Document of The World Bank Report No: ICR2677 IMPLEMENTATION COMPLETION AND RESULTS REPORT (IDA-35240 IDA-3524A IDA-45250 TF-26222) ON A CREDIT IN THE AMOUNT OF SDR 69.2 MILLION (US$ 89.05 MILLION EQUIVALENT) AND AN ADDITIONAL FINANCING CREDIT IN THE AMOUNT OF SDR 18.5 MILLION (US$ 30 MILLION EQUIVALENT) TO THE REPUBLIC OF MADAGASCAR FOR A RURAL DEVELOPMENT SUPPORT PROJECT June 27, 2013 Agriculture and Rural Development Sustainable Development Department Country Department AFC01 Africa Region CURRENCY EQUIVALENTS (Exchange Rate Effective December 31, 2012) Currency Unit = Malagasy Ariary (MGA) MGA 1.00 = US$ 0.00044 US$ 1.00 = MGA 2,272.98 FISCAL YEAR January 1 to December 31 ABBREVIATIONS AND ACRONYMS AF Additional Financing ASP Agriculture Sector Program CAS World Bank Country Assistance Strategy CDD Community-driven Development CROA Regional Committee for Orientation and Allocation CSA Agriculture Support Center DER Regional Executive Directorates FCRA Competitive Fund for Agriculture Research FOFIFA National Center for Applied Research for Rural Development GTDR Regional Rural Development Working Groups GOM Government of Madagascar IRR Internal rate of Return KPI Key Performance Indicators MAP Madagascar Action Plan MAEP Ministry of Agriculture, Livestock and Fisheries MTR Mid-term Review NGO Non-governmental Organization NPV Net Present Value OCC Opportunity Cost of Capital OP Original Project PAD Project Appraisal Document PADR Rural Development Action Plan PIU Project Implementation Unit PSDR Rural Development Support Project Vice President: Makhtar Diop Country Director: Haleh Bridi Sector Manager: Severin Kodderitzsch Project Team Leader: Ziva Razafintsalama ICR Team Leader: Anna Corsi MADAGASCAR Rural Development Support Project CONTENTS Data Sheet A. Basic Information B. Key Dates C. Ratings Summary D. Sector and Theme Codes E. Bank Staff F. Results Framework Analysis G. Ratings of Project Performance in ISRs H. Restructuring I. Disbursement Graph 1. Project Context, Development Objectives and Design ............................................... 1 2. Key Factors Affecting Implementation and Outcomes .............................................. 7 3. Assessment of Outcomes .......................................................................................... 17 4. Assessment of Risk to Development Outcome......................................................... 24 5. Assessment of Bank and Borrower Performance ..................................................... 25 6. Lessons Learned ....................................................................................................... 27 7. Comments on Issues Raised by Borrower/Implementing Agencies/Partners .......... 28 Annex 1. Project Costs and Financing .......................................................................... 29 Annex 2. Outputs by Component ................................................................................. 31 Annex 3. Economic and Financial Analysis ................................................................. 37 Annex 4. Bank Lending and Implementation Support/Supervision Processes ............ 40 Annex 5. Beneficiary Survey Results ........................................................................... 42 Annex 6. Stakeholder Workshop Report and Results................................................... 45 Annex 7. Summary of Borrower's ICR and/or Comments on Draft ICR ..................... 47 Annex 8. Comments of Cofinanciers and Other Partners/Stakeholders ....................... 49 Annex 9. List of Supporting Documents ...................................................................... 50 MAP A. Basic Information Rural Development Country: Madagascar Project Name: Support Project IDA-35240,IDA- Project ID: P051922 L/C/TF Number(s): 3524A,IDA-45250,TF- 26222 ICR Date: 04/18/2013 ICR Type: Core ICR GOVERNMENT OF Lending Instrument: SIL Borrower: MADAGASCAR Original Total XDR 69.20M Disbursed Amount: XDR 82.47M Commitment: Revised Amount: XDR 82.61M Environmental Category: B Implementing Agencies: Ministry of Agriculture Cofinanciers and Other External Partners: B. Key Dates Revised / Actual Process Date Process Original Date Date(s) Concept Review: 02/14/2000 Effectiveness: 09/20/2001 09/20/2001 11/14/2002 08/06/2004 06/27/2006 06/06/2007 04/02/2008 Appraisal: 02/13/2001 Restructuring(s): 12/17/2008 06/04/2010 02/08/2011 06/30/2011 11/27/2012 Approval: 06/19/2001 Mid-term Review: Closing: 06/30/2007 12/31/2012 C. Ratings Summary C.1 Performance Rating by ICR Outcomes: Moderately Satisfactory Risk to Development Outcome: Substantial Bank Performance: Moderately Satisfactory Borrower Performance: Moderately Satisfactory i C.2 Detailed Ratings of Bank and Borrower Performance (by ICR) Bank Ratings Borrower Ratings Quality at Entry: Moderately Satisfactory Government: Moderately Satisfactory Implementing Quality of Supervision: Satisfactory Moderately Satisfactory Agency/Agencies: Overall Bank Overall Borrower Moderately Satisfactory Moderately Satisfactory Performance: Performance: C.3 Quality at Entry and Implementation Performance Indicators Implementation QAG Assessments Indicators Rating Performance (if any) Potential Problem Project Quality at Entry Yes None at any time (Yes/No): (QEA): Problem Project at any Quality of Yes Satisfactory time (Yes/No): Supervision (QSA): DO rating before Moderately Closing/Inactive status: Satisfactory D. Sector and Theme Codes Original Actual Sector Code (as % of total Bank financing) Agricultural extension and research 13 23 Agro-industry, marketing, and trade 21 15 Central government administration 18 12 Irrigation and drainage 21 21 Other social services 27 29 Theme Code (as % of total Bank financing) Improving labor markets 40 35 Other rural development 40 40 Participation and civic engagement 20 25 E. Bank Staff Positions At ICR At Approval Vice President: Makhtar Diop Callisto E. Madavo Country Director: Haleh Z. Bridi Hafez M. H. Ghanem Sector Manager: Severin L. Kodderitzsch Joseph Baah-Dwomoh Project Team Leader: Ziva Razafintsalama Ivar Ted Serejski ICR Team Leader: Anna Corsi ii ICR Primary Author: Anna Corsi F. Results Framework Analysis Project Development Objectives (from Project Appraisal Document) The objective of the Rural Development Support Project (RDSP) is to increase incomes and reduce poverty in rural areas, while preserving the natural resource base. The project is part of the Rural Development Action Plan (Plan d'Action pour le Développement Rural - PADR), a broad-based program approved by the Government in 1999 to promote sustainable growth in agricultural production, foster food security and enhance access to basic services in the rural areas. The project would support demand-driven activities in agricultural production and technology transfer, and strengthen capacity at national, regional and community levels. Revised Project Development Objectives (as approved by original approving authority) Even though the PDO was not changed, there were considerable attempts to make it more measurable. As such, Key Performance Indicators evolved over the life of the project, maintaining the spirit and overall level of achievement of original ones (see paragraph 1.3.1). At the time the project was prepared, there was no formal “Logframe� structure. As such, the PDO and the absence of intermediate indicators were consistent with the design standards of Bank projects at the time. Formal intermediate outcome indicators were introduced at the time of the Additional Financing (2008). (a) PDO Indicator(s) Original Target Formally Actual Value Values (from Revised Achieved at Indicator Baseline Value approval Target Completion or documents) Values Target Years Indicator 1 : Average revenue increase among beneficiaries (%) Value quantitative or 0 30% 38% Qualitative) Date achieved 02/19/2001 09/25/2001 12/31/2012 Achieved: 127% of target. 38% increase for beneficiaries compared to stagnation Comments in revenues for non-beneficiaries for the same time. Indicator was reformulated (incl. % as part of 2011 project restructuring maintaining spirit and achievement level of achievement) original one. Infrastructure functioning one year after completion and mechanism in place to Indicator 2 : finance operations and maintenance (%) Value quantitative or 0 85% 90% Qualitative) Date achieved 06/19/2001 02/08/2011 12/31/2012 Comments Partially achieved: 90% of infrastructure functioning one year after completion, (incl. % however, arrangements for O&M remain unclear. Indicator introduced as part of iii achievement) 2011 restructuring to improve measurement of PDO. Indicator 3 : Subprojects functioning one year after completion (%) Value quantitative or 0 85% 90% Qualitative) Date achieved 06/19/2001 02/08/2011 12/31/2012 Partially Achieved: 90% of subprojects financed under the Additional Financing Comments are using the agricultural techniques acquired under the project one year after (incl. % completion. 80% of subprojects financed under the original project are still achievement) operational one year after completion. Indicator introduced as part of 2011 restructuring to improve measurement of PDO. Subprojects having satisfactorily developed and implemented an Environmental Indicator 4 : Action Plan (%) Value quantitative or 0 70% 91% Qualitative) Date achieved 06/19/2001 02/08/2011 12/31/2012 Comments Achieved: 130% of target. 91% of producer organizations apply the (incl. % environmental measures recommended by the EAP. Indicator introduced as part achievement) of 2011 restructuring to improve measurement of PDO. Indicator 5 : Direct project beneficiaries (number) of which female (%) Value quantitative or 0 1,060,000 (29%) 1,043,650 1,190,295 (34%) Qualitative) Date achieved 06/19/2001 02/08/2011 11/27/2012 12/31/2012 Comments Achieved: 114% of target. The project benefitted a total of 1,190,295 (incl. % beneficiaries, 34% of those are women. Core indicator introduced as part of 2011 achievement) restructuring. (b) Intermediate Outcome Indicator(s) Original Target Actual Value Formally Values (from Achieved at Indicator Baseline Value Revised approval Completion or Target Values documents) Target Years Indicator 1 : Average financial rate of return of subprojects (%) Value (quantitative 0 15% 21% or Qualitative) Date achieved 06/19/2001 09/25/2001 12/31/2012 Comments Achieved: 140% of target. Impact evaluation studies show an average IRR of (incl. % 36% for rice subprojects, 32% for peanuts subprojects, 28% for maize achievement) subprojects. Indicator 2 : Additional area provided with irrigation and drainage services (ha) Value (quantitative 0 16,000ha 14,300ha 30,069ha or Qualitative) Date achieved 06/19/2001 09/23/2008 11/27/2012 12/31/2012 iv Comments Achieved: 210% of target. Indicator reformulated for alignment with mandatory (incl. % CORE Indicator. Target was revised as part of partial cancellation of IDA credit achievement) in November 2012. Indicator 3 : Water users provided with irrigation and drainage services (number) Value (quantitative 0 28,000 25,280 31,842 or Qualitative) Date achieved 06/19/2001 02/08/2011 11/27/2012 12/31/2012 Comments Achieved. 125% of target. Introduced as mandatory core indicator. Target was (incl. % revised as part of partial cancellation of IDA credit in November 2012. achievement) Indicator 4 : Water user associations that are operational (number) Value (quantitative 0 711 643 657 or Qualitative) Date achieved 06/19/2001 02/08/2011 11/27/2012 12/31/2012 Comments Achieved: 102% of target. Introduced as mandatory core indicator. Target was (incl. % revised as part of partial cancellation of IDA credit in November 2012. achievement) Production of key crops (rice, maize, peanuts) financed through subprojects Indicator 5 : increased (% and tons) Rice: 30%; 64,200 Rice: 37%; 64,971 tons tons Value Maize: 20%, 4,600 Maize: 24%, 3,282 (quantitative 0 tons tons or Qualitative) Peanuts: 20% Peanuts: 32% 1,242 1,350 tons tons Date achieved 06/19/2001 09/25/2001 12/31/2012 Comments Achieved. For maize and peanuts, achievement of 71% and 92% respectively of (incl. % tons targets. achievement) Area cultivated with improved technologies and inputs provided through project- Indicator 6 : supported subprojects (ha) Value (quantitative 0 39,150ha 42,927ha or Qualitative) Date achieved 06/19/2001 09/25/2001 12/31/2012 Comments (incl. % Achieved. 110% of target. achievement) Indicator 7 : Infrastructure built or rehabilitated (number) Dams: 711 Dams: 643 Dams: 657 Storage facilities: Storage Storage facilities: Value 183 facilities: 165 169 (quantitative 0 Processing units: Processing Processing units: or Qualitative) 203 units: 190 205 Slaughterhouses: Slaughterhous Slaughterhouses: 28 31 es: 30 v Date achieved 06/19/2001 09/25/2001 11/27/2012 12/31/2012 Comments Achieved, except for slaughterhouses target: 93% of target. Targets were revised (incl. % as part of partial cancellation of IDA credit in November 2012. achievement) Indicator 8 : Subprojects financed and executed (number) 6,200 (4,000 under the original Value projects, 2,200 (quantitative 0 9,510 under the or Qualitative) additional financing) Date achieved 06/19/2001 09/25/2001 12/31/2012 Comments (incl. % Achieved: 153% of target. achievement) Indicator 9 : Saving schemes established and functioning (number) Value (quantitative 0 6,065 6,467 or Qualitative) Date achieved 06/19/2001 09/25/2001 12/31/2012 Comments (incl. % Achieved: 107% of target. achievement) Indicator 10 : Diagnostic study on the national agricultural research system completed Value (quantitative 0 No No or Qualitative) Date achieved 06/19/2001 09/23/2008 12/31/2012 Comments Not achieved. A preliminary diagnostic study has been prepared as a basis for the (incl. % development of the national agricultural research strategy. achievement) Indicator 11 : Research studies developed (number) Value (quantitative 0 36 26 or Qualitative) Date achieved 06/19/2001 09/25/2001 12/31/2012 Comments Not achieved. Ten research projects were cancelled as a result of the partial (incl. % cancellation of IDA credit in November 2012. achievement) Indicator 12 : National Germplasm Bank rehabilitated and material at risk regenerated Value (quantitative 0 Yes Yes or Qualitative) Date achieved 06/19/2001 09/23/2008 12/31/2012 Comments (incl. % Achieved. achievement) Indicator 13 : ACS established and functioning (number) vi Value 1 ACS to be 1 ACS established (quantitative 0 established and 40 and 21 or Qualitative) to be strengthened strengthened. Date achieved 06/19/2001 09/23/2008 12/31/2012 Comments Partially achieved. 19 ACS could not be strengthened due to difficulties in the (incl. % procurement process. achievement) Indicator 14 : CROA established and functioning (number) Value (quantitative 0 16 16 or Qualitative) Date achieved 06/19/2001 09/23/2008 12/31/2012 Comments Achieved: 100% of target. 12 CROA were established with project financing and (incl. % 4 with funds from other donors. achievement) Indicator 15 : Targeted clients satisfied with agricultural services (%) Value (quantitative 0 90% 94% or Qualitative) Date achieved 06/19/2001 02/08/2011 12/31/2012 Comments Achieved: 104% of target. Beneficiaries showed satisfaction for monitoring of (incl. % subprojects, participation of producer organization members, inputs obtained, achievement) and amount of financing received. Introduced as mandatory core indicator. Indicator 16 : Client days of training provided (number) Value (quantitative 0 727,180 753,216 or Qualitative) Date achieved 06/19/2001 02/08/2011 12/31/2012 Comments (incl. % Achieved: 104% of target. Introduced as mandatory core indicator. achievement) G. Ratings of Project Performance in ISRs Actual Date ISR No. DO IP Disbursements Archived (USD millions) 1 12/21/2001 2.71 2 05/01/2002 Satisfactory Satisfactory 2.81 3 06/11/2002 Satisfactory Unsatisfactory 2.81 4 11/20/2002 Satisfactory Unsatisfactory 5.26 5 02/19/2003 Satisfactory Unsatisfactory 7.47 6 06/13/2003 Satisfactory Satisfactory 10.63 7 12/12/2003 Satisfactory Satisfactory 15.39 8 05/28/2004 Satisfactory Satisfactory 25.56 9 07/21/2004 Satisfactory Satisfactory 27.22 10 02/10/2005 Satisfactory Satisfactory 39.37 vii 11 10/18/2005 Moderately Satisfactory Unsatisfactory 59.13 Moderately 12 12/29/2005 Moderately Satisfactory 60.84 Unsatisfactory Moderately 13 06/30/2006 Moderately Satisfactory 70.24 Unsatisfactory 14 12/11/2006 Moderately Satisfactory Satisfactory 76.64 15 06/26/2007 Satisfactory Satisfactory 83.76 16 12/19/2007 Satisfactory Satisfactory 86.91 17 05/31/2008 Satisfactory Satisfactory 90.66 18 11/25/2008 Satisfactory Satisfactory 94.59 19 04/29/2009 Satisfactory Satisfactory 96.58 20 12/07/2009 Satisfactory Satisfactory 96.58 21 06/24/2010 Satisfactory Satisfactory 96.58 22 11/29/2010 Moderately Satisfactory Moderately Satisfactory 103.27 23 05/10/2011 Moderately Satisfactory Moderately Satisfactory 110.68 Moderately 24 01/10/2012 Moderately Satisfactory 116.77 Unsatisfactory 25 10/31/2012 Moderately Satisfactory Moderately Satisfactory 126.89 H. Restructuring (if any) ISR Ratings at Amount Board Restructuring Disbursed at Restructuring Reason for Restructuring & Approved Restructuring Date(s) Key Changes Made PDO Change DO IP in USD millions Components were consolidated; (ii) implementation arrangements were improved through the inclusion of NGOs as strategic partners for the identification, design and implementation of agricultural production and off-farm sub- projects; (iii) project management was enhanced 11/14/2002 N S U 5.17 through the appointment of a project management team; (iv) due to liquidity problems, the authorized allocations of Special Accounts A and B were revised; (v) due to increasing counterpart financing problems, the disbursement percentage of expenditure Category 1 (Grants) was increased to 100 percent, while the disbursement viii ISR Ratings at Amount Board Restructuring Disbursed at Restructuring Reason for Restructuring & Approved Restructuring Date(s) Key Changes Made PDO Change DO IP in USD millions percentage of Categories 2, 3, 4 and 5 (Civil Works; Equipment, vehicles and materials; Consultants’ services, training and audits; and Operating Costs respectively) was increased to 100 percent for about one year (until December 2003); (vi) support to the environmental agencies was eliminated; and (vii) intermediate outcome indicators were brought in line with the consolidated components and project targets were adjusted by reducing the targeted number of subprojects from 1,000 to 800, the number of families benefited from 180,000 to 150,000, the percentage of subprojects geared to benefit women Reallocations of credit proceeds as follows: SDR 13.6 million to “Grants� for agricultural and nonagricultural subprojects, reflecting satisfactory performance and strong demand for subprojects; SDR 1.3 08/06/2004 S S 27.48 million to “Equipment, vehicles and materials;� SDR 3.8 million to “Operating costs� reflecting increased costs of targeting poor families in isolated communities the need to carry out a full- fledged agricultural census. SDR 11.58 million was reallocated to “Grants� for productive subprojects to permit the project to finance the continuing strong demand for 06/27/2006 MS MU 70.24 sub-projects following the rice intensification campaign; and a reallocation of SDR 12,500 to “Operating Costs;� (ii) project activities were amended to ix ISR Ratings at Amount Board Restructuring Disbursed at Restructuring Reason for Restructuring & Approved Restructuring Date(s) Key Changes Made PDO Change DO IP in USD millions clearly reflect the nature of the 2005 cyclone repairs and rice intensification campaigns; and (iii) component outcome indicators were slightly changed to make them more relevant and in line with the PDO. Closing Date of the OP was extended by 18 months to permit the finalization of all remaining subprojects and training and disbursement of the 06/06/2007 MS S 83.66 remaining Credit balance; (ii) SDR 3.4 million was reallocated to “Grants� for the completion a consolidation of subprojects and strengthening of beneficiaries’ capacity. Reallocations were approved as follows: SDR 450,000 to “Grants� for agricultural and nonagricultural subprojects SDR15,000 to “Equipment, vehicles and materials�; SDR 207,000 to “Consultants’ Services, Training and Audits�; 04/02/2008 S S 89.99 and SDR 558 “Operating Costs.� These adjustments reflected the high inflation (7.2%) and primarily the increase in the price of fuel during 2007, raising substantially the costs of visiting project areas. Closing Date of the OP was extended by 4 months to allow completion of ongoing subprojects; (ii) SDR 355,000 was reallocated to 12/17/2008 S S 95.71 “Consultants� and SDR 723,000 to “Operating Costs� to cover for expenses generated by the preparation of the exchange rate fluctuations. 