Completion Report. Project Number: Loan Number: 2449-INO(SF) October Indonesia: Rural Infrastructure Support to the PNPM Mandiri Project

Transcription

1 Completion Report Project Number: Loan Number: 2449-INO(SF) October 2012 Indonesia: Rural Infrastructure Support to the PNPM Mandiri Project

2

3 CURRENCY EQUIVALENTS Currency Unit rupiah (Rp) At Appraisal At Project Completion (15 August 2008) (9 September 2011) Rp1.00 = $ $ $1.00 = Rp9,180 Rp8,571 ABBREVIATIONS ADB Asian Development Bank BAPPENAS Badan Perencanaan dan Pembangunan Nasional (National Development Planning Agency) CDD community-driven development CIO community implementing organization CSP country strategy and program DGHS Directorate General of Human Settlements DPIU district project implementation unit km kilometer KPP kelompok pengguna dan pemelihara (community group responsible for the operation and maintenance of community infrastructure) MDG Millennium Development Goal MPW Ministry of Public Works MTPRP midterm poverty reduction plan O&M operation and maintenance PCMU project coordination and monitoring unit PIU project implementation unit PNPM Mandiri Program Nasional Pemberdayaan Masyarakat Mandiri (National Program for Community Empowerment) PPIU provincial project implementation unit RPJMN rencana pembangunan jangka menengah nasional (national medium-term development plan) RRP report and recommendation of the President (i) (ii) NOTES The fiscal year (FY) of the government and its agencies ends on 31 December. FY before a calendar year denotes the year in which the fiscal year ends, e.g., FY2009 ends on 31 December In this report, $ refers to US dollars.

4 Vice-President S. P. Groff, Operations 2 Director General K. Senga, Southeast Asia Department (SERD) Director J. Lindborg, Indonesia Resident Mission (IRM), SERD Team leader Team members S. Hasanah, Senior Project Officer, IRM, SERD Wolfgang Kubitzki, Principal Portfolio Management Specialist, IRM, SERD Suzana, Senior Operations Assistant, IRM, SERD In preparing any country program or strategy, financing any project, or by making any designation of or reference to a particular territory or geographic area in this document, the Asian Development Bank does not intend to make any judgments as to the legal or other status of any territory or area.

5 CONTENTS Page BASIC DATA MAP ii vi I. PROJECT DESCRIPTION 1 II. EVALUATION OF DESIGN AND IMPLEMENTATION 1 A. Relevance of Design and Formulation 1 B. Project Outputs 2 C. Project Cost 7 D. Disbursements 8 E. Project Schedule 8 F. Implementation Arrangements 8 G. Conditions and Covenants 9 H. Consultant Recruitment and Procurement 9 I. Performance of Consultants, Contractors, and Suppliers 10 J. Performance of the Borrower and the Executing Agency 10 K. Performance of the Asian Development Bank 10 III. EVALUATION OF PERFORMANCE 11 A. Relevance 11 B. Effectiveness in Achieving Outcome 12 C. Efficiency in Achieving Outcome and Output 12 D. Preliminary Assessment of Sustainability 13 E. Impact 13 IV. OVERALL ASSESSMENT AND RECOMMENDATIONS 14 A. Overall Assessment 14 B. Lessons 14 C. Recommendations 15 APPENDIXES 1 Project Logical Framework 16 2 Participating Provinces and Districts 23 3 Gender Action Plan 24 4 Types of Built Infrastructure 27 5 Project Cost and Financing Plans 28 6 Disbursements 30 7 Flow of Funds 32 8 Project Implementation Schedule Project Implementation Arrangements Compliance with Loan Covenants Economic and Financial Analysis 44

6 BASIC DATA A. Loan Identification 1. Country 2. Loan Number 3. Project Title 4. Borrower 5. Executing Agency 6. Amount of Loan 7. Project Completion Report Number B. Loan Data 1. Appraisal Date Started Date Completed 2. Loan Negotiations Date Started Date Completed 3. Date of Board Approval 4. Date of Loan Agreement 5. Date of Loan Effectiveness In Loan Agreement Actual Number of Extensions 6. Closing Date In Loan Agreement Actual Number of Extensions 7. Terms of Loan Interest Rate Maturity (number of years) Grace Period (number of years) Indonesia 2449-INO(SF) Rural Infrastructure Support to the PNPM Mandiri Project Republic of Indonesia Directorate General of Human Settlements, Ministry of Public Works Original loan amount: $50.0 million (SDR31.85 million) Actual loan amount: $48.1 million (SDR31.34 million) PCR: INO June June August August September November January January 2009 None 30 June September 2011 None 1% during the grace period; 1.5% thereafter 32 years 8 years 8. Disbursements a. Dates Initial Disbursement 16 March 2009 Effective Date 28 January 2009 Final Disbursement 09 September 2011 Original Closing Date 30 June 2011 Time Interval 30 months Time Interval 29 Months

