percent at the end o f the project. 13. Baseline cost: US$17,470,000 (US$15,670,000 IDA andUS$1,800,000 Beneficiaries) 14. GEFalternative scenario (OP15): IDA funding will be used for irrigation rehabilitation (infrastructure work) and capacity strengthening o f water users associations for the management o f the irrigation schemes. There will be no additional GEF hnding to this component. Aspects o f interests to GEF, such as environmental management in relation to agricultural improvement i s covered under component 1 and the environmental management at the watershed or landscapelevel with global environmental impacts are found under component 3. 15. Expected local and global benefits under the GEF alternative (OP15) Environmental benefits with significant impact on irrigation schemes will be created through GEF incremental funding under component 1, 3 and 4. Reduced sedimentation of irrigation infrastructure (which reduced O&M costs) will be a result from overall GEF increment. 16. GEF Alternative costs: US$17,470,000 (US$15,670,000 IDA and US$1,800,000 Beneficiaries) 17. Incremental cost: US$O GEF Component 3: Watershed Development The objective o f the component i s to lay thefoundations for sustainable management of watersheds including irrigated and rainfed agriculture, the conservation of the natural heritage, and improved productivity of the natural resources. 18. Baseline: This component will finance the a) planning and capacity building for the sustainable management o f watershed and b) investments for watershed protection. The project will finance technical assistance to prepare a watershed managementplan for each o f the four project zones. For this an additional 0.77 million US$will be provided through a PPF. It will include (i)zoning and description o f land use systems, ecosystems, settlements, institutions and partners, (ii)strategic analysis o f erosion problems for downstream sedimentation and natural resources degradation; (iii) a specific and detailed analysis to define project activities, and (iv) establishing a baseline for monitoring and evaluation o f component results. The project will also support land tenure security through the installation of intercommunal `Land Tenure Window, that assist in recording non-titled property rights, regularize land rights and secure secondary landrights. 19. The project will also invest in watershedprotection. The planning will have identifiedthe 'hot spots' o f erosion that have a significant impact on downstream irrigation infrastructure. Through participatory negotiations, local strategies will be developed for controlling erosion, halting gullies and reducing the quantity of sedimentstransported to downstream irrigation areas. The project will finance the setup of such strategic anti-erosion works favoring biological methods and techniques. Possible mechanical works will be built, favoring local manpower. 120 20. Expected results under the baseline scenario: Successful implementationo f this component will result in4 master watersheds plans that will provide a diagnosis o f natural resources and identify pathways o f interventions for sustainable land and water management at the watershed level. In addition, hot-spot erosion will be identified, strategies developed for their control and erosion control works implemented preferably with the participation o f concerned stakeholders. Through improved landtenure security, farmers will be more willing to invest into andprotect their landfrom degradation. 21. Baseline cost: US$1,910,000 (US$l,820,000 IDA and US$90,000 Beneficiaries) 22. GEF alternative scenario (OP15): GEF contributionwill complement IDAfunding by addressinglonger-termenvironmentaland landdegradation issues at the watershed level, that negatively impact lowland anduplandagricultural production systems as well as global environmental goods and services. Most important degrading landuses are pasturemanagementbasedonperiodic burning, extensive agriculturalpractices basedon slashing (primary forest or fallow vegetation) and burningto produce food crops such as upland rice. Additional destructive forest extraction practices concern logging, charcoal production, firewood collection, over-extraction o f NTFP, and hunting o f lemurs and small mammals. These activities contribute to natural resource degradation, depletion o f vegetation cover and biodiversity. (see also Annex 16 for land degradation analysis). Often, these extensive land use practices do not even allow farmers to achieve satisfactory incomes. 23. GEF funding will be used to address these landdegradation issues through aparticipatory and integrated approach, and will provide technical assistanceto develop landuse alternatives that should encourage local population to take responsibility and engage in the sustainable management of their natural resources. The approach will include: a) establishing a participatory zoning with the stakeholdersat the sub-watershedlevel to determine optimal landuse according to topography, current land use and land rights, diagnosis o f soil fertility and soil production potential, location and characteristics of water sources and streams, and the origin andpathways o f erosion; b) environmental awareness raising campaigns; c) training and capacity strengthening in alternative sustainable NRM practices according to stakeholders' needs; d) provision o f support to community to obtain landrights (GELOSE) and of technical assistance to preparenatural resourcemanagementplans, and e) provision of support to environmental and other communicationand negotiation platforms that influence natural resources management at the watershed level. In addition, interventions will be targeted at increasing vegetation cover on communal land and to a lesser extent on private land, which includes improved pasture management without fire, afforestation and reforestation, natural regeneration o f native vegetation, and provision to protect natural habitats (forests, wetlands, lakes) and associated biodiversity; 24. Expected local and global benefits under the GEF alternative (OP15) DirectimDact and local Globalenvironmental benefits environmentalbenefits 1.2. Participatory watershed Increasedawareness of Improvedcapacity of management plans stakeholders on stakeholders to integrate the environmental issues at the creation of global watershed level and improved environmental benefits into capacity of stakeholders for their activities. This will environmentally sensitive result inthe design and decision-making and planning implementation o f that impacts environmental participatory watershed conditions at the watershed management plans where the level positively, and at the creation of global same time provides local environmental benefits at the 121 stakeholderswith watershed level will be environmentalservices that consciously integrated(such improvelandproductivity and as soil (carbon) protection, livingconditions biodiversityconservation, water resourcesprotection, increasedcarbons sequestrationthroughSLM