06/04/2010 S S 96.58 Closing Date of OP was x ISR Ratings at Amount Board Restructuring Disbursed at Restructuring Reason for Restructuring & Approved Restructuring Date(s) Key Changes Made PDO Change DO IP in USD millions extended by two and half years to coincide with the AF closing date, following authorization to exceptionally resume project activities in light of OP/BP 7.30. Closing Date of AF was extended by 18 months to compensate for delayed effectiveness and permit successful completion of project activities; (ii) SDR 37,500 and SDR 232,500 was reallocated to “Consultants services, training, audits� an “Operating Costs,� 02/08/2011 MS MS 107.10 respectively, to accommodate a new strategy for managing the project’s vehicle fleet; (iii) the threshold l for prior review in NCB was restored to USD 100,000, reflecting improved procurement capacity; (iv) outcome indicators were revised to improve measurement of PDO. Closing date of OP was extended by 18 months to avoid loss of unspent IDA funds due to the disbursement suspension in effect under OP/BP 7.30. Closing date of the OP was made to coincide with the AF 06/30/2011 MS MS 114.85 closing date; (ii) SDR 392,610 was reallocated to “Grants� to cover additional project management costs to maintain core functions of the Project Implementation Unit (PIU) dur the suspension of disbursement. US$ 6.5. million was cancelled so as to focus efforts on 11/27/2012 MS MS 120.43 sustainability of ongoing subprojects and project targets were adjusted accordingly. xi I. Disbursement Profile xii 1. Project Context, Development Objectives and Design 1.1 Context at Appraisal 1.1.1 This Implementation Completion Report (ICR) describes the experiences, achievements and lessons of the Madagascar Rural Development Support Project. The project was implemented in two stages: the Original Project (OP - Credit 3524-MAG) approved June 19, 2001 and closed December 31, 2012, and an Additional Financing (AF - Credit 4525-MAG) with the same project development objectives, design and implementation arrangements, approved October 23, 2008 and closed at the same time as the OP on December 31, 2012. The AF scaled up the coverage of project activities/investments in small-scale socio-economic infrastructure and services, with the goal of increasing the impact of the OP. 1.1.2 At the time of appraisal, Madagascar was one of the poorest countries in the world, with an overall poverty rate of 70 percent, rising to 80 percent in rural areas. Sixty percent of rural poor were classified as extremely poor, being unable to meet their minimum caloric consumption needs. Agriculture provided a livelihood for 75 percent of the country’s population; however, performance of the sector had been disappointing despite economic liberalization and the adoption of a number of macroeconomic reforms in the nineties. Major constrains to the development of the sector remained, including lack of rural infrastructure, poorly functioning rural institutions, limited access to credit, burdensome regulatory limitations, and high transport costs. Further, agriculture practices were poorly developed, suffered from a lack of support services, and mostly were carried out in an environmentally unfriendly manner. 1.1.3 Bank-supported and other rural poverty reduction and environment projects carried out in Madagascar in the previous decade had helped to alleviate these conditions. However, these interventions were characterized by a sub-sectorial focus, lack of synergies and integration, and a supply driven approach. As such, they had a limited and localized impact, as evidenced by stagnation in yield growth for virtually all crops and production systems during that time. Massive demand for basic services and more innovative approaches in rural areas indicated the ongoing poverty reduction challenges facing the country. 1.1.4 Government’s Strategy and Rationale for Bank Involvement. The need for sustained growth in agriculture as a means to alleviate rural poverty had been emphasized in the Interim Poverty Reduction Strategy paper finalized by the Government of Madagascar (GOM) at the end of 2000. In addition, at the time of project preparation the GOM had adopted a rural development strategy as well as the Rural Development Action Plan (PADR), which identified the following core policy themes along which programs and projects were to be formulated and implemented: • Sustainable growth in agricultural production through increased use of improved technologies; clear linkages with the environment and adoption of environmentally friendly practices; strengthened capacity for maintenance of rural infrastructure; and acceleration of the land titling process; • Partnerships in rural development, accompanied by the rationalization of the role of the state in rural development and the strengthening of producer organizations’ participation and rural finance institutions; • Institutional and regulatory reforms, including the promotion of decentralization and the reform of the regulatory framework; • Regional food security, through the improvement of rural transport and emergency preparedness; and 1 • Rural social services, including improved access to clean water and sanitation. 1.1.5 The adoption of the PADR represented a major step in the direction proposed by the Bank’s Country Assistance Strategy (CAS – IDA/R97-7), which encouraged all stakeholders from the public and private sectors to come together to define a rural development strategy relying on empowered local communities. The PADR was consistent with the CAS in (a) ensuring a pro-poor orientation of growth; (b) building human capital and institutional capacity; (c) strengthening the public sector’s ability to deliver quality services and create a business- friendly environment; and (d) promoting natural resources management to reduce environmental degradation. As such, Bank support for the sector was justified as a catalyst to environmentally sustainable economic development, business creation, more rational public investment in the sector and export growth, and more generally as a bridge to the continuation of much needed support in the agriculture sector. The Original Project (OP) became an integral part of the PADR by supporting the first three pillar of the Government’s rural strategy, namely sustainable growth in agricultural production, promotion of partnerships in rural development, and institutional and regulatory reforms. Similarly, the Additional Financing (AF) adhered closely to the new CAS (38135-MG, 2007-2011), which was aligned with the Government’s new development strategy called Madagascar Action Plan (MAP), by responding to the challenge of removing constraints to investments and growth in rural areas through economic inclusion to strengthen communities’ productive potential and integrate them into the broader economy. 1.1.6 A Bank sector review mission carried out in 2000 identified the challenges and opportunities for alleviation of rural poverty in Madagascar and developed strategic options for policy and programmatic action. Finding that, despite the existence of a relatively balanced incentive structure, agricultural productivity had remained almost stagnant during the previous decade, the Bank recommended the adoption of more innovative approaches to rural development focused on (a) demand-driven projects implemented through decentralized mechanisms; (b) increased farmers participation in decision making and financial contribution; and (c) improved infrastructure and technological services. According to the Project Appraisal Document (PAD), project design sought close alignment with (a), (b), and (c). 1.2 Original Project Development Objectives (PDO) and Key Indicators (as approved) 1.2.1 As stated in the PAD, the Project Development Objective (PDO) was to increase incomes and reduce poverty in rural areas, while preserving the natural resource base. 1 The project was part of the Rural Development Action Plan (PADR), a broad-based program approved by the Government in 1999 to promote sustainable growth in agricultural production, foster food security and enhance access to basic services in the rural areas. The project was designed to support demand-driven activities in agricultural production and technology transfer, and strengthen capacity at national, regional and community levels. 1.2.2 The key performance indicator was: 1 The Development Credit Agreement for the OP as well as the Financing Agreement for the AF include a similar albeit more specific PDO formulation: “to contribute to the implementation of the PADR and selected elements of the PADR by fostering development of revenue increasing activities, thus increasing incomes and reducing poverty in rural areas, while preserving the natural resource base.� This formulation makes reference to the Government Rural Development Action Plan which over time became irrelevant and was substituted by the Madagascar Action Plan (MAP). The MAP maintained support for the main elements stated in the PDO as formulated in the PAD, which the ICR uses. 2 • Productivity of major crops, livestock, fish and type of processed agricultural products increases by 35% one year after the completion of subprojects. 1.3 Revised PDO (as approved by original approving authority) and Key Indicators, and reasons/justification 1.3.1 The PDO was never formally revised, but several adjustments were made to make it more measurable. Key Performance Indicators (KPI) evolved over the life of the project, maintaining the spirit and overall level of achievement of original ones. Intermediate outcome indicators were introduced at the time of the AF reflecting the adoption of a formal “Logframe� structure, which was not a design standard of Bank projects at the time the project was designed. Specifically, on two occasions changes were made to the KPI. In 2008, taking advantage of the Additional Financing review process, indicators were revised to better align them to the expected outcomes to be achieved during the project, with productivity-related indicators moved from the PDO-level to the intermediate outcome-level. In 2011, several of the key indicators were again revised as part of project restructuring to facilitate measurement of progress achieved toward the PDO, focus more sharply on outcomes to be achieved through the inclusion of several new performance indicators, add several mandatory core indicators, and remove several indicators that the Project interventions would influence only indirectly. The changes to the PDO-level indicators and key intermediate outcome-level indicators are summarized below. Original Project Additional Financing 2011 Project Restructuring PDO-level indicators • Productivity of major crops, • Percentage of • Average revenue increase livestock, fish and type of beneficiaries households among beneficiaries (%) processed agricultural registering 30% increase products increases by 35% in revenues at the end of one year after the the project completion of subprojects • Infrastructure functioning one year after completion and mechanism in place to finance operations and maintenance (%) • Subprojects functioning one year after completion (%) • Subprojects having satisfactorily developed and implemented an Environmental Action Plan (%) • Direct project beneficiaries (number), of which female (%) - members of organizations supported by the project - members of organizations supported by the project plus their households Intermediate outcome-level indicators • Additional area under Mandatory CORE Indicators: irrigation systems (ha) • Additional area provided with irrigation and drainage services (ha); • Water users provided with irrigation and drainage services 3 (number); • Water users associations that are operational (number) • About 1000 sub-projects • Number of sub-projects • Sub-projects financed and financed annually during financed and executed executed (number); project life, of which at least • Average financial rate of return 40% benefit women of sub-projects (%) • About 400 savings schemes • Saving schemes • Saving schemes established and have been established established (number) functioning (number) • About 5,000 groups of • Number of producer • Area cultivated with improved producers have been organizations that have technologies and/ or inputs strengthened received support services provided through the sub- projects (ha); • Infrastructure built or rehabilitated (number) (Dams, Processing units, Storage facilities, Slaughterhouses) • Percentage of beneficiary • Production of key crops households with 50% financed through the sub increase in production projects (increase in %, suppl. volumes following value: crops in tons) (Rice, subproject assistance. Maize, Peanuts) • About 30% of resources • FCRA institutional audit • Diagnostic study on the national allocated for competitive and restructuring plan agricultural research system research and extension prepared and approved by completed (yes/no); grants awarded to private the Government • Research studies developed sector and NGOs (number); • National Germplasm Bank rehabilitated and material at risk regenerated (yes/no) 1.4 Main Beneficiaries 1.4.1 The primary target population was the same for both phases. The OP intended to reach with at least one subproject an estimated 180,000 families, identified as poor smallholders living in remote, low-density areas with no access to any subsidy for productive investments and with inadequate infrastructure and services. The AF targeted an incremental 70,000 families with the same profile, organized into 6,000 producer organizations. The planned project area covered all of the 20 agro-ecological regions of the country, with 117 districts (out of the total 119) to be covered under the AF. Special attention was to be given to targeting mechanisms to ensure effective representation and participation of women, including giving priority to subprojects proposed by women’s groups. It was anticipated that the entire rural population in the area served by each subproject would also benefit indirectly. In addition, national and provincial 2 institutions in the rural sector and associated with the project would benefit from improved capacity to formulate and analyze policy, and to promote public/private coordination, as a result of project- financed institutional development activities. 2 As part of an effort to decentralize administration, Madagascar's six administrative provinces were subdivided into 22 regions in 2004. The regions became the highest subdivision level when the provinces were dissolved in accordance with the results of the 2007 referendum. The regions are further subdivided into 119 districts, 1,579 communes, and 17,485 fokontany. 4 1.5 Original Components (as approved) 1.5.1 The OP planned to finance five components: Component 1: Productive Investment (US$71.39 million, 67% of total project cost) financed about 1,000 matching grants per year. These grants went to organized rural producer organizations and village groups and were intended to finance demand-driven small-scale productive infrastructure, agricultural production, and off-farm investments in rural areas. Component 2: Support Services (US$11.88 million, 11% of total project cost) financed extension services, technical support, and training to build capacity of farmers and producers organizations for the implementation of the demand-driven productive investments and other small scale non-productive activities. Component 2 also financed agricultural research through a competitive grant program, and it sponsored research for strategic and upstream research. Component 3: Community Development (US$6.13, 6% of total project cost) financed demand- driven activities aimed at strengthening the rural communities and producer organizations. Activities supported by Component 3 included: (a) preparation of development and business plans for financing under Component 1, (b) development of organizational and managerial capabilities, and (c) events for promoting beneficiaries’ participation. Component 4: Capacity Building and Policy Development (US$5.22, 5% of total project cost) financed technical assistance and training to build capacity in implementing agencies to reinforce the PADR process at national and regional levels, to formulate and analyze rural policy, to establish adequate agricultural statistical systems, and to strengthen environmental assessment. Component 5: Project Administration and Monitoring (US$11.47, 11% of total project cost) financed project administration and coordination activities, including supervision, monitoring and impact evaluation (M&E). 1.6 Revised Components 1.6.1 The original five components were revised and consolidated into four components as a result of the 2002 project and overall portfolio restructuring approved by the Board to improve consistency of project design, lower transaction costs, and increase development effectiveness. Specifically: (i) capacity building and training activities for farmers organizations (Component 2) were integrated with the financing of productive investments (Component 1); (ii) outreach efforts and sponsored research activities were consolidated into a single component (Component 2); (iii) Component 3 – Community Development was folded into Component 1 – Productive Investments; and (iv) strengthening of the environmental units of Ministries involved in rural development (under Component 4) was discontinued, since similar activities were being supported under the Madagascar Environmental Program. These changes to the components are summarized below: Original Components Revised Components Component 1: Productive Investment Component 1 – Productive Investments and Support Services Component 2: Support Services Component 2 – Agricultural research and Training Component 3: Community Development Component 3 – Capacity Building and Policy Development 5 Component 4: Capacity Building and Policy Component 4 – Project Administration and Development Monitoring Component 5: Project Administration and Monitoring These four components were maintained in the AF, with virtually identical allocations by component as under the OP. These allocations were subsequently amended to help advance activities that were performing well (productive subprojects) at the expense of those that were performing less well (research). 1.7 Other significant changes 1.7.1 Suspension of disbursement: Following the change in Government on March 17, 2009, Bank Management triggered OP/BP 7.30 (Dealing with de facto Governments), leading to a temporary ban on written communication with Government agencies and a freeze on disbursement. The ban was still in effect on the date when the OP was scheduled to close (April 30, 2009), thereby preventing the processing of the closing. On May 6, 2010, exceptional authorization was granted retroactively to extend the closing date of the OP to make it coincide with the closing date for the AF and to allow for the completion of pending subproject activities. On the same date, the AF was declared effective. Due to an administrative oversight, however, authorization to continue disbursements under the OP was not granted until June 2011. As a result, even though the OP was able to resume implementation, about US$1.6 million IDA funds remained idle for more than two years. OP/BP 7.30 remained in effect at the time of closing. 1.7.2 Additional Financing: As noted above, the AF loan (US$30 million), approved in October 2008, was designed to scale up financing and beneficiary coverage to strengthen the impact of the well-performing OP, which had become the centerpiece of Government’s agenda for promoting direct investments in the rural sector. While preserving the basic design of the OP and maintaining the same implementation structures, the AF was also intended to provide bridge financing until the launch of the Government’s Agriculture Sector Program (ASP) and the putting in place at the national level of new agricultural support services. The Bank’s internal review process resulted in changes to the KPI, as noted, while implementation arrangements were adjusted to reflect the Government’s decentralization process from provinces to regions. 3 The project’s Financial Management (FM) arrangements were also modified, moving from transaction-based disbursement procedures under the OP to report-based disbursement under the AF, which was to finance 100 percent of each disbursement category. 1.7.3 Restructuring: The project was restructured a total of 10 times, reflecting the project’s life span and the political turbulence and natural shocks it experienced. The following changes were made: • November 15, 2002 (Level 1 approved by the Board): (i) Components were consolidated as described above; (ii) implementation arrangements were improved through the inclusion of NGOs as strategic partners for the identification, design and implementation of agricultural production and off-farm subprojects; (iii) project management was enhanced through the appointment of a project management team modeled on private sector practices; (iv) due to liquidity problems, the authorized allocations of Special Accounts A and B were revised; (v) 3 The 4 Provincial Project Implementation Units (PPIUs) were decentralized into 8 Regional Project Implementation Units (RPIUs) responsible for coordination and implementation of project activities under Component 1. 6 due to increasing counterpart financing problems, the disbursement percentage of expenditure Category 1 (Grants) was increased to 100 percent, while the disbursement percentage of Categories 2, 3, 4 and 5 (Civil Works; Equipment, vehicles and materials; Consultants’ services, training and audits; and Operating Costs respectively) was increased to 100 percent for about one year (until December 2003); (vi) support to the environmental agencies was eliminated; and (vii) intermediate outcome indicators were brought in line with the consolidated components and project targets were adjusted by reducing the targeted number of subprojects from 1,000 to 800, the number of families benefited from 180,000 to 150,000, the percentage of subprojects geared to benefit women from 40 percent to 25 percent, the number of producer groups from 5,000 to 3,000, the percentage of producers that adopt improved agricultural technologies from 75 percent to 50 percent, and the number of rapid rural appraisals from 600 to 400. These changes were made because it was recognized that the original targets were overly ambitious. • August 6, 2004 (Level 2 approved by the Country Director): Reallocations of credit proceeds were approved as follows: SDR 13.6 million to “Grants� for agricultural and nonagricultural subprojects, reflecting satisfactory performance and strong demand for subprojects; SDR 1.3 million to “Equipment, vehicles and materials;� SDR 3.8 million to “Operating costs� reflecting increased costs of targeting poor families in isolated communities and the need to carry out a full-fledged agricultural census (with a decrease in “Consultants, training, audit� category following inclusion of training as part of “Grants� for productive subprojects). • June 27, 2006 (Level 2 approved by the Regional Vice President): (i) SDR 11.58 million was reallocated to “Grants� for productive subprojects to permit the project to finance the continuing strong demand for subprojects following the rice intensification campaign; and a reallocation of SDR 12,500 to “Operating Costs� (with a decrease in “Grants� for productive infrastructure, “Works�; and “Consultants, training, audit� categories) (ii) project activities were amended to clearly reflect the nature of the 2005 cyclone repairs and rice intensification campaigns; and (iii) component outcome indicators were slightly changed to make them more relevant and in line with the PDO. • June 6, 2007 (Level 2 approved by the Country Director): (i) the Closing Date of the OP was extended by 18 months to permit the finalization of all remaining subprojects and training and disbursement of the remaining Credit balance; (ii) SDR 3.4 million was reallocated to “Grants� for the completion and consolidation of subprojects and strengthening of beneficiaries’ capacity (with a decrease in the “Consultants, training, audit and “Operating Costs� categories). • April 2, 2008 (Level 2 approved by the Country Director): Reallocations were approved as follows: SDR 450,000 to “Grants� for agricultural and nonagricultural subprojects; SDR15,000 to “Equipment, vehicles and materials�; SDR 207,000 to “Consultants’ Services, Training and Audits�; and SDR 558,000 to “Operating Costs.