7 iii b. Loan 2449-INO(SF): Amount ($ 000) Category or Subloan Original Allocation Last Revised Allocation Amount Canceled Net Amount Available Amount Disbursed Undisbursed Balance 01 Civil Works 45, , , , Consulting 2, , , , Services 03 Audit, Surveys and Studies 04 Interest Charge Unallocated Total 50, , , , Notes: (i) The reduction in the total revised allocation was due to fluctuations of exchange rates. (ii) Figures may not add up to the totals given because of rounding. c. Loan 2449-INO(SF): Amount (SDR 000) Category or Subloan Original Allocation Last Revised Allocation Amount Canceled Net Amount Available Amount Disbursed Undisbursed Balance 01 Civil Works 29, , , , Consulting 1, , , , Services 03 Audit, Surveys, and Studies 04 Interest Charge 05 Unallocated Total 31, , , , a The undisbursed balance of SDR0.51 million (equivalent to $0.8 million) was canceled when the loan account was closed on 9 September Local Costs (Financed) - Amount ($) NA - Percentage of Local Costs NA - Percentage of Total Cost NA C. Project Data 1. Project Cost ($ million) Cost Appraisal Estimate Actual Foreign Exchange Cost NA NA Local Currency Cost NA NA Total 2. Financing Plan ($ million) Cost Appraisal Estimate Actual Implementation Costs Borrower-Financed ADB-Financed Other External Financing Total

8 iv Cost Appraisal Estimate Actual IDC Costs Borrower-Financed ADB-Financed Other External Financing Total ADB = Asian Development Bank, IDC = interest during construction. 3. Cost Breakdown, by Project Component ($ million) Component Appraisal Estimate Actual A. Community Facilitation and Mobilization B. Rural Infrastructure Improvement C. Audits, Studies, Surveys D. Consulting Services E. Project Management F. Unallocated/Contingencies G. Interest During Implementation Total Project Schedule Item Appraisal Estimate Actual Date of Contract with Consultants National Project Management Consultant Apr May 2009 Regional Project Management Consultant (RPMC) Apr May 2009 South Sumatra RPMC Lampung Apr May 2009 RPMC Jambi and Riau Apr May 2009 Completion of Engineering Designs Jul 2009 Jul 2009 Civil Works Contract Date of Award Sep 2009 Aug 2009 Completion of Work Sep 2010 Dec 2010 Equipment and Supplies Dates First Procurement Nov 2009 Oct 2009 Last Procurement Mar 2010 Jun 2010 Completion of Equipment Installation Sep 2010 Nov 2010 Start of Operations Completion of Tests and Commissioning Oct 2010 Nov 2010 Beginning of Start-Up Nov 2010 Dec 2010 Operation and Maintenance (O&M) of Built Facilities O&M Training for Communities Jun 2010 Jun 2010 Preparation of O&M Plans Jun 2010 Jun 2010 Implementation of O&M Plans Aug 2010 Jun 2011 Aug 2010 Jun 2011 Monitoring and Evaluation Monitoring of project preparation and Oct 2008 Dec 2010 Oct 2008 Dec 2010 implementation Report on physical and financial progress Mar 2009 Dec 2010 Mar 2009 Dec 2010

9 v Item Appraisal Estimate Actual Audit accounts and financial statements Conduct of spot audits and complaint investigation Assessment of project impact Jun Sep 2009 and Jun Sep 2010 Jun Dec 2009 and Jun Dec 2010 Dec 2009 and Dec 2010 Jun 2009 and Jun 2010 Jun Dec 2009 and Jun Dec 2010 Dec 2009 and Dec Project Performance Report Ratings Rating Implementation Period Development Objectives Implementation Progress 28 Jan June 2011 Satisfactory Satisfactory D. Data on Asian Development Bank Missions Name of Mission Date No. of Persons No. of Person-Days Specialization of Members a Fact-finding 19 Feb 14 Mar a, b, c, d, e Appraisal 9 26 Jun a, d, f, g Inception 18 May 1 Jun b, d Review 1 23 Nov 7 Dec d, h Review 2 22 Mar 22 Apr d, h Review Nov d, h Review 4 6 Jun 18 Jul d, h Project completion review 16 Mar 9 May d, h, i a a = environmental economist, b = social sector specialist, c = environmental engineer, d = senior project officer, e = project officer, f = social development specialist, g = counsel, h = principal portfolio management specialist, i = senior operation assistant.