etc) 1.3. Supportto environmental Informationexchange Improvedinformation communicationplatforms betweenstakeholders exchanges favors Allows for harmonizingof coordinationand approaches andcreating collaborationand allowsfor synergies betweendonors and strategic decision makingby projects especiallybetween various stakeholdersto environmentalprojects and addressglobal environmental developmentprojects inareas issues, such as biodiversity with highbiodiversityand conservation, habitat natural habitats protection,and carbon Allows for coordinated sequestration interventionswithin the projectarea on environmental andrural development activities 1.4. Awareness campaigns, Improvedknowledgeand Increasedknowledgeand trainingandcapacity capacityinregards to land awarenesson global strengtheningon environmental degradationimpacts as well as environmental issues at the issues existingalternativesby local and regionallevelwill o Ruralpopulation allow for strategic decision o Localandregionalstaff makingby various (technical services, NGOs) stakeholders(rural 9Newly createdor reinforced development, environment, NRMfarmers groups or privatesector etc) to engage associations with improved in SLM activities that create capacity global environmental benefits (carbon sequestration,increase in agro-biodiversity) 1.5. Supportto of community Secured communityland Maintainecosystem's basedlandtenure security rightswill provideincentives integrity through sustainable for improvedNRMpractices extractionand harvest of Establishedmanagement products from natural habitats plansprovidecommunities Protect biodiversityby guidelines on volumes for maintaininghabitats extraction, management Avoided deforestationdue to practices, and inform on long- community landrights term productivityofresources (avoided carbonloss) Stimulates environmental stewardshipof communities Will improveproductivity and profitability ofNR use. 2.2. Revegetationof communal Plantedfodder grassesand Improvedabove andbelow land(pastures, reforestation, improvedpasture ground carbon sequestration protectionof natural forests) management will contribute (fodder grasses, reforestation) to Avoided carbon loss (pasture o improvedcattlenutrition fires, deforestation,reduced andproductivity,which forest product extraction) enables improvedcrop and Regenerationofnative livestockintegration vegetationincreases above o regenerationof vegetation and below ground o reducedsheet erosion biodiversity 122 Reforestationwill contribute Reducedpressure on primary to improved forests, leadsto improved o Fuelwoodandconstruction protectionof wood supply o biodiversity o Erosioncontrol o importantenvironmental Regenerationof natural regulatoryservices such as vegetationwill water source protection o providemultiple products (Marovoay) for extraction(fuelwood, medicinalplants, wild fruits and other food plants) o reintroduce native biodiversitywithin productionlandscape Protectionofnaturalhabitats will contributeto o Biodiversityconservation of manyendemic and endangeredspecies o Protect ecosystem regulatoryservicesand functions 27. GEF Alternative costs: US$4,330,000 (IDA, Beneficiaries andGEF) 28. Incremental cost: US$2,420,000 GEF Incremental costs will be occurring for awareness campaigns and information exchange, technical assistance to communities and local NGO and technical staff, participatory processes for innovation development, testing and adapting proposed technologies on farm, dissemination of improvedtechnologies, participatorymonitoring o f development processes. Component 4: Project Management The objective of this component is to manage and use resources in accordance with the project's objectives andprocedures, and toput inplace apolicyframework that isfavorable to upscaling of the project at the national level. 29. Baseline: Management of the project, including (a) provision o f technical assistance, training, oflce equipment and vehicles, minor oflce upgrading works, auditing and evaluation studies, and incremental operating costs in support o f project management, (b) overall project planning, quality oversight, procurement, financial management, and monitoring o f project activities; and (c) outsourcing of quality oversight through independent financial and technical audits, and evaluation of project activities. Project management will encompass all four target watersheds as well as national level coordination. Project monitoring will be undertaken at internal and external levels. This component will also includesupport to nationalpolicies. 30. Expected scenario under the baseline scenario: Successful implementation of this component will result in efficient implementation arrangements, effective oversight, monitoring and evaluation o fproject activities. 31. Baseline cost: US$3,450,000 (IDA) 32. GEF alternative scenario (OP15) GEF funding will contribute to the project monitoring and evaluation system by financing the satellite images and their interpretation to monitor the global and environmental indicators in order to assess impact o f project activities on land degradation, carbon sequestration, biodiversity, habitat protection, and area under SLM. Inaddition, a community-based monitoring systemwill be supported. GEF will also contribute technical assistanceto M&E andthe project implementationteam. 123 Activities DirectimDact Local and globalenvironmental benefits Designingandimplementinga Improvedcapacity ofproject Quantificationof environmental M&Esystemto monitor localand staff andimproved benefits globalenvironmentindicators, understandingofthe to be includedineconomic andprovisionof supportto underlyingcauses,processes analysis of the project projectimplementationteam and dynamics associated Inform global community, with landdegradation policy makers,research, and Improvedenvironmental development communities on informationsystem and projectoutcome. environmental indicators State-of-the-Artknowledge will be available at local level 35. GEFAlternative costs: US$4,310,000 (GEF+IDA) 36. Incremental cost: US$860,000 GEF Incremental costs will cover the reinforcement of the M&E systemwith GIS andthe participatorymonitoring at the locallevel. Incremental Cost Matrix 37. The incremental costs are calculated as the difference between the GEF alternative scenario and the IDA baseline scenario. The results are reported inthe matrix below. As most o f the decisions, practices and technologies that the beneficiaries o f the project will adopt cannot yet be determined, the analysis favors a qualitative approach. Component1 Category Estimated LocalBenefit GlobalBenefit Expenditures (US!
Группа Всемирного банка · Project Appraisal Document
Madagascar - Irrigation and Watershed Management Project
Открыть оригинал документа
Полный текст размещён на сайте публикующей организации. lawenc.com индексирует метаданные и ведёт на официальный источник.
Полный текст
Основные сведения
Организация
Группа Всемирного банка
Тип документа
Project Appraisal Document
Страна
Мадагаскар
Источник
Всемирный банк