� These adjustments reflected the high inflation (7.2%) and primarily the increase in the price of fuel during 2007, raising substantially the costs of visiting project areas. • December 17, 2008 (Level 2 approved by the Country Director): the Closing Date of the OP was extended by 4 months 4 to allow completion of ongoing subprojects; (ii) SDR 355,000 was reallocated to “Consultants� and SDR 723,000 to “Operating Costs� to cover for expenses generated by the preparation of the AF and exchange rate fluctuations (with a decrease in “Grants� for subprojects category). 4 This extension applied only to category 1 (A) and 1 (B) of the IDA Credit Agreement (IDA-3524-MAG). All other disbursement categories were supposed to close as scheduled on December 31, 2008. 7 • June 4, 2010 (Level 2 approved by the Managing Director as part of the authorization to exceptionally resume project activities in light of OP/BP 7.30): the Closing Date of the OP was extended by two and half years to coincide with the AF closing date. • February 8, 2011 (Level 2 approved by the Country Director): (i) the Closing Date of the AF was extended by 18 months to compensate for delayed effectiveness and permit successful completion of project activities; (ii) SDR 37,500 and SDR 232,500 was reallocated to “Consultants, training, audits� and “Operating Costs,� respectively, to accommodate a new strategy for managing the project’s vehicle fleet (with a decrease in the “Materials, vehicles, equipment� category); (iii) the threshold level for prior review in NCB was restored to USD 100,000, reflecting improved procurement capacity; (iv) outcome indicators were revised to improve measurement of PDO. • June 30, 2011 (Level 2 approved by the Regional Vice President): (i) the Closing date of the OP was extended by 18 months, resulting in a cumulative extension of five and half years, to avoid loss of unspent IDA funds due to the disbursement suspension in effect under OP/BP 7.30. The closing date of the OP was made to coincide with the AF closing date; (ii) SDR 392,610 was reallocated to “Grants�, and SDR 71,440 to “Equipment, vehicles and materials� (with a decrease in all other categories) to cover additional project management costs to maintain core functions of the Project Implementation Unit (PIU) during the suspension of disbursement. • November 27, 2012 (Level 2 approved by the Country Director): (i) US$ 6.5. million was cancelled so as to focus efforts on sustainability of ongoing subprojects; and (ii) project targets were adjusted by reducing the targeted number of beneficiary from 1,060,000 to 1,043,650. 2. Key Factors Affecting Implementation and Outcomes 2.1 Project Preparation, Design and Quality at Entry 2.1.1 Soundness of the background analysis: The analytical basis and justification for the OP rested on the documented lessons from several previous operations in the rural sector across the country and elsewhere since the early nineties. The OP project was the direct outcome of its precursor operations, the main lessons of which had been incorporated in the government rural strategy and its action plan (PADR), and included: (a) the need to adopt an integrated sectoral focus to address rural growth and development nationwide; (b) the effectiveness of decentralization in reducing bureaucracy and reinforcing accountability; (c) the importance of using a demand-driven approach and encouraging beneficiary participation at all stages of the investment cycle for building ownership and ensuring sustainability; (d) the importance of building partnerships with all sector stakeholders and in particular the private sector for efficient provision of technical services; and (e) the need to link research with on the ground extension services and community investments. Lessons were also taken from a similar program financed by the Bank in Mexico 5 showing that successful rural investment programs need to promote a regional approach. The considerable amount of background analysis and consultation carried out during the rural strategy formulation provided a solid justification for Bank support for the PADR. The project was the first operation to support the PADR and was viewed as a vehicle for maintaining momentum in the participatory implementation of the country’s rural development policy. As such, it was seen as an instrument for intensifying economic activity within the viable 5 The Mexican Rural Development in Marginal Areas Project. 8 small farm sector, a growth enhancer for the non-farm sector, and a safety net for poor rural families in isolated areas with natural resource limitations. 2.1.2 Assessment of project design: Overall, the project design was relatively simple and well understood by the counterpart team. The level of complexity was no greater than that of similar community-driven development (CCD) operations supported by the Bank in the Region and elsewhere. It had a few significant shortcomings, including the (i) inadequate composition of teams staffing the national and provincial level implementation units, which did not reflect the orientation and scope of activities, causing bottlenecks for implementation; 6 (ii) fragmentation of activities among several components and subcomponents: separation of farmer organizations’ capacity building and training from the related investments, increasing the project’s transaction costs, and separation of research activities and associated research outreach and technology transfer, isolating research activities from on-farm reality; (iii) processing of subprojects on an individual basis, forgoing economies of scale; and (iv) wide range of activities, impeding a focused effort. In addition, the initial absence of an integrated monitoring and evaluation system hampered proper coordination, team work and information management. Most of these shortcomings were proactively corrected early on, primarily as part the 2002 overall portfolio restructuring described above, resulting in improved development effectiveness and accelerated implementation. The AF design overestimated the capacity of farmer organizations to mobilize cash contributions for matching grants, which proved a significant obstacle to smooth implementation and completion of activities, as less than half of producer organizations were able to fully contribute as stipulated at subproject design stage. It needs to be recognized, however, that the capacity of farmer organizations to mobilize cash contributions was adversely affected by the prolonged political crisis that broke out just before the AF became effective. 2.1.3 Project Development Objectives (PDO) and Indicators: Project objectives under the OP and AF were quite ambitious and not always directly attributable to project supported intervention and in some cases likely to be influenced by factors far beyond the control of the project. It was indeed unrealistic to expect that the project by itself would increase income and reduce poverty in rural areas 7 . However, objectives were reasonable given conditions on the ground in rural Madagascar, were aligned with country and sector strategies, and remained broadly consistent with the Borrower’s rural priorities until final project closing. The higher-level development objective – not expected to be measured within the life of the project – sought to reduce poverty by alleviating wide-ranging constraints, including agriculture. The single Key Indicator, however, did not fully capture the scope or meaning of the PDO, and was not accompanied by Intermediate Results Indicators, reflecting the absence of a formal “Logframe� structure at the time the project was prepared. More precise measurement was needed to gauge the actual increase in income and reduction in poverty, which are not easily measured and imply a 6 Despite the focus of PSDR to increase revenues of producer organizations, the team initially did not have commercial and financial expertise. Despite the large budget allocation for productive infrastructure projects, the team initially did not have civil engineering expertise. Despite the clear need for proper compliance with agreed environmental safeguards in view of Madagascar’s unique environment, the team initially did not have dedicated environmental expertise. The concentration of strategic, administrative and operational coordination functions in the hands of a single project director was incompatible with the vast magnitude of project activities. 7 The project was designed in 1999-2000 when all projects followed the Logical Framework with inputs (costs) and output and impact indicators. There was no requirement that PDO impact indicators be attributable to the project. As such, the PDO and the absence of intermediate indicators were consistent with the design standards of Bank projects at the time. 9 well-formulated and executed M&E plan. Some targets established for the OP were overly ambitious, representing an unrealistic benchmark to adequately measure performance, and they were not always in line with capacity on the ground. Revised and additional performance indicators were subsequently included, and targets were reduced to levels that the project could realistically achieve. However, the opportunity to focus the PDO and make it more measurable was never taken up 8. 2.1.4 Integration/Decentralization: Integration of the project into programs and policies being implemented at the local level proved to be more complex than anticipated. During the design stage, it was imagined that the Regional Rural Development Working Groups (GTDRs) would serve as the primary clearing houses for identifying regional priority investments, ensuring regional participation and inter-sectoral coordination, and eventually approving subproject proposals. In practice, the GTDRs did not have enough capacity to carry out these functions, and their role in the subproject cycle was too loosely defined. Because the legal status of the GTDRs was not formally recognized until 2004, for quite some time the project operated in an institutional vacuum at the regional level. Later, questions emerged about the sustainability of the GTDRs, as the Government’s commitment to continue financing them was weak. During the AF phase, the role originally assigned to the GTDRs was played by the newly created Regional Executive Directorates (DER) of the Ministry of Agriculture. In retrospect, it is clear that the GTDRs’ capacity to perform the integration function at local level was significantly overestimated, resulting in overly ambitious targets. It should be recognized, however, that at the time of project design Madagascar’s institutions were in a constant flux, making it difficult to support emerging decentralization efforts. Under these circumstances, it was nearly impossible for an original design to predict accurately which institution would remain in place. The only sensible approach was to keep the project design flexible, and build on a strong Project Implementation Unit (PIU) as the main tool to deliver an ambitious program at the local level in a difficult environment. 2.1.5 Communally versus individually managed subprojects: The design requirement that subprojects be implemented by producer groups comprising no less than 10 individuals from different households proved to be a constraint, as such groups did not always exist in areas being targeted by the project. Even when such groups did exist (or in cases where they were formed for the purpose of implementing a subproject), the lack of a commitment to collaborative work often undermined their effectiveness, leading to the abandonment of many subprojects. Mainly for this reason, the eligibility criteria for subprojects were gradually relaxed, and over time the project supported increasing numbers of subprojects that were implemented by more natural-sized producer groups dominated by members of the same household, or producer groups comprising fewer than 10 members. 2.1.6 Adequacy of Government’s commitment: The project was formulated within the policy and institutional framework of the PADR. The Government was fully-committed to the project methodology, objectives and activities under both stages, and it collaborated closely with the Bank to improve key design elements and resolve problems at each stage. This commitment flowed through into the implementation phase, as reflected for example in the swift response to the 2004 rice crisis that temporarily derailed the project (see section 2.2), in the Government’s 8 Even though the PDO was not changed, there were considerable attempts to make it more measurable. The PSDR team tried to measure the project impact both inside and outside beneficiary areas, which was the standard at the time. 10 strong support for the Additional Financing, and in its decision to allocate public funds during the disbursement freeze period (2009-10) to allow project staff to continue working. 2.1.7 Risk assessment: Risks were correctly projected by the PAD based on previous experiences with similar projects, but there were omissions both in the types of risks that were identified and in the planned risk minimization measures. For example, after identifying the risk associated with political interference in the selection of subprojects, the PAD might have called for the establishment early on of a strong monitoring and evaluation system 9. Also, the pervasive lack of adequate technical assistance in rural areas should have been recognized, and a mitigation framework designed and implemented earlier on, especially in light of the project’s productive goals and consequent need for sustained, specialized support. The risk analysis obviously could not have predicted the political crises of 2001 and 2009, which affected the entire Bank portfolio in Madagascar, and magnified the risks already identified at appraisal. The political crises were in the end the major implementation bottlenecks. 2.2 Implementation Box 2.2.1: Developments in the project’s external environment The external environment in which the project operated was characterized by significant developments that affected its performance and required extensive flexibility and continuous adjustments. Madagascar is characterized by a state of increasing fragility, with chronic political instability and vulnerability to economic and natural shocks as the main external risks. Over the twelve years of project implementation, the country experienced five years of political crisis on two occasions, the most recent of which has been ongoing since 2009 with an unconstitutional regime change. After having been one of the most rapidly growing African economies, Madagascar plunged into a deep political crisis at the beginning of 2002, which led to a freezing of Madagascar's assets abroad, a suspension of foreign exchange trading and a closure of the treasury bond market for several months. Economic activity fell sharply, poverty increased, and revenue losses of the public sector were immense. Farm produce prices, especially in isolated areas of the country, were cut in half, causing an unprecedented drop in rural incomes and putting future production at risk. Following the resolution of the 2002 crisis, the country experienced positive developments in social, economic and governance indicators until early 2009 when the second, ongoing political crisis erupted, wiping out years of achievements, setting the development clock back, and leading to a decline in economic growth. The 9 The Bank team, however, feels that no M&E system would have survived the political pressure coming from the top, where the President himself had a personal interest in agriculture policy and frequently made personal policy changes. 11 prolonged and severe political crisis has caused major economic and social impacts that have been exacerbated by a series of external shocks and a withdrawal of most external funding -- the economy has stalled, poverty has sharply increased, the ability to deal with exogenous shocks is severely curtailed, and infrastructure has deteriorated. Madagascar is among the countries with the highest risk level of natural disasters, such as droughts, cyclones, locust plagues, and floods, which regularly undermine growth and poverty reduction. Moreover, in the last three years, the rise of rice prices has created the context for an impending food crisis. Each year, an average of three to four cyclones make landfall on parts of Madagascar causing on average damage of US$50 million per cyclone. During the life of the project, the country suffered from several major cyclones, some of them the most violent cyclones in Madagascar in the last 40 years (Gafilo in 2004). In 2008, three consecutive cyclones affected 84 percent of the national territory, causing economic losses equivalent to 4 percent of GDP. Damages from the 2012 cyclone season have been estimated to be in the same range. In 2010, about 60 percent of extremely poor households suffered a catastrophic event or a combination of catastrophic events (e.g. cyclones, floods, droughts, locust infestations, plant diseases) that adversely affected their economic well-being. As such, these risks are an integral risk to the country's development and the Bank's program. Extended political crisis and natural disasters have increased the vulnerability of large segments of the population. The rural population, whose income is highly dependent on agriculture, has been especially affected by the natural and man-made crises, as a weakening demand for food (associated with the overall decline of the economy) has been compounded by a series of agricultural supply disruptions caused by natural disasters. As a result of these combined shocks, many rural households have depleted their assets and lost their main source of income, leading to sharp declines in income and welfare in rural areas. 2.2.2 The 2002 and particularly the 2009 political crises greatly disrupted project implementation and had a pronounced social and economic impact especially in rural areas. The launch of the project was complicated by the political climate leading up to the presidential elections of December 2001 and the subsequent period of conflict and uncertainty in the country that paralyzed the project for six months (until July 2002). The ensuing lack of clarity regarding decision-making authority between and within Government agencies prolonged the period required to achieve full implementation, particularly for activities in which the Ministry of Agriculture was supposed to play a leadership role (i.e., the competitive grant scheme for applied research, capacity building, and policy development). However, after the new Government assumed power in July 2002, the enabling conditions for effective implementation were restored rather swiftly. As a result, the project got back on track rather quickly, greatly expanded its presence and impact in the field, and performed well until the closing of the OP, with the exception of the difficulties of 2004 and 2005 (see section 2.2.3). 2.2.3 The outbreak in late 2008 of civil disturbances leading in early 2009 to the change in government and the subsequent triggering of OP/BP 7.30 came at an unfortunate time, with the OP about to close and the AF not yet effective. Political paralysis, coupled with the severe fiscal constraints experienced by the Government following the suspension of most development assistance, resulted in a major disruption of the project, as funds were not available to cover project management costs during the 18-month disbursement freeze. The capacity of the Project Implementation Unit (PIU) eroded rapidly because of staff defections, affecting the capacity of the Borrower to maintain effective control over project records, management systems, and physical assets, both at the central and regional level, and posing a fiduciary and reputational risk for the Bank. In May 2010, the Bank’s senior management granted exceptional authorization for the AF to be declared effective, and from that point on activities under Component 1 resumed swiftly with the recruitment of new staff, payment of outstanding invoices, and identification and evaluation of new subprojects. Activities under Component 2 and 3 continued to suffer delays as 12 direct transfer of funds to government agencies was prohibited, even after the project was granted an exemption to the disbursement freeze. 