10

11 I. PROJECT DESCRIPTION 1. The Rural Infrastructure Support to the PNPM Mandiri Project was an integral part of Indonesia s core poverty reduction program, the National Program for Community Empowerment (PNPM Mandiri). The largest community empowerment program in the world, PNPM Mandiri was launched in 2006 to reduce poverty through community-driven development (CDD) and direct support for improvements in basic infrastructure and essential social services for poor rural and urban communities. All members of the community are encouraged to be involved in the development process. 2. Although the government has succeeded in reducing the incidence of poverty, from around 25% in 1998 to 15.4% in 2008, Indonesia still faces significant challenges before it can achieve the Millennium Development Goal (MDG) target of 7.5% poverty incidence by At the time the project was designed, almost 50% of the population was living on less than $2 a day. 1 About two-thirds of Indonesia s poor live in the rural areas. Rural poverty has many causes: insufficient access to health, education, safe water, and sanitation services; lack of economic opportunities; poor rural connectivity. Areas with difficult geography are isolated by a shortage of infrastructure and inadequate market linkages. Nearly one out of five rural villages is inaccessible even to four-wheel-drive vehicles for part of the year. Disadvantaged rural people are especially vulnerable to external shocks and could easily fall back into poverty in the event of sickness, a natural disaster, harvest failure, or rising food prices. 3. In recognition of the development impact of improved infrastructure on poverty reduction, the government s medium-term development plan (RPJMN) for emphasized the need to increase investments in infrastructure in general and rural infrastructure in particular. The country strategy and program (CSP) of the Asian Development Bank (ADB) for Indonesia for had a similar focus The project, as a vital part of PNPM Mandiri, was meant to improve the socioeconomic conditions of rural communities in the project area by giving the poor and near poor in these communities better access to basic rural infrastructure. Its output comprised the following: (i) an increase in the capacity of the communities to engage in planning and development, and (ii) improvements in village services and infrastructure with the help of community development grants. (The project design monitoring framework, which defines the project impact, outcome, output, and performance indicators and targets, is in Appendix 1.) Block grants were provided to poor communities in 1,724 villages in 37 districts in the provinces of Lampung, Jambi, Riau, and South Sumatra. (The list of participating districts is in Appendix 2.) About 2 million poor people, or about 400,000 households, benefited from the project. II. EVALUATION OF DESIGN AND IMPLEMENTATION A. Relevance of Design and Formulation 5. The project was formulated according to RPJMN , which gave strategic direction to the government s poverty reduction efforts. The RPJMN reflected the emphasis placed on rural infrastructure investments in the National Poverty Reduction Strategy as an 1 ADB Report and Recommendation of the President to the Board of Directors on the Proposed Loan to the Republic of Indonesia for the Rural Infrastructure Support to the PNPM Mandiri Project. Manila. 2 ADB Country Strategy and Program: Indonesia, Manila.

12 2 important means of reducing poverty and regional disparities by improving access to services and markets and creating jobs, and thereby achieving the MDGs. 6. The project design was in line with ADB s strategy for Indonesia as set out in the CSP for , which also drew attention to the significance of investments in rural infrastructure in reducing poverty, lessening regional disparities in service delivery, and accelerating the achievement of the MDGs. 7. During project formulation, ADB cooperated closely with government counterparts from the Ministry of Public Works (MPW), the National Development Planning Agency (BAPPENAS), the Ministry of Finance, and the Coordinating Ministry for People s Welfare in integrating the project design and implementation arrangements with PNPM Mandiri. The project supported the implementation of the PNPM Mandiri core program. 3 In accordance with the PNPM Mandiri design, the project adopted the CDD approach, handing over to communities the power to decide on the use of block grants. Communities were authorized to design and implement infrastructure development projects according to their development priorities. Community-owned projects have been shown to perform better than projects directed and administered by government agencies, and to have a higher rate of participation among community members, especially among the poor and women. The project design drew on lessons from the other successful CDD projects implemented by ADB and other development partners in Indonesia. 8. Adopting the design, funding modality, and implementation structure of PNPM Mandiri rendered the project fully consistent with the government s development program and strategies. It ensured strong ownership of the project and greatly improved its prospects of achieving the common goals of improved service delivery and poverty reduction, and thereby meeting the MDG targets. At project completion in 2011, a follow-on project 4 had just been approved. This suggests that the government considered the project relevant, both at appraisal and at completion. 9. The project covered 1,724 villages in 37 districts in the provinces of Lampung, Jambi, Riau, and South Sumatra. Rural poverty incidence in these four provinces was higher than the national average. The selection of project villages was based on a list of less-developed subdistricts (kecamatan) 5 issued by the Coordinating Ministry for People s Welfare (MENKO KESRA). The final list of villages and subdistricts approved by the MPW was contained in Decree No. 274/KPTS/M/2009 of 9 March The project was implemented as formulated during appraisal. No change in the project design was required during implementation. B. Project Outputs 1. Community Facilitation and Mobilization 10. Assistance was provided to communities to further empower them and strengthen their capacity to identify development priorities and to plan and implement specific investment 3 Centered on promoting community empowerment, establishing and strengthening community implementing organizations, building the capacity of communities to plan and implement projects according to their needs, and providing investment grants for projects. PNPM Mandiri s other tier, its support program, is focused on the delivery of specific services to sectors. 4 ADB Report and Recommendation of the President to the Board of Directors on the Proposed Loan to the Republic of Indonesia for the Rural Infrastructure Support to the PNPM Mandiri II Project. Manila. Loan 2575-INO ($84.24 million) was approved on 12 November 2009 and declared effective on 1 February A village is part of a subdistrict. The village selection criteria are in Appendix 2.