2.2.4 Political interference was a major risk to the project between 2004 and 2005. Following the 2004 rice crisis, the project fell victim of its own visibility, as the regime in power placed intense political pressure on the project to finance a rice intensification campaign and support cyclone repair works outside of the approved work plan. This pressure resulted in significant over-commitment of credit proceeds and loss of financial and monitoring control by the project team. 10 Strong remedial measures were taken at that time, resulting in a declaration of misprocurement and ineligible expenditures, major cancellations of subprojects, hiring of a fiduciary oversight company, launching of financial and procurement audits, strengthening of the operational manual, and commissioning a review by Madagascar’s anticorruption committee. Thanks to these remedial measures, the political interference stopped, and the project regained its institutional reputation as well as its perceived integrity. 2.2.5 Frequent turnover in MAEP and PIU personnel affected the performance of the project. In addition to the disruptions caused by the political crises, the project had to absorb changes in the management team in mid-2003 (following the 2002 restructuring) and in 2005 (as a result of the declaration of misprocurement). These changes eventually served to strengthen implementation capacity, but they were disruptive in the short run because newly hired staff had to climb a steep learning curve. Similarly, resignations of several mid-level staff in 2007-08 precipitated by the looming closure of the project created further disruptions, although these did not affect the top management of the PIU. The fact that both the Minister of Agriculture and the president of the project Steering Committee changed frequently throughout the life of the project made it difficult to establish an effective interface between the project and its host Ministry, a problem which persisted until the end. 2.2.6 The uninterrupted political support to PSDR was a positive factor. Despite the political crises and the strong political pressure to which it was subjected at various times during its implementation, the project remained an important if not the most important element of the country’s rural development agenda during both the OP and AF phase. The successive governments showed a consistent commitment to rural development and to protecting and supporting the project as a major instrument for the implementation of their rural development policy and for improving productive assets of poor rural households. 2.2.7 Institutional capacity to appraise and support productive investments was uneven: Weak institutional capacity of producer organizations and lack of qualified and experienced service providers in sufficient numbers was another reality that the project had to deal with since its launch. The institutional assessment at the time noted that excessive centralization of decision- making was undermining the capacity of communities and local governments to manage their own affairs, which the PAD recognized was key for the success of the project. To mitigate the risks associated with the limited institutional capacity of local entities and groups, the project called for the participation of collaborating institutions and private strategic partners, through the adoption of a wholesale approach to subproject implementation, which replaced the demand- driven approach for capacity building that was followed since the project launch. Despite the 10 The project engaged in two major operations outside the approved Annual Work Plan: a rice intensification campaign and repairs to cyclone-damaged infrastructure, for a total of 4,200 subprojects. Project funds were over- committed by nearly SDR 23 million. 13 uneven quality of the technical assistance, this approach proved to be successful, as reflected in an increased rate of disbursement. That said, the ability of strategic partners to rapidly develop the capacity to effectively manage subproject activities may have been overestimated. Farmers associations continued to suffer from important institutional weaknesses, which impaired their capacity to make the strategic business choices needed to maximize profits and react to shocks. The challenges of giving farmers long-term support links to markets, financial institutions and expert advice persisted throughout the life of the project, although they were partially addressed under the AF by the creation of “technical pools,� which ensured quality at entry and technical support of subprojects after their approval. 2.2.8 Strategic reorientations helped focus the project and increase impact and visibility but created delays. Initially, the project did not have a consistent strategy to prioritize among subprojects. As a CDD project, it attempted to respond to the needs of beneficiary communities, which it did by supporting any and all proposals that were in conformity with the very broad eligibility criteria. 11 Starting in 2005, after evidence had emerged that many subprojects were closing as soon as project financing had run out, eligibility criteria for subprojects were tightened and screening procedures were introduced in an effort to improve the likelihood that subprojects would be sustainable. These changes led to a more strategic and market-oriented approach. The project’s strategic direction was further adjusted in 2011 following the resumption in disbursements and the ensuing portfolio restructuring, which was designed to concentrate IDA resources on high pay-off activities, achieve critical mass, re-orient activities geographically and exploit synergies with other Bank operations. The changes introduced at that time improved the quality of many subprojects but also slowed the pace of implementation and impaired the project’s ability to disburse. In addition, the introduction under the AF of a requirement that some subproject beneficiaries pay part of their contribution in cash (previously they had been allowed to pay their contributions entirely in kind) proved to be an obstacle to subproject implementation. 2.2.9 PSDR became a critical instrument for rapid response to climate shocks. Project investments were negatively impacted by several major cyclones that strongly hit Madagascar especially in 2004 and 2005. Several subprojects were wiped out and had to be rebuilt, and selected beneficiaries were exceptionally allowed to receive a second grant. From its early stages, the project helped mobilize resources for damaged agriculture infrastructure during many of the cyclone seasons, and played a key role in containing the drought in the south. 2.2.10 Mid-term Review contributed to improved performance quality. A timely Mid-term Review (MTR) with good follow-up was influential in correcting and strengthening the implementation trajectory of the OP. The MTR Action Plan recommended: (i) launching a systematic and dedicated capacity building effort for producer organizations; (ii) improving the realism of subproject design; (iii) ensuring greater selectivity of proposals; (iv) establishing closer linkages between subprojects and the research programs; (v) improving access to markets and increase the effectiveness of slash-and-burn control measures by pursuing closer coordination with other programs and projects; and (vi) further accelerating project implementation by strengthening the national and regional project teams. The Bank team worked with PSDR to implement the plan, and most recommendations were implemented with positive effects. 11 According to the PAD, virtually any agricultural production and off-farm activities were in principle eligible for financing, including but not limited to the establishment of seed and plant nurseries, improvement in existing farming systems, diversification of production systems and a wide range of crops, the development of minor stockbreeding, aquaculture, apiculture, home garden production, handicrafts, various types of woods and iron work, etc. 14 2.2.11 A “Quality of Supervision Assessment� review conducted by the QAG in 2004 rated overall project supervision as “satisfactory.� 2.3 Monitoring and Evaluation (M&E) Design, Implementation and Utilization 2.3.1 Design: The M&E system was designed to reflect the comprehensive scope of PSDR and adopt an M&E approach that would allow for full participation of beneficiary rural communities through the establishment of rural management committees. Additional elements of the system were an internal monitoring conducted by M&E staff at both the central and local level, and periodic supervision and evaluation of project performance and impacts. The overall design of the M&E system had some weaknesses, as it was primarily focused on the measurement of inputs and outputs, rather than results and impacts. The Results Framework reflected the old-style project Logframe mentality and lacked intermediate outcome indicators that would have allowed project performance to be tracked during the early years of implementation. 2.3.2 Implementation and utilization: Overall, the performance of the M&E system was adequate but not exceptional. During the first year of project implementation, the system suffered from considerable deficiencies, and as a result it did not permit an adequate assessment of project performance. These shortcomings were subsequently rectified, and the M&E system was strengthened through integration with the overall project Management Information System (MIS). Following these improvements, the project M&E unit was able to collect and report on performance indicators. However, data collection and reporting procedures were often informal and ad hoc, and performance indicators remained heavily focused on inputs and outputs. Processing and timely dissemination of M&E data was not supported by a well-defined and transparent system, thus limiting the usefulness of the data and making it difficult to communicate the achievements of the project to larger audiences. In order to enhance M&E capacity to measure project performance, and considering the difficulties of assessing the cumulative impact of a myriad of small-scale subprojects, the project financed periodic one-off impact assessment studies to measure progress achieved against performance indicators. 12 These helped adjust the design and implementation of subsequent support to productive investments, including the AF. Additional efforts were made to strengthen the system under the AF through the development of an excel-based tool that allowed the use of M&E data for decision-making. 2.4 Safeguards and Fiduciary Compliance 2.4.1 Safeguards/Environment: The project was a Category B with an Environmental Management Plan and a Pest Management Plan. Implementation arrangements for mitigation measures were specified in MOUs negotiated between PSDR regional offices and subproject beneficiaries. Implementation of mitigation measures was done by the beneficiaries themselves, with the support of local service providers. PSDR technical staff were responsible for monitoring implementation and for ensuring that no adverse environmental impacts were occurring. Activities carried out to ensure compliance with applicable environmental safeguards policies were in general satisfactory, and environmental themes were incorporated in training courses. Bank supervision mission identified minor implementation issues of some mitigation measures, 12 The impact assessment studies introduced IRR calculations as a common denominator for all subprojects, thereby allowing measurement and comparison of impact of subprojects across the portfolio. 15 which were promptly corrected. On the whole, mitigation measures associated with individual subprojects were implemented reasonably well by subproject beneficiaries. 2.4.2 Although the implementation of environmental safeguards policies was generally satisfactory throughout the life of the project, implementation performance declined following the disbursement freeze instituted in March 2009, as the support being provided by local service providers to assist beneficiaries in the implementation of mitigation measures stopped. Similarly, monitoring by PSDR technical staff of the rate of adoption of mitigation measures and assessment of the environmental impacts of project-financed activities ceased. Environmental monitoring and support activities were resumed and strengthened under the AF through the establishment of technical pools. 2.4.3 During project preparation, two safeguards policies were triggered: OP/BP 4.01 (Environmental Assessment) and OP 4.09 (Pest Management). During implementation, questions about the applicability of OP 4.37 (Safety of Dams) were raised with regard to a number of proposals relating to the rehabilitation or development of small-scale irrigation schemes. Considering that safety risks associated with irrigation systems being financed under the Project were minimal and that the affected subprojects were at an advanced stage of design or implementation and could not be stopped without risking timely achievement of the PDO, the policy was not triggered. However, as a precautionary measure and to make sure that the spirit of the policy was respected, PSDR was asked to prepare guidelines for the management of small dams and applied these guidelines systematically to all irrigation schemes financed by the project. 2.4.4 Procurement: The project established early on a satisfactory system to carry out procurement processes. However, as the procurement load became heavy due to the number, diversity, type and technical complexity of the subprojects, the project experienced delays requiring continuous need for specialist consultants to ensure compliance with procurement procedures. 2.4.5 In view of the problem faced by the project in 2005 as a result of the financing of cyclones repairs and the rice intensification campaign outside of the approved work plan (see section 2.2.3), the Bank launched an independent procurement audit which revealed significant irregularities with mini-tractors procured under direct contracts by the PIU, as well as suspicion of fraudulent pricing. The audit also found potential problems with other three contracts, namely cyclone repair works, procurement of fertilizers, and agri-business contract with a local consulting firm. After a thorough internal review, the Bank officially declared misprocurement in June 2006 on the purchase of 750 mini-tractors for a total commitment of US$1.6 million, requested reimbursement of US$ 435,024 and canceled US$ 718,327 in credit proceeds. 2.4.6 Following the extensive corrective measures taken in 2005, the procurement function improved significantly and remained overall satisfactory throughout the OP, with a well- established and functioning procurement unit. During the 2009 political crisis, procurement staff were placed on unpaid leave and the procurement function lapsed. The AF made significant efforts to rebuild procurement capacity, including the hiring of a Fiduciary Firm responsible for supervising all procurement processes under the project. However, minor weaknesses in the procurement system mostly due low capacity remained which delayed implementation of the AF. 2.4.7 Financial Management: The financial management aspects of the project were handled by project staff directly contracted by the PIU, assisted by fiduciary agents (accounting firms) that were contracted over specific periods of the project to provide support and capacity building. 16 Overall, the project had appropriately qualified and experienced personnel in charge of financial management. The internal controls were spelt out in a Procedures Manual, amendments to which were made on a periodic basis to take into account changes in the project’s operations. The project had adequate financial management arrangements over the course of the implementation period. In most cases, the quarterly financial reports and annual audit reports were submitted to the Bank on a timely basis, and recommendations made subsequent to implementation support missions were appropriately implemented. The audit report for the year ending on 31 December 2010 (IDA Credit 4525 MAG) was submitted four months past the due date, owing to late commencement of the procurement process for external audit services. 2.4.8 At the time of preparation of the ICR, the project had not yet submitted the documentation needed to determine the outstanding balances in the Designated Accounts. The amount that needs to be accounted for USD 7.1 million and the project team is working to ensure the submission of the related statements of expenditure before the end of the grace period (30 June 2013). The financial management of the project was rated as moderately satisfactory at the project closing date. 2.5 Post-completion Operation/Next Phase 2.5.1 Transition arrangements to regular operations: An agreement signed between the Ministry of Agriculture and PSDR in December 2012 establishes the transitional arrangements for efficient support, supervision and monitoring after project closure. Such arrangements include technical support on the part of the Ministry for the execution/completion of subprojects, the formal reception by the Ministry of infrastructure subprojects, and the provision of technical and socio-organizational assistance to water user organizations for management and maintenance of infrastructures after the project closure. The regions need to continue investing in social and institutional capital during the transition period and after. There are potential transitional risks from technical and organizational weaknesses in monitoring and vigilance, but not so much from the interruption of programs due to changing political and/or economic priority. 2.5.2 While PSDR launched an important process, critical shortcomings persist in rural Madagascar, affecting the livelihoods and competitive conditions of small producers. These have become even more critical as a result of the prolonged and severe political crisis, which has caused major economic and social impacts that have been exacerbated by a series of external shocks, and a withdrawal of most external funding. These concomitant events have caused a dramatic increase in the incidence of poverty, especially among the most vulnerable population, with a reduction of social protection safety nets, income-earning opportunities, and maintenance of lifeline infrastructure. The short-term Bank response has been the launch of the Emergency Infrastructure Preservation and Vulnerability Reduction project, which among other things is supporting: (i) distribution of improved seed, fertilizer, tools and associated technical assistance to enable rapid increases in production of food staples, mainly rice but also other crops; and (ii) rehabilitation and maintenance of small-scale agricultural infrastructure, primarily micro- irrigation systems. The initial set of subprojects to be financed consist of subprojects pre-selected for financing under PSDR, for which technical design studies have already been prepared. 2.5.3 The longer-term Bank response to the remaining rural development challenges of Madagascar has been to shift strategic approach and focus on linking smallholders into commercial value chains and finding ways to support small and medium enterprises to identify and implement profitable commercial agriculture opportunities with some punctual public support. For this purpose, a new operation is being prepared. 17 2.5.4 On the Government side, efforts are ongoing to build on the positive experiences of PSDR in setting up a decentralized funding mechanism for demand driven interventions and the provision of still much needed technical support and assistance. Specifically, the Government has designed and is putting in place rural development funds which will function as basket funds for rural development subprojects at the regional level. Additionally, Agriculture Support Centers (CSA) are being established and strengthened to provide technical assistance, taking advantage of capacity building efforts developed under PSDR. Further, CSA are using procedures and processes developed and tested under PSDR for the identification, selection and financing of subproject proposals (list of activities eligible for support, level of subsidy, repayment periods, etc.). 3. Assessment of Outcomes 3.1 Relevance of Objectives, Design and Implementation Rating: High overall relevance 3.1.1 The development objective of PSDR was highly relevant when the project was prepared, and it remains highly relevant today, more than a decade later. Promoting the development of revenue-increasing activities in rural areas is fundamental for the future growth and poverty reduction in the country. The objective also remain broadly consistent with the 2007-11 Country Partnership Strategy and the 2012 Interim Strategy Note, explicitly via the challenge of reducing vulnerability and improving resilience in addressing the most urgent problems posed by the crisis, including a dramatic rise in poverty levels. The follow-on Emergency Infrastructure Preservation and Vulnerability Reduction operation focuses on preserving key lifeline infrastructure and reducing household vulnerability, for which preservation of productive capacity in agriculture is key. The types of investments, although smaller in scope, remain similar to PSDR with a focus on reducing household vulnerability in crisis-affected areas through investments designed to boost agricultural production, safeguarding food security, and providing short-term employment opportunities. PSDR was one of the GOM’s core investment instruments for putting the government’s investment strategy into operation and the new project continues to build on its strengths, core objectives and vision. 3.2 Achievement of Project Development Objectives 3.2.1 Achievement of the PDO was substantial and is rated Moderately Satisfactory overall, based on the result of independent studies which show the many important achievements of the project under both phases (OP and AF). Overall, as shown by the evaluation studies and the impressive degree of penetration in the rural areas, PSDR had a positive impact on rural development in Madagascar. The project benefitted more than 10 percent of rural population and was active in more than 70 percent of rural communes, half of which located in remote areas, more than doubling the appraisal target of beneficiary groups. The project had a positive impact on income and quality of life, production, and farmers’ organizational skills and savings. 3.2.2 As a national project that operated for more than a decade throughout the entire country and financed thousands of subprojects, many of which were located in extremely remote locations, PSDR had a huge footprint. Because PSDR was the centerpiece of the Government's rural development strategy for so many years, it received a great deal of publicity, to the point that it would be fair to say that there are very few people in Madagascar who have not heard about PSDR. The project was one of the few programs available to the government that could be used to channel resources to rural communities throughout the country. Another consequence of 18 having had such an extensive reach is that PSDR ended up being the de facto national agricultural extension service for quite some time and an important interface between farmers, agribusiness, and sector institutions. 3.2.3 The project’s impact, however, did not extend beyond the sum of its achievements. As a national program with a strong demand-driven approach that directly responded to the communities’ needs, PSDR became eventually involved in a wide range of activities. Further, as a result of its national presence, the project was pulled in different directions to address the challenges arising from the external shocks, rather than becoming an instrument for long-term development assistance. Despite the strategic reorientations carried out during implementation, aimed at strengthening its strategic focus and embracing a market-oriented approach, the project and the thousands of subprojects it financed did not reach a critical mass nor had a strong catalytic effect on the rural space, with added sustainability concerns. 3.2.4 Indicator 1: Average revenue increase among beneficiaries (30%). Achieved: Subproject impact evaluations 13 carried out during 2012 show that, as a result of project interventions: • Average beneficiaries’ revenues have increased by 38 percent, compared to stagnation in revenues for non-beneficiaries for the same time; • More than 80% of beneficiaries report an increase in income of more than 30%, with beneficiaries engaging in agriculture activities registering a higher increase, followed by those engaging in productive infrastructure (see Table 3.2.3.1); • Average revenue per beneficiary per subsector has increased by about 84%, a positive trend that was maintained until the end of the project; • Increase in income was primarily supported by increase in production, particularly for households with access to irrigated land and improved technologies for rice, corn and peanut production. The vast majority (88 percent) of beneficiaries report an increase in production compared to the situation before the project; • Overall, more than 80 percent of beneficiaries report an improvement in their livelihoods 14, and 92 percent consider that subprojects have responded to their priority needs; • Higher scale of production and social cohesion facilitated the farmer organizations link to institutions of micro-finance. The majority of producer organizations are now members of financing institutions, although they use them mostly for saving purposes, and 6,467 saving schemes have been established. Table 3.2.3.1: Percentage of beneficiaries registering an increase in income of more than 30% Type of subproject 2007 2008 2009 2010 2011 Agriculture 86.0 90.7 81.4 93.0 88.4 Livestock 77.8 78.6 80.2 75.4 74.6 13 Impact Evaluation of Productive Investment Subprojects under the Rural Development Support Project (PSDR), December 2012; and Economic Evaluation of Productive Investment Subprojects under the Rural Development Support Project (PSDR), May 2012, which focused on a representative sample of subprojects financed in 2007 and 2008 under the OP. 14 Improved income had a positive impact on the living conditions of families in terms of better housing, access to basic social services (education, health) and agricultural inputs. 19 Fisheries 80.0 85.0 76.7 83.3 83.3 Social Infrastructure 75.9 82.7 80.2 82.9 83.7 Economic Infrastructure 65.4 69.2 69.2 74.4 79.5 AGR 70.9 78.9 64.5 73.5 75.8 Aggregate 75.2 81.4 76.9 78.2 80.1 Source: Beneficiaries survey 2012/Cabinet Miaramita 3.2.5 Indicator 2: Infrastructure functioning one year after completion and mechanism in place to finance operations and maintenance (85%). Partially Achieved: Ninety percent of financed infrastructure is still functional one year after completion, is in very good condition, and has been built according to specifications. However, the arrangements for operation, management and maintenance remain unclear and not well understood by most beneficiaries, particularly regarding subdivision of responsibilities, determination of O&M costs, and mode of payment and collection of maintenance fees, which are often insufficient. That said, the following elements point to a positive trend: (i) legally binding community agreements (DINA) describing roles and responsibilities for O&M activities are in place for the majority of producer organizations; (ii) Management Committees have been formed for each infrastructure site, composed of representatives of beneficiaries and users associations; (iii) basic training on O&M has been provided and organizations have been restructured and their governance bodies put in place; and (iv) saving funds have been created by some organizations to cover ordinary maintenance. 3.2.6 Indicator 3: Subprojects functioning one year after completion (85%). Partially Achieved: Ninety percent of beneficiaries under the AF are using the agricultural techniques and technology acquired under the project one year after completion of subprojects, 15 while 80 percent of subprojects financed under the OP are still operational one year after completion 16. The majority (90 percent) of beneficiaries are willing to continue their subproject after project closing, although only 72 percent report having the material means and 47 percent the financial resources to do so. End-of-project impact evaluation studies show that 66 percent of subprojects have a satisfactory IRR (21%) compared to 58 percent in 2008. These figures are indicative of the increased awareness of the need for farms modernization on the part of producers, who were for a long time weakened by the lack of access to training and technical assistance. 3.2.7 Indicator 4: Subprojects having satisfactorily developed and implemented an Environmental Action Plan (70%). Achieved: Field assessments conducted in December 2012 as part of the project final evaluation estimate that 75 percent of subprojects have adopted an Environmental Action Plan (EAP) during implementation, while 87% of beneficiaries states that they comply with such Plan. The positive result was consolidated during the AF, through the improved supervision carried out by the “technical pools�. The final subproject impact evaluation reports that 91 percent of producer 15 This indicator has been measured based on a comparative analysis of information related to productive subprojects implemented by beneficiaries during the 2010-2011 and 2011-2012 agricultural seasons. 16 Data from field teams’ reports. 20 organizations apply the environmental measures recommended by the EAP, with reforestation being the measure adopted by most of beneficiaries. The project went beyond a narrow approach covering only the screening of subprojects and the specification of mitigation measures to one that actively promoted sound environmental management by: (i) developing alternative agricultural practices to slash-and-burn production systems; (ii) launching efforts to include environmental considerations into commune development plans; (iii) exploring revenue- generating potential of non-timber forest products; and (iv) facilitating forest-management transfer contracts with communes. Further, links between PSDR and the Third Environmental project helped to a large extent maintain population support for the national park system and prevent continuing encroachment. 3.2.8 Indicator 5: Direct project beneficiaries (1,060,000), of which female (29%). Achieved: The project benefitted a total of 1,190,295 beneficiaries, 34 percent of those being women. Through strong participation in the elaboration and implementation of subprojects, and their gradual integration in production activities, women are now enjoying an increased share of responsibility within the producer organizations, and play an important role in the maintenance of activities. 3.3 Efficiency 3.3.1 The evaluation of PSDR efficiency was made difficult by the lack of a complete set of quantitative data. An economic and financial analysis reviewed 4 types of productive subprojects (rice, beef, dairy, and poultry) collectively representing about 64% of the most common subprojects financed by end-2012. The productive investment subprojects (Component 1) plus associated project management expenses represented about 95 percent of the overall project cost, thus, the analysis of the efficiency of productive investments gives an adequate assessment of the overall project efficiency. The methodology and results are detailed in Annex 3. Main results are as follows: 3.3.2 Overall project’s efficiency: Using cash flows from four types of subprojects (rice, beef, dairy and poultry), financial and economic indicators were calculated to evaluate the overall project profitability. Both financial and economic NPV were positive. Financial and economic IRR were greater than the opportunity cost of capital (OCC) of 12 percent, at 38% and 26% respectively. 3.3.3 IRRs of subprojects: The efficiency analysis on the beef, dairy, rice and poultry subprojects used financial and economic indicators such as net present value (NPV) and internal rate of return (IRR), with conservative key assumptions. IRRs ranged from highs of 38% and 29% for dairy and rice respectively, and lows of 11% for beef. NPVs were high and positive, except for the beef model. Findings suggest that financial returns are robust for all subproject types except beef, even under conservative assumptions. 3.3.4 The analysis for the beef model was carried out under both a “before financing� and “after financing� scenario 17. The beef model subproject “before financing� was not financially viable given that the IRR was less than the OCC, and the NPV was negative. This was also true in 17 The analysis under a “before financing� and “after financing� scenario was applied only to the beef model because, based on data available, the beef model was the only subproject that required loan repayment. For the other subprojects, which were financed by grants, the “before financing� and “after financing� scenario did not apply. 21 the “after financing scenario, with some improvement in the indicators. With a change in the discount rate from 12% to 11%, which is realistic considering the limited opportunities in the project area, the NPV after financing was positive and suggested an attractive subproject 18. With a discount rate of 10 percent, the NPV before and after financing was positive. 3.3.5 For the dairy model, the subproject was financially viable (IRR of 38% and positive NPV) but in terms of economic efficiency (labor market adjustment) the project was not attractive with a negative economic NPV and an ERR less than the OCC. For the rice model, conservative assumptions on yield were considered, and the financial and economic efficiency indicators showed profitable rice subprojects, with an IRR of 29% and positive NPV. 3.4 Justification of Overall Outcome Rating Rating: Moderately Satisfactory 3.4.1 The overall outcome rating of Moderately Satisfactory is based on: (a) original and continuing high relevance of project objectives and design to the needs of the poorest rural population in Madagascar; (b) an overall moderately satisfactory outcome for PDO achievement, tempered by concerns about O&M practices and sustainability of subprojects. With some exceptions, the project was able to meet/exceed intermediate outcome targets under both the OP and AF, although with delays and bottlenecks caused by political crises and misprocurement; and (c) indications of economic and financial efficiency based on an analysis of a limited sample of successful productive investments. Further, there is evaluative evidence of substantial institutional development at the local level and in community associations. Overall, the project provided an appreciable contribution to the improvement of agricultural development and rural incomes in support of Madagascar’s rural development strategies. 3.5 Overarching Themes, Other Outcomes and Impacts (a) Poverty Impacts, Gender Aspects, and Social Development 3.5.1 Poverty impact: Poverty reduction in rural areas was a specific objective of the project. However, as mentioned in section 2.1.3, there is an attribution problem, considering the difficulty of directly linking to the project specific impacts on poverty reduction in rural areas. Consequently, the ICR focuses its analysis on poverty reduction among beneficiaries. The results of economic and financial analysis, beneficiary responses about wellbeing, employment and income show the project’s actual and potential poverty impact. A 2012 beneficiaries’ survey reports that the percentage of poor beneficiaries decreased by 33 percentage points as a result of the project (see Graph 3.5.1.1). A high proportion of productive investments were subsistence support which alleviates poverty. The large number of agricultural production investments, improved irrigation infrastructure and access to microfinance services has positioned many rural families for a higher level of productive activity; poverty reduction is now latent in the productive assets of many poor communities. Further, poverty targeting was accurate and independent studies confirm that the project reached its targeted poor clientele under both phases. Specifically, PSDR was successful in reaching out to population in isolated areas, which, given the strong correlation between poverty and isolation, is a positive sign as far as its poverty impact is concerned. 18 The analysis with different discount rates was done for the beef model because the NPV for this model was negative. For other subprojects it was not applied given that their NPVs were high and positive. 22 Graph 3.5.1.1: Poverty trend among beneficiaries: Source: Survey of beneficiaries’ households, 2012, Cabinet Miaramita 3.5.2 Gender: The project was serious about reaching women and made significant progress. Women’s participation in productive subprojects exceeded the 29% target: over 34% of productive subprojects were presented/executed by associations led by women. More than 30 percent of the members of farmer organizations supported by the project were women. More and more women are integrated in productive activities and some of them run farms. Where they were strongly represented (e.g., rice husking, livestock, sewing, handicrafts), the organizations had generally higher levels of savings and income than those dominated by men. This is especially satisfying since the PAD expected women to benefit primarily from productive subprojects. (b) Institutional Change/Strengthening 3.5.3 The project contributed to a re-dynamization of rural development, the re-emergence of synergy and collaboration among actors responsible for rural services and a renewed interest in the modernization and diversification of agriculture activities, especially in the absence of formal agriculture extension services in Madagascar. The numerous partnerships facilitated by the project (between farmer organizations and private enterprises, micro-credit institutions, government, commercial banks, research organizations and other donors) show notable evolution. However, the decision to build a strong PIU with parallel implementation structures -- which was seen as necessary to deliver a multitude of subprojects in a difficult environment -- did not allow the project’s integration into the regional entities, and eventually the building of a strong decentralized institutional structure. 3.5.4 At the local level, improved organization and professionalization motivated by a sense of empowerment and genuine appropriation of the project’s goals to curb chronic food insecurity boosted the status and legitimacy of the farmers and producer organizations. Institutional growth was also demonstrated by the following: increased capacity of groups benefiting from subprojects to participate in planning and management of local development, manage and administer funds; demonstrated efforts to form strategic alliances with the municipalities and other local organizations for commercial purposes; expressed awareness of the importance of maintaining their investments; increased bargaining power vis-à-vis the various partners (local authorities, financial institutions) and new-found readiness to approach local authorities about community needs. (c) Other Unintended Outcomes and Impacts (positive or negative) 23 3.5.5 PSDR survived political and internal operational issues to become the Government’s flagship sector investment program. It is widely-viewed as a model of a new relationship between the national government and the private sector through a shift away from the traditional top-down approach, all major catalysts to regional commitment to the program. At the public/private sector level, PSDR has become the brand-name public policy instrument for support to small and medium-sized producers. 3.6 Summary of Findings of Beneficiary Survey and/or Stakeholder Workshops 3.6.1 Beneficiary Surveys. A comprehensive beneficiary survey 19 was conducted in 2005 to gauge beneficiary organizations’ attitude. Survey results showed: (i) overall 77 percent of beneficiaries were satisfied with the project and its response to their expectations and priority needs. The project was perceived as having improved and diversified production activities thus contributing to reduced food insecurity; (ii) most beneficiaries reported being satisfied about the management of funds which was found to be transparent by 73 percent of beneficiaries; (iii) most beneficiaries believed farmers organizations had reached a greater level of maturity; (iv) women’s participation in producer organization had increased, with more than 32 percent of members being women; and (v) a positive trend in the establishment of links with micro finance institutions was perceived, with 53 percent of producer organizations becoming a member. The main areas of weakness were found to be: (i) training and the lack of customization to the specific needs of beneficiary producers; (ii) the lack of adequate monitoring at the regional level; (iii) the lack of a quality control mechanism for the provision of inputs; and (iv) the identification of market access. In addition, a few Borrower studies based on beneficiary surveys were carried out during project implementation. Findings are summarized in Annex 5. 3.6.2 Two separate rounds of stakeholder consultations were conducted to obtain feedback and distill lessons learned. The first round of discussions involved a workshop held in July 2012 with the key agencies and individuals with strong influence over the direction of agriculture and rural development strategies and policies. The second round of discussions, held in November 2012, involved a wide range of participants including representatives from 22 Producer Organizations (one each from the 22 Regions benefitted by PSDR), the Ministries of Agriculture, Fisheries and Livestock, the Regional Directorates for Rural Development, the Agriculture Support Centers (CSA), the private sector, and researsch institutions (FOFIFA). Key feedback obtained include: (i) so as to ensure homogeneity, criteria for group formation need to include the level of education in addition to the degree of expertise and experience; (ii) regular internal meetings, establishment of a network of partners, collaboration between the project and the line Ministry are key for effective management and coordination; (iii) mainstreaming of environmental considerations is essential, regardless of project scale; (iv) beneficiaries’ contribution needs to be established based on poverty level rather than geographical location; and (v) the establishment of a permanent monitoring system of subproject is fundamental for follow up and learning. Overall, the beneficiaries acknowledged the important contribution made by PSDR to the improvement of their quality of life and were generally satisfied with the support received from the project. In particular, it was noted that the project, through the introduction of beneficiaries’ contribution, helped promoting a change of mentality towards entrepreneurship and thus increasing ownership of activities and investments financed. 19 Beneficiaries’ evaluation of PSDR project, 2005, Cabinet MIARA-MITA covered 10.1 percent of subprojects implemented from project start until mid-2005, covering the country’s 22 regions. 