13 3 projects efficiently and transparently. Community facilitators supported the following activities: (i) awareness campaigns to familiarize community members with PNPM Mandiri; (ii) community facilitation, including poverty mapping in the villages, identification of problems and needs, evaluation of community implementation capacity, and development of mechanisms for planning and decision making; (iii) establishment and capacity building of community implementing organizations (CIOs); (iv) preparation of village midterm poverty reduction plans (MTPRPs) and related annual investment plans to be funded through block grants; (v) provision of technical guidance to communities during the implementation of activities included in the village MTPRPs; and (vi) formulation and implementation of operation and maintenance (O&M) plans to ensure the sustainability of the completed facilities. 11. At project completion, block grants had been provided to 1,724 villages, exceeding the appraisal target of 1,650 villages, 1,724 CIOs had been established, and 1,724 MTPRPs and 1,724 annual investment plans had been prepared by the communities To assist the communities in the project villages, 1,179 community facilitators were contracted and mobilized. 7 Women were to compose about one-third of the total. The facilitators were divided into teams, with each team having five facilitators (three responsible for social development matters and two for technical subjects). Each team was assigned to assist up to seven villages. The recruitment of community facilitators was coordinated by the provincial project implementation units (PPIUs). 13. The performance indicators stated in the report and recommendation of the President (RRP; footnote 1) for this output were as follows: (i) at least 50% of the project communities would acquire the capacity to plan CDD projects by 2010; (ii) the participation rate for women and the poorest community members in planning and decision-making meetings would reach at least 40% by 2010; and (iii) the district public works services would consider community feedback in designing future projects and programs. 14. After development priorities were identified by the villages, the village MTPRPs were presented at kecamatan forums, where all villages within the kecamatan were represented. The forums were expected to prioritize village development needs for implementation under block grant assistance, taking into account overall kecamatan development priorities (e.g., development of intervillage roads and water supply schemes) and efficiency considerations to avoid duplication of activities. Discussions with district public agencies indicated that many of the village MTPRPs developed with project assistance were discussed and considered in the annual district medium-term development meeting (musrembang). The district public works agencies confirmed that many community investment plans as specified in their MTPRPs have been included in the public work agencies annual plans and financed under the local government budgets. 15. The project completion and earlier review missions found wide variation between villages in the level of community participation. In some villages, the empowerment process was extremely successful. Entire communities were involved in planning and decision making. But in other communities, community participation in poverty mapping, investment prioritization, and proposal preparation was a continuing challenge. 6 Appendix 1 (Design and Monitoring Framework, DMF) of the RRP (footnote 1) - performance targets for the impact. 7 Community facilitators were financed from government funds.