24 4. Assessment of Risk to Development Outcome Rating: Substantial 4.01 Sustainability of subprojects is determined by the financial viability, the maintenance of organizational structures for implementation and the continued adoption of improved techniques, the existence of markets and compliance with environmental action plans for sustainable production. This rating takes into account the following: (a) the percentage of subprojects remaining functional by 2012 is high for the AF, but lower for the OP; (b) economic analysis of selected productive subprojects suggests varied degrees of sustainability; (c) the ability of producer organizations to negotiate with financial partners and establish savings mechanisms has grown, but their management capacity needs further strengthening, and their financial viability and access to credit remain weak; (d) networks of technical partners are being put in place, but their development remains precarious due to factors external to the subprojects, namely the isolation of production sites and the difficulties of communication for disseminating both technical and commercial information; (e) access to new markets remains a constraint, particularly due to isolation, lack of communication, and insufficient marketing activities (fairs and word of mouth are currently the most common channels used to market the products); (f) the socio-organizational capacity of producer organizations has improved but needs further strengthening. O&M practices are still weak but the probability of survival of producer associations that are in charge of operation and maintenance is promising due to follow up support of local authorities; (g) two out of three producer organizations have adopted the recommended technologies in the implementation of their subprojects. Given that such adoption has had a positive impact on productivity and income, prospects for broader dissemination are good; (h) sound environmental management is increasingly taking place which is an indicator of enhanced awareness of the importance of sustainable production. 5. Assessment of Bank and Borrower Performance 5.1 Bank Performance (a) Bank Performance in Ensuring Quality at Entry Rating: Moderately Satisfactory 5.1.1 Preparation of the project spanned some 18 months and included extensive background analysis and consultations at the central and regional level. The project design was innovative and consistent with the Government’s rural strategy and priorities and the Bank’s CAS, and reflected the core lessons of previous, similar operations. It was innovative in (i) adopting an approach to fostering co-participation between the central and regional governments on agricultural and rural development, and to institutional collaboration across sectors; and (ii) introducing a demand- driven approach to rural development and poverty reduction. However, as mentioned in section 2.1.2, the design of both the OP and AF presented some shortcomings (fragmentation of activities among several components, wide range of activities, processing of subprojects on an individual basis, decentralized approach based on the GTDRs) and was not always in line with capacity on the ground, causing delays in the initial phase. The Results Framework was reasonable, but the design of indicators to capture the full implications of revenue increase and poverty reduction was limited, while other indicators reflected unrealistic benchmarks to adequately measure performance. (b) Quality of Supervision Rating: Satisfactory 25 5.1.2 Supervision was characterized by: (i) regular, well-documented missions, with randomized field visits to communities and subprojects, and considerable follow-up with TTLs based in the field; (ii) a focus on fiduciary capacity and performance, with consistent, intensive supervision of procurement in the post-crisis period (2004-05); (iii) clear identification of capacity needs; (iv) strong MTR including comprehensive Action Plans, and sustained follow-up with PSDR, with some weaknesses though on the M&E aspects; and (v) intense, pragmatic, rigorous efforts to salvage the operation post-crises; (vi) quick diagnose of implementation problems and strong proactivity in seeking restructurings, especially during difficult times. (c) Justification of Rating for Overall Bank Performance Rating: Moderately Satisfactory 5.1.3 This rating reflects the Moderately Satisfactory preparation performance and the Satisfactory rating for supervision. It reflects the Bank’s efforts to innovate and test a CDD model. However, the project’s operational strategy and content under the OP was not matched by institutional capacity on the ground in key areas. This rating is reinforced by design flaws (e.g., on the beneficiaries’ contribution) in the AF which might have been avoided or at least mitigated by further analysis at appraisal and more timely action during supervision. Overall, during supervision, Bank teams were conscientious in working with the government and other partners in facilitating the implementation of the project. They clearly and accurately identified in a timely manner the operational bottlenecks and the appropriate remedial responses to be undertaken. 5.2 Borrower Performance (a) Government Performance Rating: Moderately Satisfactory 5.2.1 Government showed high degree of support for a new, decentralized and demand-driven project approach giving substantial latitude to the regions to determine their rural investment priorities. But, political pressure resulting in misprocurement and likely fraud and corruption, as well as severe political crises, put PSDR at risk with significant impact on its implementation, and effort to restore normalcy was arduous. Once conditions improved, Government reconfirmed PSDR as the hallmark of its agricultural and rural development program and its support, including budgetary, has not waned. Similarly, commitment to the Additional Financing and the follow-on operation has been strong, despite the ongoing political crisis. (b) Implementing Agency or Agencies Performance Rating: Moderately Satisfactory 5.2.2 The Ministry of Agriculture showed consistently strong, committed effort, even with formidable obstacles from 2009 onwards, to collaborate with the Bank team in moving the project forward. It was very responsive in restoring enabling conditions for effective project implementation following political and internal crises. However, implementation of component 2 and 3 (under the direct responsibility of the Ministry) was somewhat weak and characterized by poor accountability in the use of funds and lack of clarity about incremental benefits generated. The Project Implementation Unit had a monumental set of responsibilities, and was successful in gradually establishing an adequate regional field presence for the processing of a large number of subprojects all over the country. However, it focused too much on quantitative targets at the expense of qualitative ones, namely provision of technical assistance and screening of subprojects. (c) Justification of Rating for Overall Borrower Performance 26 Rating: Moderately Satisfactory 5.2.3 On balance, overall Borrower performance is rated Moderately Satisfactory based on the Moderately Satisfactory performance of both the Government and the implementing agency. The rating reflects: (i) the complex realities of the initial (2002) and the closing (2009) years which brought the project to a complete freeze; and (ii) the professionalism and dedication of the technical and administrative teams, close collaboration with the Bank in the post-crisis periods to rescue the project and maintain it alive, role in bringing PSDR to a successful close, and accomplishments in physical targets and environmental issues. 6. Lessons Learned 6.1.1 The following lessons are among the more important: Innovative CDD operations in the rural sector, especially when operating in a fragile environment, require an extended horizon, a high degree of flexibility, continuous support for progressive capacity-building and repeat financing to maximize impact, learning and sustainability. The time required for projects such as PSDR to implement and mature greatly exceeds Bank averages. This is even more relevant in fragile conditions, like those experienced in Madagascar, where projects need to constantly adapt to changing external circumstances. The importance to such projects of: (i) a high degree of flexibility to respond to external shocks and adapt in terms of the type of support they provide; (ii) unique outreach and implementation arrangements for channeling support where it is needed; and (iii) experienced, multi-skilled, cross-sector Bank and Borrower project teams with continuity of leadership cannot be over-stated. CDD projects can indeed become critical instruments for rapid response to external shocks. Differentiated targeting and support, and clarity of subprojects’ objectives are key factors to ensuring project sustainability. This type of project should differentiate between target groups and subproject objectives, particularly between (i) the poorest segments of the rural population, usually characterized by very low capacity, for whom the goal is ensuring food security; and (ii) farmers with proven farming experience and an entrepreneurial approach to production and marketing for whom the goal is to ensure financial sustainability over time. Financing criteria and the level of subsidy provided through matching grants should be adjusted to the degree of vulnerability of the targeted group, geographical location, and the nature of the subproject. Beneficiaries belonging to group (i) may need to receive grants, with beneficiary contributions assigned in-kind rather than cash, whereas group (ii) beneficiaries may be more apt to be linked to micro-credit, with beneficiary contributions in cash. Under the project, the capacity of farmer organizations to mobilize cash contributions was overestimated and not adjusted to local conditions. Such capacity was further adversely affected by the prolonged political crisis that broke out just before the AF became effective. Quality of subprojects needs to be vigorously pursued by paying continuous adequate attention to financial and commercial aspects. The promotion of new and economically viable businesses requires strong long-term technical and organizational support, including rigorous screening of potential agro-businesses. The market analysis is the critical point of a large number of small investment projects in rural areas. Under the project, the level of market analysis was very heterogeneous. In those cases where a link to the market was identified early on, with the establishment of commercial contracts that guaranteed a certain degree of security for the additional production generated by the subproject, subprojects remained functional with increased chance of sustainability. As such, subprojects should be promoted based on real market demand and according to the identified priority, commercially viable sub-sectors within each region. 27 Leveraging maximum impact from resources allocated for subprojects requires the identification of investment proposals in a transparent, strategic and participatory manner, promoting competition for support through targeting mechanisms which ensure high quality, sustainable subproject proposals consistent with project development objectives and sector priorities. A hard budget constraint, made known to communities and local authorities in the participatory planning process, is helpful for introducing a degree of rigor in prioritization. The sourcing and quality of technical assistance needs focused attention from the earliest stages of project development. Allocating a portion of subproject cost for technical assistance cannot resolve the fundamental problem of the scarcity and quality of such services in rural areas. CDD projects need to support an intense effort to build up such services over time, with a focus on: (i) building a sustained, long-term support for producer organizations; (ii) responding to the specific needs of potential entrepreneurs; (iii) tailoring the support to the unique technical, institutional and financial endowments of each region; (iv) ensuring adequate presence in rural areas through massive decentralization of staff and providers; (v) grouping service providers by area of specialty and introducing a black list of non-performers; and (vi) supporting "training of trainers." Targeted groups/beneficiaries need to be functional to ensure long-term sustainability of investments. Several evaluation studies show that many organizations appear to have been formed expressly for the purpose of securing a subproject and that beneficiaries tend to revert back to their “natural� grouping centered on households or to produce individually once they have accessed funds for implementing their productive activity. This shows that the conditions for effective collective action are more complex than shared vulnerability alone. As a general rule, support to production should target members in its natural production unit, while support to producer associations should focus around services where farmers can benefit from economies of scale (i.e., procurement, marketing, and training). As such, project design needs to include flexibility to allow for financing of smaller groups and/or individual enterprises. When CDD projects are heavily focused on demand-driven approaches tensions can arise between responding to the communities’ needs and building a critical mass necessary for sustained growth and rural development. As part of project design, a clear strategic and prioritazation framework, adjustable over time, is fundamental to ensure that CDD projects are strategically focused and have a catalytic impact on the rural space, rather than responding to ad hoc needs thereby dissipating the focus from the original objectives. Implementation arrangements for national programs requiring a strong and massive PIU are not conducive to strong alliances with local-level governments and institutions and decentralization efforts. The instability and lack of clarity in the institutional structures call for the establishment of sizable PIUs that ensure a substantial presence at the local level needed for implementation of projects such as PSDR. Such massive, parallel, decentralized system, however, is not an incentive to empower regional organizations and eventually has a feebler impact at the decentralized level. In the end, this may be a missed opportunity as evaluation study showed that subprojects that have been successful are those who have been able to develop and benefit from the development of partnerships and networks with other local institutions, producers federations and potential buyers. This has created a synergetic effect beneficial for subprojects in terms of improved accessibility to local services and markets. Establishing strong ‘firewalls’ in participatory planning and sub-project selection procedures is fundamental to protect projects from political influence. CDD projects can 28 easily fall victim to political manipulation, especially in fragile situations. Strong political interference and potential fraud and corruption should not be considered sufficient reasons to close an otherwise well performing project. Participatory mechanisms, strong and proactive corrective measures coupled with intensified supervision can go a long way in creating a stronger system of checks and balances and eventually putting a much needed project rapidly back on track. Strong corrective measures also provide an incentive to curb future corruption attempts. The development of synergies and collaboration with parallel operations through formal agreements can improve impact on the ground and sustainability. PSDR collaborated actively with several ongoing operations to avoid overlaps and enhance impact on the ground. Synergies were created with CELCO (the implementing agency for the Third Environmental Project – PE3) and ANGAP (now Madagascar National Parks) to help facilitate the selection by local populations living in and around national parks of alternative sources of income. Similarly, PSDR established important synergies with the Integrated Growth Poles (PIC) project which, focusing on strengthening farmers’ negotiating capacity and linkage to the markets, helped develop productive chains and increase the viability of PSDR-supported productive activities. 7. Comments on Issues Raised by Borrower/Implementing Agencies/Partners (a) Borrower/implementing agencies 20 (b) Cofinanciers N/A (c) Other partners and stakeholders N/A 20 The draft ICR was sent to the Government for review; however, the Bank did not receive any feedback. As a result no comments are presented in Section 7 above or letter entered in Annex 7. If, following submission of the ICR, the Bank receives any comments, they will be filed in the project archive/WBDoc. 29 Annex 1. Project Costs and Financing A. Original Project (3524-MAG) (a) Project Cost by Component (in USD Million equivalent) Actual/Latest Appraisal Estimate Percentage of Components Estimate (USD (USD millions) Appraisal millions) Productive Investments 62.33 68.09 109.25 Support Services 11.60 2.30 19.85 Community Development 6.00 0.00 0.00 Capacity Building and Policy 5.03 2.52 50.05 Development Project Administration and 10.71 17.48 163.21 Monitoring PPF 0.60 0.00 0.00 Total Baseline Cost 96.27 90.39 93.90 Physical Contingencies 1.68 0.00 0.00 Price Contingencies 8.14 0.00 0.00 Total Project Costs 106.09 90.39 85.20 Front-end fee PPF 0.00 0.00 .00 Front-end fee IBRD 0.00 0.00 .00 Total Financing Required 106.09 90.39 85.20 (b) Financing Appraisal Actual/Latest Type of Estimate Estimate Percentage of Source of Funds Cofinancing (USD (USD Appraisal millions) millions) Borrower 8.00 8.97 112.11 Local Communities 9.04 9.17 101.49 International Development 89.05 84.72 95.14 Association (IDA) B. Additional Financing (45250-MAG) 30 (a) Project Cost by Component (in USD Million equivalent) Actual/Latest Appraisal Estimate Percentage of Components Estimate (USD (USD millions) Appraisal millions) Productive Investments 28.20 19.23 68.20 Support Services 1.50 0.22 14.80 Capacity Building and Policy 2.10 0.55 26.15 Development Project Administration and 4.30 3.90 90.77 Monitoring PPF 0.00 0.00 0.00 Total Baseline Cost 36.10 23.91 66.22 Physical Contingencies 0.00 0.00 0.00 Price Contingencies 0.00 0.00 0.00 Total Project Costs 36.10 23.91 66.22 Front-end fee PPF 0.00 0.00 .00 Front-end fee IBRD 0.00 0.00 .00 Total Financing Required 36.10 23.91 66.22 (b) Financing Appraisal Actual/Latest Type of Estimate Estimate Percentage of Source of Funds Cofinancing (USD (USD Appraisal millions) millions) Borrower 0.00 0.00 .00 Local Communities 6.10 2.34 38.41 International Development 30.00 21.31 71.03 Association (IDA) 31 Annex 2. Outputs by Component 2.1 The components and main accomplishments under the project were as follows: 2.2 Component 1: Productive Investments and Support Services was to finance matching grants to organized rural community associations for about 4,000 and 2,200 small-scale investment subprojects under the Original Project and Additional Financing respectively. The component was to finance the following types of subprojects, both under the OP and AF: (i) investments in productive rural infrastructure; (ii) productive agricultural activities; (iii) productive non-agricultural activities; and (iv) the provision of advisory services and capacity building support for producer organizations. Following the 2002 restructuring, the component was also to finance extension services, technical support, and training to build capacity of farmers and producers organizations for the implementation of the demand-driven productive investments and other small scale non-productive activities. Total cost at project completion was 109% and 68% respectively, of the appraisal estimate. 2.3 The Project financed a total 9,510 subprojects benefiting around 238,000 families (without repetition) or some 1,190,000 persons (average 5 persons/family) implying that the Project reached more than 10 percent of the total rural population, and 50 percent of population living in remote rural areas. The project exceeded its subproject target under both Phase I (6,892 vs. target of 4,000) and Phase II (2,618 vs. target of 2,200). 2.4 About 85 percent of subprojects fell under the productive agricultural activities category, with productive infrastructure accounting for 12 percent and productive non-agricultural activities for 3 percent of total subprojects. In terms of cost, infrastructure development represented more than 44 percent of total component costs, agricultural activities 53 percent and non-agriculture activities 2 percent (see table 2.4.1). Table 2.4.1: Number and cost by subproject category (cumulative) No of financed % of total Cost (OP+AF) Subproject category % allocations subprojects subprojects (USD million) Productive 1177 12,38 33.02 44,37 Infrastructures Agricultural activities 8032 84,46 39.96 53,71 Non-agricultural 301 3,16 1.42 1,92 activities TOTAL 9510 100,00 74.40 100,00 Component 1 subprojects Source: Completion Report, Mamokatra S.A., December 2012 2.5 Productive Infrastructure. More than half of productive infrastructure subprojects were hydro-agricultural development indicating the importance attached to rice production, at least from the perspective of food security. At project closure, the number of small dams financed was 657 against a forecast of 643, a rate of realization of 102 percent. However, considering that 85% of subprojects supported agricultural production, the demand for infrastructure in support to upstream and downstream production operations appears small, with only 2 percent of subprojects dealing with processing of agricultural products and storage. These figures suggest a weak integration of subproject activities into market outlets. Achievements include: 32 • More than doubled additional area provided with irrigation and drainage services (30,069 ha vs. target of 14,300); • Water users provided with irrigation and drainage services largely surpassed the target (31,842 vs. 25,280); • Water user associations that are operational surpassed the target (657 vs. 643); • Infrastructure built or rehabilitated: 657 dams (vs. target of 657), 205 processing units (vs. target of 190), 169 storage facilities (vs. target of 165); and 28 slaughterhouses (vs. target of 30). 2.6 Productive agricultural activities. Subprojects focused on the development of techniques for rice, maize and peanuts cultivation, with positive increase in production. Specifically, achievements include: • The area cultivated with improved technologies and/or inputs provided through the subprojects increased by 42,927 ha, surpassing the target of 39,150ha; • Increase in rice production by 37 percent and 64,971 tons compared to a target of 30 percent and 64,200 respectively; • Increase in maize production by 24 percent and 3,282 tons compared to target of 20 percent and 4,600 respectively; • Increase in peanut production by 32 percent and 1,242 tons compared to target of 20 percent and 1,350 respectively; • Satisfactory average financial rate of return of subprojects (21% vs. target of 15%); • Area cultivated with improved technologies and inputs provided through Project-supported subprojects surpassed the target (42,927 ha vs. 39,150). • Saving schemes established and functioning surpassed the target (6,467 vs. 5,982); • Targeted producers largely satisfied with agricultural services received (94 percent vs. a target of 90 percent); • More than 753,000 client days of training provided. 2.7 Productive non-agricultural activities. These activities represented about 3 percent of total subprojects. The survey results consistently show that more than 70 percent of rural households engage in non-agricultural activities (crafts in particular, typically over 10%) as secondary activity and with a relatively negligible contribution to income (on average less than 5%). Requests for support in this area were limited, due to difficulties in the supply of inputs (fibers, scrap, etc.) and links to market. 2.8 The most common subprojects financed under each phase are ranked as per Table 2.6.1, including numbers of families benefited in each case. The types of subprojects are remarkably similar between phases, with the exception of the exit of vanilla production, beekeeping, and dairy cows raising. 33 Table 2.6.1: Ranking of Commonly-Financed Subprojects, Phase I (3524-MAG) and Phase II (4525-MAG) No of Families No of Families Subprojects Subprojects subprojects benefited subprojects benefited 3524-MAG 4,416 112,768 4525-MAG 1,923 34,953 Rice Rice 2,223 50,608 984 11,936 intensification intensification Hydro- agricultural 479 19,080 Marine fisheries 188 2,616 development Hydro- Marine fisheries 323 7,067 agricultural 178 12,762 development Beekeeping 236 5,693 Peanuts 108 1,403 Fish farming 226 6,353 Pig breeding 101 1,261 Cattle fattening 217 5,172 Cattle fattening 96 1,041 Vanilla 198 4,898 Vegetable crops 69 910 Dairy cows 176 4,927 Husking 67 1,186 raising Onion 173 4,646 Fish farming 67 869 Sheep/goat Sheep/goat 165 4,324 65 969 raising raising Source: PSDR 2.9 The table below shows that about 75% of OP subprojects proposed were not financed, and about 57% of AF proposals. The rejection rate by subproject category was higher for productive infrastructure, and lower for both agricultural and non-agricultural activities. In aggregate, the rejection rate was 71%. The reasons for non-financing were diverse. One basic reason is that the PIU did not track receipt of proposals in real time, so for quite some time they were unaware that many more proposals had been submitted than could ever be funded, and consequently they were slow to announce publically that receipt of new proposals was being suspended. The fact that so many proposals were received may have created false expectations among some potential beneficiaries, but it also allowed the PIU to be more selective in deciding which subprojects to finance (which was a good thing). Proposals were rejected for many different reasons: (i) incomplete proposal, or poor quality proposal; (ii) group submitting the proposal did not meet the eligibility requirements; (iii) funding requested for activities that were not eligible for project funding (e.g., power tillers, establishment of perennial crops); (iv) funding requested for activities deemed commercially non-viable; (v) need to maintain a regional balance in distribution of subprojects (in some regions the local authorities mobilized many beneficiary groups to submit proposals quickly in an effort to capture a large share of PSDR benefits). 34 Table 2.9.1: Subprojects Proposed, and Financed/Completed, 2001-2012 Original project Additional Financing Subproject categories % % Proposed Financed Proposed Financed financed financed Productive Investment 27 106 6892 25% 6090 2618 43% subprojects Productive infrastructures 4 247 881 21% 911 296 32% Agricultural activities 21 810 5752 26% 5076 2280 45% Non-agricultural activities 1 049 259 25% 103 42 41% Source: PSDR 2.10 Technical assistance. Technical assistance performance was fair under the OP but improved under the AF. The uneven quality of the "partenaires strategiques" under the OP was recognized and solved by developing an informal "black list" of poorly performing partners, and grouping the technical assistance providers by their areas of specialty. Technical assistance was provided to about 5,800 subprojects under the OP and about 3,737 subprojects under the AF. 2.11 According to a 2012 survey, only half of producer organizations were satisfied with technical assistance provided on the ground, due to (i) insufficient support for the elaboration of subproject profiles, (ii) lack of educational materials available to the beneficiaries, (iii) lack of consideration of the different crops’ seasonality (iv) inadequate and/or insufficient supervision, and (v) lack of transparency and insufficient communication. In direct response to the challenges experienced under Phase I, PSDR made a significant effort to compensate under the AF with the establishment of “pool techniques�. As a result, training and subproject supervision was delivered on a more permanent basis. This new arrangement allowed the project to: (i) monitor subproject activities in real time thanks to an effective and timely feedback system; and (ii) prioritize interventions. 2.12 Component 2: Support Services was to finance support for the strengthening of the capacity of main agricultural research institutions and better dissemination of the stock of new and promising technologies. Total cost at completion was about 20% and 15% of the appraisal estimate, respectively. One reason explaining why the cost of this component at completion was far below the initial estimates is that the public agency that was charged with implementation of Component 2 (FOFIFA) had very weak capacity and was very slow to submit and then implement the agreed work plans. Further, under the AF, after PSDR was granted an exemption to the disbursement freeze that had been implemented after OP 7.30 was triggered, direct transfers of funds to government agencies was prohibited. 2.13 Thematic research. Two public institutions FOFIFA (National Center for Applied Research for Rural Development) and FIFAMANOR (Fiompiana Fambolena Malagasy Norveziana) received funding under PSDR to conduct thematic research activities to address medium or long-term constraints to the intensification, diversification and/or sustainability of production systems. FOFIFA research activities, covering the entire country, focused on six themes including plant and animal genetics, soil fertility, phytosanitary management, social sciences and the agri-food sector. FIFAMANOR worked in one target region on two themes, (i) 35 conservation and management of genetic resources and (ii) agro-food research. Technical sheets for each regional crop were developed and used for developing training materials for service providers supporting the development of investment subprojects, thereby ensuring the link between research and productive activities on the ground. Support under this component has contributed to obtaining better performing varieties of tubercles, and to improving the ration formula for feeding dairy cows. The dissemination of research results was done through the institutions’ own extension services and partner NGOs. It should also be noted that support for thematic research has enabled these institutions to acquire scientific and technical equipment to improve research results. 2.14 Applied research. The Competitive Agricultural Research Fund (FCRA) established under PSDR financed 26 research projects under the OP; eight of these subprojects focused on data collection to fill significant gaps in terms of basic scientific and technical data. Proposals for applied research themes generally came from productive investment subprojects. Just over half of the 140 themes proposed during the OP came directly from producers facing problems in the implementation of subprojects; 21 percent of proposals came from branches of the Ministry of Agriculture, while requests from the private sector and research institutes accounted for only 8.7% and 6.7% respectively. Results were disseminated through 5 documentaries, CDs, and FCRA and PSDR websites. Under the AF, several research projects could not be developed due to delays in procurement processes and were cancelled. 2.15 Main results under this component are the following: • Research studies developed did not achieve the target (26 vs. 36). Ten research projects, which were experiencing delays in processing, were cancelled following the partial cancellation of the AF credit in November 2012; • The National Germplasm Bank was rehabilitated and material at risk regenerated; • The national agricultural research strategy has not been completed. A preliminary diagnostic study on the national agricultural research system has been prepared as a basis for the development of the strategy. • Eighty seven percent of producers have adopted improved agricultural technologies developed under the project. 2.16 Component 3: Capacity Building and Policy Development was to finance technical assistance and training to build capacity in implementing agencies at central and regional level to formulate and analyze rural policy, to establish adequate agricultural statistical systems, and to strengthen environmental assessment. Total cost at completion was about 50% and 26% of the appraisal estimate, respectively. Similarly to Component 2, the cost of this component at completion was far below the initial estimates, reflecting: (i) the very weak capacity of the public agency responsible for implementation of Component 3 (UPDR); and (ii) the prohibition of direct transfers of funds to government agencies following the disbursement freeze that had been implemented after OP 7.30 was triggered. 2.17 The component implementation was the responsibility of the for Rural Development Policy Unit (UPDR), which played a strategic role in supporting decision making through the preparation of studies aimed at assessing the impact of macro-economic, legislative, regulatory and tax provisions on rural areas and at supporting the development and monitoring of rural policies. Main results under the component are the following: Under the OP: 36 • The National Rural Development Program (PNDR) was developed and approved; • All planned Regional Rural Development Programs (PRDR) were prepared on the basis of the PNDR; • The Rural Development Policy Letter was developed and used for the elaboration of the PNDR; • The agriculture census was completed and disseminated; • The agricultural statistics system is operational, Regional Agricultural Statistics Units (URSA) were expanded and strengthened and newsletters on the market information system (SIM) were developed and published. Under the AF: • Planning for the elaboration of the Agriculture Sector Plan (PSA) was developed, with regional workshops carried out in all 22 regions; • Agriculture Support Centers (CSA) were partially strengthened (21 vs. a target of 40), and CSA was established; • Regional Grant Approval Committees were established (12 with PSDR financing and 4 with other donor funds for a total of 16 as per the target). 2.18 Component 4: Project Administration and Monitoring financed project administration and coordination activities, including supervision, monitoring and impact evaluation (M&E). Total component cost under each phase was 163% and 91% respectively of estimated cost at appraisal. The amount disbursed under Component 4 exceeded the budgeted amount and was in excess of the 15% rule of thumb figure that is often used as a reasonable amount. The main reason why Component 4 was overspent, is that key PIU staff were kept on board during the extended period under OP 7.30 when disbursements were frozen (their salaries and benefits were paid using project funds that were in the designated accounts when the disbursement freeze went into effect). 37 Annex 3. Economic and Financial Analysis 3.1 The evaluation of PSDR efficiency was made difficult by the lack of a complete set of quantitative data. An economic and financial analysis reviewed 4 types of productive subprojects (rice, beef, dairy, and poultry) collectively representing about 64% of the most common subprojects financed by end-2012. The productive investment subprojects (Component 1) plus associated project management expenses represented about 95 percent of the overall project cost, thus, the analysis of the efficiency of productive investments gives an adequate assessment of the overall project efficiency. 3.2 Overall project’s efficiency: Using cash flows from four types of subprojects (rice, beef, dairy and poultry), financial and economic indicators were calculated to evaluate the overall project profitability. Both financial and economic NPV were positive. Financial and economic IRR were greater than the opportunity cost of capital (OCC) of 12 percent, at 38% and 26% respectively. 3.3 Economic and financial results of individual subprojects. Because of data availability, the analysis was undertaken for four types of productive subprojects (rice, beef, dairy, and poultry) collectively representing about 64% of the most common subprojects financed by end- 2012. 3.4 Beef. Two financial indicators were calculated: the NPV and IRR before and after financing 21. Before financing, using a 12 percent discount rate, the project was not financially viable because the IRR was less than the assumed opportunity cost of capital (OCC). Furthermore, the NPV was negative. This was also true for the after financing, with some improvement in the indicators. An IRR equal to11.5% (Table 3.4.1) meant that the cost and benefits of the beef project were almost equal and the project might have been financially attractive enough to be implemented. TABLE 3.4.1: FINANCIAL COST BENEFIT ANALYSIS INDICATORS FOR BEEF Before financing After financing FRR 10.9% 11.5% NPV -685 -238 3.5 With a change in the discount rate from 12 to 11 percent, which is realistic given the limited opportunities in the project area, the NPV before financing was negative (-32 Ar), while the NPV after financing was 274 Ar, which yielded an attractive subproject. With a discount rate of 10 percent, the NPV was 705 Ar before financing and 862 Ar after financing (see Table 3.5.1). 21 The analysis under a “before financing� and “after financing� scenario was applied only to the beef model because, based on data available, the beef model was the only subproject that required loan repayment. For the other subprojects, which were financed by grants, the “before financing� and “after financing� scenario did not apply. The financial cost benefit analysis “before financing� means the exclusion of total debt service (including the interest repayment on the loan disbursement at the start of the project, the unpaid loan balance after loan repayment, and loan repayment (principal)). The cost benefit analysis “after financing� included the cash “before financing� plus the loan disbursement deducted from the total debt service. 38 Table 3.5.1: NPV and changes in the discount rate – beef Base discount rate Discount with 11% Discount with 10% (12%) NPV before -685 -32 705 financing NPV after financing -238 274 862 3.6 Sensitivity of beef subprojects. In order to test the robustness of the beef subprojects, a sensitivity analysis was performed on different financial and economic indicators (NPV, IRR). Two types of sensitivity analysis were used: the Variable-by-Variable (VBV) analysis and the scenario analysis. The VBV consisted of increasing the total cost and decreasing total benefit of the subprojects with different amounts (+5%, +10% and 20%). The scenario analysis made those changes simultaneously. The VPV analysis suggested that the project was sensitive to an increase of aggregate costs or decrease of aggregate benefits: from a decrease of 5% and plus in benefits and an increase in costs of 5% and plus, the project became unprofitable (negative NPV and EIRR less than the OCC. See table 3.6.1). Table3.6.1: VBV sensitivity analysis of beef subprojects IRR NPV After financing Before financing After financing Before financing Base case 11,52% 10,95% - 238 - 685 Costs (+5%) 9,90% 10,15% - 1 116 - 1 254 Costs (+10%) 8,49% 9,39% - 1 994 - 1 994 Costs (+20%) 6,12% 7,99% - 3 751 - 2 960 Benefits (-5%) 7,82% 8,48% - 2 120 - 2 235 Benefits (-10%) 4,21% 5,81% - 4 002 - 3 786 Benefits (-20%) - - - 39 478 - 33 145 3.7 Dairy. Financial and economic analyses were performed, with the use of a subsidy rate of 80% for the financial analysis, and of a conversion factor for labor of 0.5 for economic analysis. The project was financially viable (Table 3.7.1) since the IRR (38%) was greater than the OCC of 12 percent, and the NPV showed that the project would generate 6.8 million Ariary. In terms of economic efficiency indicators, the project was not attractive; with a negative economic NPV and relatively small IRR (<12%). Table 3.7.1: Efficiency indicators for dairy model Financial efficiency Economic efficiency indicator indicator NPV (Ariary) 6,816,037 -577,893 IRR/EIRR 38% 9.2% 3.8 Sensitivity of dairy subprojects The sensitivity analysis showed that dairy subprojects were robust since the financial NPV is positive for increase of costs from 5% to 20%, and decrease of benefits by the same percentage change. Similarly, the IRR was greater than the OCC for these changes. On the contrary, the economic NPV was negative with an increase of costs (or 39 a decrease of benefits) of 10% and more. Likewise, the economic IRR was less than the OCC for an increase in costs (or a decrease in benefits) by 10% and more (table 3.8.1). Table 3.8.1: VBV sensitivity analysis of dairy subprojects Economic Financial IRR Financial NPV IRR Economic NPV Cost (+5%) 33,4% 7 596 912,12 13,3% 484 886 Cost (+10%) 28,4% 6 231 783,29 10,5% - 598 340 Cost (+20%) 20,3% 3 501 525,64 5,8% - 2 764 791 Benefit (-5%) 33,1% 7 148 810,07 13,1% 406 480 Benefit (-10%) 27,4% 5 335 579,20 10,0% - 755 151 Benefit (-20%) 16,8% 1 709 117,45 3,6% - 3 078 413 3.9 The scenario approach for the economic indicators showed that the dairy subprojects were not sensitive to simultaneous changes in costs increase and benefits decrease. The financial NPV was negative and the financial IRR less than the OCC only when the costs increased by 10% coupled with a decrease of more than 20% in benefit or the benefits decreased by 20% with a simultaneous increase of costs by 10% and more. Hence, these efficiency indicators were not sensitive to small changes in aggregate costs and benefits (5%-10%). Conversely, the economic NPV and IRR are very sensitive to small changes (increase of costs by 5% and more and a decrease in benefits by 5% or more yields negative efficiency indicators. Table 3.9.1). Table 3.9.1: Scenario sensitivity of dairy subprojects Benefit (-5%) Benefit (-10%) Benefit (-20%) Cost (+5%) 28,0% 22,8% 12,9% IRR Cost (+10%) 23,5% 18,7% 9,3% Cost (+10%) 16,0% 11,7% 2,6% Cost (+5%) 5 783 681 3 970 450 343 989 NPV Cost (+10%) 4 418 552 2 605 322 - 1 021 140 Cost (+10%) 1 688 295 -124 936 - 3 751 398 Cost (+5%) 10,3% 7,3% 1,1% ERR Cost (+10%) 7,7% 4,8% -1,4% Cost (+20%) 3,1% 0,3% -6,4% Cost (+5%) - 676 745 - 1 838 376 - 4 161 638 ENPV Cost (+10%) - 1 759 971 - 2 921 602 - 5 244 864 Cost (+20%) - 3 926 422 - 5 088 053 - 7 411 315 3.10 Rice. Conservative assumptions on yield over time have been assumed for the rice subprojects. Quantities of inputs under a “without project� situation were on average half of the quantities “with the project situation�. The financial cost of labor for rice (2500 Ar. /day) was adjusted to an economic cost of labor (1250 Ar. /day), or a conversion factor of 0.5. Other inputs were costed using current market prices. For the rice output price, the current export parity price (633 USD/T or 1012 Ar. /kg) was derived from the calculation done by the impact evaluation study as shown in Table 3.10.1. The financial and economic efficiency indicators showed profitable rice intensification subprojects, with a financial IRR (29%) greater than the OCC and a positive financial NPV (339873 Ar.). With assumptions on adjustment of labor costs, rice output 40 price, and on inputs quantities, the economic NPV was approximately 4.2 Million Ar., indicating a high profitability of the rice subprojects. Table 3.10.1: Export parity for rice Item Value US Dollar/ AR rate 1600 Paddy Lac Alotra 500 White rice (60%) 833 Processing cost 20 Bags – new 50 kg 10 Fumigation 10 Transport to Port 40 Loading/Agency Fee (2.5%) 23 Bank finance (3%) 28 Commercial margin (5%) 48 Export Price AR/kg 1012 Current Export Price (US dollars/m ton) 633 3.11 Sensitivity of rice subprojects. In order to assess the strength of the assumptions for the rice subprojects a sensitivity analysis was performed on different financial and economic indicators, such as financial and economic NPV. The VBV consisted of increasing the aggregate cost or decreasing the aggregate benefit of rice subprojects with different amounts (+5%,+10% and 20%). The intensification rice subprojects were not profitable financially with a total cost increase of over 20% or a decrease in benefit of the same amount. Nevertheless, the projects were economically sound, with a positive economic NPV with VBV. For the scenario analysis, the NPV for the rice model was sensitive to a simultaneous change in aggregate costs by 5% and more (decrease) and increase in aggregate benefits by 10% and more. But the sensitivity analysis for the Economic NPV indicated that the project was still viable; the rice subprojects are robust in the VBV and Scenario analyses (table 3.11.1). Table 3.11.1: Sensitivity of efficiency indicators of rice subprojects Economic Scenario analysis Financial Variable by variable Financial NPV NPV NPV Economic NPV cost +5% Benefit - cost +5% 205 033 4 088 181 5% 53 200 3 780 871 cost +5% Benefit - cost +10% 70 194 3 990 180 10% - 98 633 3 473 560 cost +5% Benefit - cost +20% - 199 486 3 794 176 20% - 402 300 2 858 938 cost +10% Benefit Benefit -5% 188 040 3 878 872 -5% - 81 640 3 682 869 cost +10% Benefit Benefit -10% 36 206 3 571 561 -10% - 233 473 3 375 558 cost +20% Benefit Benefit -20% -267 461 2 956 940 -5% - 351 319 3 486 865 41 Annex 4. Bank Lending and Implementation Support/Supervision Processes (a) Task Team members Responsibility/ Names Title Unit Specialty Lending Ivar Serejski Task Team Leader Alassane Sow Economist Isinael Ouedraogo Economist Jean Delion Rural development Specialist Devendra Bajgain Natural resources Specialist Jumana Farah Natural resources Specialist Ousmane Seck Ag. Extension Specialist Gervais Rakotoarimanana Financial Specialist Menahem Prywes Economist Jerome Gauthier Livestock Specialist Eileen Murray Community Development Specialist Bienvenu Rajaonson Environment Specialist Ziva Razafintnsalama Social Development Specialist Sylvain Rambeloson Procurement Specialist Marie-LouiseA h-Kee Procurement Specialist Luc Lapointe Procurement Specialist Amadou Tidiane Toure Quality Enhancement David Freese Disbursement Officer William Marke Disbursement Officer Raj Soopramanien Counsel Susanne Holste Rural Roads Specialist Irene Xenakis Quality Enhancement Willem Zijp Peer Reviewer Adolfo Brizzi Peer Reviewer Jacob Kampen Peer Reviewer Michele Rajaobelina Program Assistant Virginie Vaselopulos Program Assistant Rondro Malanto Rajaobelison Program Assistant Supervision/ICR Martien Van Nieuwkoop Sector Manager (TTL until 2005) AFTA1 Lead Operations Officer (TTL 2005- Sofia U. Bettencourt AFTN2 08) Juerg Brand Consultant AFTN1 Joseph Byamugisha Financial Management Specialist AFTME Jean-Christophe Carret Sector Leader AFTSN Renee M. Desclaux Senior Finance Officer CTRLD Cynthia Mireille Faure Team Assistant MNCMI Paul-Jean Feno Senior Environmental Specialist AFTN1 Olivier Jenn-Treyer E T Consultant AFTN1 Jean Charles Amon Kra Sr Financial Management Specialist AFTME Ellena Rabeson Operations Officer AFCS4 42 Rondro Malanto Rajaobelison Program Assistant AFCS4 Bienvenu Rajaonson Senior Environmental Specialist AFTN1 Francois Marie Maurice Sr Financial Management Specialist AFTME Rakotoarimanana Sylvain Auguste Rambeloson Senior Procurement Specialist AFTPE Liliane Randrianarivelo Consultant AFTME Alain Pierre Randrianjohary Program Officer AFTN1 Lova Niaina Ravaoarimino Procurement Specialist AFTPE Lead Agriculture Economist (TTL Michael Morris AFTA2 2008-2010) Senior Rural Development Specialist Ziva Razafintsalama AFTA2 (TTL 2011-2012) Anna Corsi Land Administration Specialist LCSAR Consultant Majid Benabdellah Economist FAO (b) Staff Time and Cost Staff Time and Cost (Bank Budget Only) Stage of Project Cycle USD Thousands (including No. of staff weeks travel and consultant costs) Lending FY97 17.33 FY98 94.80 FY99 151.18 FY00 36 125.69 FY01 88 472.12 FY02 1 0.36 Total: 125 861.48 Supervision/ICR FY00 0.25 FY01 0.08 FY02 48 136.85 FY03 40 112.09 FY04 37 186.78 FY05 38 150.23 FY06 52 100.85 FY07 52 117.37 FY08 34 186.61 FY09 9 0.00 FY10 14 84.29 FY11 19 88.34 FY12 27 96.80 FY13 13 70.25 Total: 383 1331.51 43 Annex 5. Beneficiary Survey Results 5.1 The Project was the subject of a comprehensive beneficiary survey and a few survey- based studies, which give important insights into PSDR achievements. The results are summarized below. Beneficiaries’ Evaluation of PSDR (Cabinet MIARA-MITA, 2005) 5.2 The evaluation was conducted on the basis of a sample of 600 subprojects implemented or under implementation since the beginning of the project until mid-June 2005 and distributed in the country’s 22 regions. Agriculture production activities represented 45% of the sample subprojects, with livestock representing 37% of the sample. Results were as follows. 5.3 Overall, the project has responded to the priorities and expectations of beneficiaries and greatly contributed to promote a new dynamism in rural development. Further, access to rural services has favored a renewed interest in the modernization of agriculture and livestock activities. The majority of beneficiaries have experienced an increase in production following the implementation of subprojects, with positive impact on income and livelihood. The responses provided by beneficiaries with regard to the project key indicators revealed that the project was on track to achieve its objectives, particularly regarding the rate of adoption of new technologies and the increased ability to establish savings. However, overall satisfaction with technical assistance providers was low, while satisfaction with participatory processes was high, thus fostering the consolidation of farmer organizations. 5.4 Productive infrastructure subprojects. Overall, beneficiaries were satisfied with the implementation of productive infrastructure subprojects, which were perceived as having a positive impact on production and consequently on income. Beneficiaries felt that they actively participated in the decision –making regarding subprojects identification and selection, but such participation weakened during the implementation phase with construction firms taking over responsibility. Collaboration between local authorities and technical assistance providers was weak, with lack of clear definition of roles for the construction and maintenance of infrastructures. Less than half of beneficiaries are satisfied with the service provided by technical assistance providers. 5.5 Livestock activities. The majority (63%) of beneficiaries were satisfied with the subprojects and the impact on production, livelihood and local development. Subprojects were considered a priority for almost all beneficiaries, who participated actively in their selection, implementation and maintenance. More than half of beneficiaries believed that the subprojects contributed to the achievements of their objectives, and were satisfied with service providers’ performance, while 80% perceived that funds were allocated in a transparent manner. 5.6 Agricultural activities. The majority of beneficiaries (79%) perceived that subprojects financing agricultural activities have responded to their expectations, while less than half (42%) experienced an increase in income as a result of enhanced production. The majority (75%) believes that the project improved their living conditions, with 69% feeling responsible for the material and financial inputs provided by the subprojects. Only 37% of beneficiaries were satisfied with their participation in the procurement processes, and about half were satisfied with service providers. 5.7 Non-agricultural activities. On average, beneficiaries’ expectations were only partially met, due to delays in the allocation of funds, insufficient training, and poor participation in some 44 phases of the subprojects’ implementation. However, participation in the design phase was high with significant ownership of selected subprojects. The performance of service providers was considered satisfactory by less than half of beneficiaries. Ex-Post Evaluation of Productive Investment Subprojects under PSDR (Cabinet RConseil, 2005) and Ex Post Evaluation of productive investment subprojects under PSDR (Savaivo, 2008) 5.8 The results of the 2005 and 2008 evaluations, summarized in Table 5.8.1 below, show that the project was increasingly on track to achieve its objectives, with overall satisfaction of the majority of beneficiaries. Table 5.8.1: Synthesis of 2005 subprojects evaluation Indicator Results as of 2005 Results as of 2008 Increase in income 79% of beneficiary households 75% of beneficiary households saw a 30% increase in income saw a 30% increase in income. The decrease compared to 2005 was explained by the high inflation rate (55%), and the occurrence of natural disasters (cyclones) during the 2004-2007 period. Development of a recovery 28% of producer organizations 61% of beneficiaries cost system have developed a system for established a cost recovery cost recovery system, with higher percentages for agricultural and non-agricultural activities (65% and 64% respectively), and for producer organizations headed by women. Profitability of investments IRR 21% IRR 33.5% Sustainability of subprojects 49% of producer organizations 55% of producer organizations activities are members of financing are members of financing institutions; 61% have savings, institutions; and 39% set up revolving funds Increase in productivity On average, 92% of 65% saw a satisfactory subprojects have experienced productivity an increase in production Adequate adoption of 87% of trained beneficiaries 42% adopted the new technology have adopted the new technologies promoted by the technologies project Satisfaction 51% of beneficiaries are 96% satisfied with the satisfied that the objectives subprojects’ results, 95% with have been achieved; 73% are the amount of financing satisfied of the technical allocated, 94% with training assistance provided; and 69% offered, 93% with the are satisfied that the monitoring carried out by the 45 subprojects respond to their technical assistance providers, needs and priorities and 92% with the management of funds within the producer organizations. Environmental compliance 72% of beneficiaries comply 68% of producer organizations with the Environmental Action comply with the Plans Environmental Action Plans Economic Evaluation of Productive Investment Subprojects under PSDR (Cabinet MIARA- MITA, 2012) 5.9 The study was based on a random sample of 205 subprojects financed between 2007 and 2008, targeting 1,260 beneficiaries in 16 regions. The study highlighted the following project strengths and weaknesses: Project strengths: • Positive economic and social impact that helped improve living conditions and increase income; • Response to the huge needs for support in the different regions of the country and especially in remote and isolated areas (12% of subprojects took place in remote areas); • Access to training and mentoring for producers; • Inclusion of gender aspects in the implementation of activities; • Promising sustainability of activities, as evidenced by high rate of adoption of new technologies, increased access to micro-finance institutions and saving capacity; Project weaknesses: • Subproject selection characterized by the promotion of a large range of activities resulting in dispersed efforts and inadequate/inefficient monitoring and supervision; • Weak performance of service providers who lack professionalism thus having a negative impact on quality of implementation; • Inadequate monitoring of subprojects after the design and formulation phase. 46 Annex 6. Stakeholder Workshop Report and Results 6.1 Two separate rounds of stakeholder consultations were conducted to obtain feedback, and distill lessons learned. The first round of discussions involved a workshop held in July 2012 with the key agencies and individuals with strong influence over the direction of agriculture and rural development strategies and policies. The second round of discussions, held in November 2012, involved a wide range of participants including representatives from 22 Producers Organizations (one each from the 22 Regions benefitted by PSDR), the Ministries of Agriculture, Fisheries and Livestock, the Regional Directorates for Rural Development, the Agriculture Service Centers (CSA), the private sector, and researsch institutions (FOFIFA). 6.2 Key feedback obtained during the July 2012 workshop include: (i) so as to ensure homogeneity, criteria for group formation need to include the level of education in addition to the degree of expertise and experience; (ii) regular internal meetings, establishment of a network of partners, collaboration between the project and the line Ministry are key for effective management and coordination; (iii) mainstreaming of environmental considerations is essential, regardless of project scale; (iv) beneficiaries’ contribution needs to be established based on poverty level rather than geographical location; and (v) the establishment of a permanent monitoring system of subproject is fundamental for follow up and learning. 6.3. During the November 2012 workshop, the beneficiaries acknowledged the important contribution made by PSDR to the improvement of their quality of life and were generally satisfied with the support received from the project. In particular, it was noted that the project, through the introduction of beneficiaries’ contribution, helped promoting a change of mentality towards entrepreneurship and thus increasing ownership of activities and investments financed. 6.4 Main achievements highlighted by participating stakeholders included: • Increase in production and income for agriculture and off-farm activities (e.g., handicraft); • Organizational strengthening, increased capacity to manage resources and projects (e.g., shared responsibility for maintenance and operation of infrastructure, improved procurement capacity); • Improvement of infrastructure; • Decrease in social conflicts for access to resources (i.e., water). 6.5 Main obstacles emphasized by participating stakeholders included: • Delay in obtaining financing and in procurement of material and equipiment, which delayed activities in the field (conflict with production calendar year); • Difficulties in fulfilling the beneficiaries’ contribution, which in some cases led to the inability to realize the infrastructure work, while in other cases farmers had to indebt themselves; • Poor quality of some infrastructure work and material; • Lack of technical capacity to implement subprojects, and difficulties purchasing certain materials/inputs (e.g. fertilizaers and pesticides); • Insufficient and/or inadequate training for use of equipment and for production in specific regional condition; weak link between research and technical assistance; • Difficulties in maintining the membership of community associations, unless there is a high degree of family connection; • Insufficient support for market access; 47 • Lack of appropiate storage facilities; • Heavy focus on agriculture production, with little support for fisheries and livestock subprojects; 6.6 Main concerns raised for sustainability of activities: • Other sources of financing; • Training/capacity building for identification of market opportunities and access; • Technical assistance and training for O&M; • Sensitization of beneficiaries to strengthen ownership of investments, rationale for contribution, and participation of all community members in O&M. 48 Annex 7. Summary of Borrower's ICR and/or Comments on Draft ICR Unofficial Translation Final Report Rural Development Support Project Republic of Madagascar A. Executive Summary 7.1 The preliminary results of the ex-post evaluation carried out during the project (over the period 2004-06) confirm that the subprojects reached the targeted recipients and had a considerable impact both on the level of productivity and income. The project had reached close to three-quarter of the country’s rural communes, half of which in isolated areas. 7.2 As of December 31, 2012, this trend is maintained as PSDR’s results show that 9,510 rural organizations (versus a target of 6200) and nearly 1.2 million people (i.e., approximately 10% of the rural population of Madagascar) in 240,000 rural households (versus a target of 150,000) directly benefited from the project. 7.3 Component A achievements are interesting, with results higher than the targets in the majority of cases and even well beyond in certain cases (1177 productive infrastructure subprojects delivered; 8032 agricultural subprojects financed and 301 non-agricultural subprojects financed). Some beneficiaries were able to continue their activities beyond PSDR and/or extend or diversify their activities. Some beneficiaries acquired negotiation skills in particular with their commercial partners. 7.4 With regard to Component A.4. (Support Services): (i) technical and organizational trainings were capitalized by the majority of the recipients; (ii) compliance with environmental action plans indicates a special attention to sustainable production; (iii) experiences under the initial phase allowed for the identification and filling of the gaps for environmental management: each Regional Executive Directorate (DER) now has an environmental specialist responsible for following up closely the implementation of subprojects environmental aspects. 7.5 With regard to Component B, thematic and applied research adequately responded to the farmers’ concerns and the analysis of their problems. More than 80% of subprojects beneficiaries adopted the innovating technologies brought within the framework of the thematic research support between 2001 and 2009 (during the initial financing); 26 research projects, financed by FCRA were implemented. 7.6 With regard to Component C, the UPDR contributed in a substantial way to the strengthening of sectors of national scale and to the strategic support for the elaboration of the rural development national plan (PNDR). In addition, the establishment and support to eleven GTDR during the initial financing and contributions for the establishment of twelve Regional Steering and Allocation Committees (CROA) and the strengthening of twenty one Agricultural Service centers (ASC) during the additional financing phase are substantial achievements that promoted the synergy of actors in rural development and the strengthening of the process of regionalization and decentralization. 49 7.7 With regard to Component D, the changes in strategies, approaches and work methodologies, and the restructuring made during the life of the project influenced management costs and project management. PSDR intervention is deemed relevant, effective and efficient for the period 2001-2012. 7.8 Partnership: • The “Partenaires Strategiques� have varying capabilities across regions and sectors and cover different situations; • Relationships with financial institutions have influenced the development vision of producer organizations and their activities could develop; • Despite problems encountered in developing a system for the recovery of funds to cover the operating expenses for the continuity of productive investment subprojects, a little less than half of the producer organizations have been able to achieve financial savings except members with savings in kind (source: surveys EEDR MAMOKATRA, December 2012). 7.9 Impact: • The project had a significant impact on increasing the income improving the living conditions of farmers who benefited from subprojects; 80% of project beneficiaries have seen their incomes increased by at least 30%; • Thanks to the project, the majority of beneficiaries experienced an increase in the volume of production, which will have an impact on improving food security; • It can be argued that 88% of subprojects remain functional one year after reception (source: Economic Assessment Results MIARAMITA confirmed by the PSDR Results Framework, December 2012); • The project had a significant impact on increasing income of subprojects beneficiaries. Indeed, following the economic assessment by the Cabinet MIARAMITA, 80% of project beneficiaries have seen their incomes increased by at least 30%. 7.10 Sustainability: • The appropriation of subprojects by the beneficiaries is the result of the use of participatory planning techniques, the involvement of vulnerable producers and women (34%) in the process of decision making. These ownership elements reflect the interest to sustain subproject activities. • Mechanisms for subprojects ownership on the part of beneficiaries are generally operational (organizational and functional structure); • Environmental protection is an achievement of the Project. The Environmental Action Plans (EAP) are complied with by 75% of beneficiaries according to the Results Framework of December 2012. The survey conducted as part of this study confirmed that more than 88% of producer organizations are complying with the terms of the EAP. 7.11 We can conclude and confirm that the overall objective of the project has been achieved as 80% of targeted rural households feel that their incomes increased by at least 30%. Given the good performance of PSDR and the many complaints collected during the field surveys, best practices from the PSDR experience deserve to be capitalized and elaborated in the framework of commercial agriculture where each stakeholder will continue to benefit and improve their income. B. Borrower’s Comments on Draft ICR The Borrower did not send any comments following reception of the Bank’s draft ICR. See Main Text, footnote 20. 50 51 Annex 8. Comments of Cofinanciers and Other Partners/Stakeholders N/A 52 Annex 9. List of Supporting Documents Project Appraisal Document, Report No. 21516-MAG, May 16, 2001 Project Paper for the Additional Financing, Report No. 44953-MG, September 23, 2008 Credit and Financing Agreements Aide Memoires Financial Management Supervision Reports Procurement Reports Implementation Supervision Reports (ISR) Restructuring Papers Client Completion Report: “Evaluation du Project de Soutien Au Developpement Rural (PSDR) – Achevement du Project�, EEDR MAMOKATRA S.A., December 2012 Evaluation externe consolidée du Project de Soutien au Developpement Rural, Cabinet ADAPT, July 2004 Rapport definitif relatif a l’evaluation par les beneficiares du Project PSDR, Cabinet MIARA-MITA, 2005 Evaluation ex-post de sous-projets d’investissements productif du Projet de Soutien du Developpement Rural, Cabinet RConseil, October 2005 Evaluation ex-post de sous projets d’investissement prductif du Projet de Soutien au Developpement Rural (PSDR), Savaivo, August 2008 Evaluation économique des sous projets d’investssiment productif du Projet de Soutien au Developpement Rural (PSDR) – Janvier 2007 au Juin 2008, Cabinet MIARA-MITA, May 2012 Evaluation d’impacts des sous projets d’invesitissement productif duProject de Soutien au Developpement Rural (PSDR), Savaivo, December 2012 53 MAP 54