14 4 16. In some villages, the socialization activities seemed rushed, such that many community members were not fully involved in planning and decision making. Sound poverty mapping exercises and joint identification of problems and needs did not take place as planned; instead the village head made the decisions. Many CIO members in these villages did not fully understand the purpose and content of the MTPRPs, and the plans were not considered in medium-term development planning for the district. 17. The participation of women also varied in degree. In most villages, women, including heads of households, were actively involved in all stages of the project. But in some villages, the majority of women had no part in decision making and in the execution of civil works. Feedback from these women suggests that they hesitated to participate because they were busy with family matters, the meeting halls were too far from their houses, or the meetings were usually held in the evening. In these villages, women s participation was limited to providing food and refreshments during meetings and construction works. At project completion, about 32,330 women (39% of the 83,230 women included in the project) had participated in all stages of project implementation. The project target of a 40% participation rate for women was considered to be achieved. Of the 1,179 community facilitators, 411 (35%; better than the target of 30%) were female However, the membership of women in the CIOs and user groups was far below the target. According to feedback obtained during field visits, women were hesitant to join the CIOs mainly because of the volume of their domestic duties. In villages where the participation of women was high, the use of infrastructure and the number who benefited from such use were also usually high. The membership of women in CIOs helped the latter communicate better with women and induced more women to participate. CIOs and user groups where women were actively involved were also better managed, particularly with respect to financial reporting. The targets under the project gender action plan are compared with the achievements in Appendix Effective community empowerment requires experienced and qualified facilitators, and good guidance from consultants and district administrators. While the majority of the community facilitators employed under the project performed well, the project completion and earlier review missions saw the need for more support and training, chiefly for those facilitators who lacked experience. The regular meetings at which community facilitators from different districts were able to share experiences and learn from successful community empowerment approaches proved very useful. In retrospect, more efficient guidance and support from the district and provincial consultant teams would have helped the community facilitators deal with the challenges in the community empowerment process. 20. Overall, good governance and accountability arrangements were established in the project communities. In some villages, however, the CIOs did not follow the agreed procurement procedures, and did not comply with the project guidelines regarding the sharing of information about fund utilization with the general public. Community members in these villages were unaware of the specific use of the block grants, and public information boards with information about bidding procedures, detailed expenditures, employment arrangements, wages, etc., were not in place. 8 Targets for women participation are defined in the DMF - Appendix 1 of the RRP (footnote 1).

15 5 2. Rural Infrastructure Improvement 21. Block grants were provided to villages according to the infrastructure priorities identified in their MTPRPs and at kecamatan community forums. The infrastructure improvement and expansion priorities involved the rehabilitation or construction of village roads and pathways, bridges and culverts, small piers, village irrigation schemes, water supply and sanitation facilities, and drainage works. The MTPRPs should also include arrangements for the operation and maintenance of the built infrastructure. 22. The project provided the first cycle of block grant assistance to 1,724 project villages. Each participating village received a block grant of Rp250 million (equivalent to about $27,500) to finance the rural infrastructure improvements. The block grants were released in three tranches according to the following procedure: (i) with the assistance of a community facilitator, the CIO prepared a community/village 9 investment plan; (ii) once the plan was approved by the district project implementation unit (DPIU), the district project manager 10 and the CIO coordinator signed a community contract; (iii) the project manager submitted a request for the transfer of funds to the district treasury office; (iv) the district treasury office transferred 40% of the block grant directly into the community s bank account as an advance payment; (v) after the civil works reached 36% completion (as certified by the district project manager), the CIO requested the second tranche of the block grant (40%); (vi) after the civil works reached 72% completion, the CIO requested the third and last tranche of the block grant (20%); (vii) when the civil works were fully completed, the CIO met with the community to report to its members and to reach agreement on the preparation of the final report; and (viii) after the final report was approved by the district manager, the latter formally handed over the built facility to the CIO, representing the community members. 23. The project approach, which conformed to the PNPM Mandiri guidelines, had the following features: (i) the CIO representative committee, including those members representing women and minority groups, should endorse the village investment plan; (ii) the activities should provide proportional benefits and employment opportunities to women and minority groups; (iii) the improvements in village infrastructure should lead to improvements in the livelihood of the poor or near poor; (iv) the investments should be immediately useful to the poor and near-poor; (v) the investments should be economically viable and financially sustainable; (vi) the investments should have no adverse environmental, social, or economic impact; (vii) alternative designs should be evaluated to establish the cost-effectiveness of the proposed design; (viii) the work should use local labor and materials and, to the extent possible, employ technology that is simple or otherwise appropriate to local conditions; and (ix) proper O&M plans should be prepared by the CIO, and O&M of project facilities should be fully implemented and financed by the communities. The district governments were to provide certain O&M activities that were beyond the technical capacity of the communities. 24. While most of these conditions were met, many villages did not assess the economic viability of investment projects, and some villages did not include effective O&M arrangements in their investment proposals. 25. The block grants were mostly used to finance the construction or upgrading of transport facilities (rural roads, bridges). About 1,630 kilometers (km) of rural roads, 5,483 bridges, 52 boat landings, more than 142 km of drainage canals, 26 km of irrigation cannals, 355 domestic 9 The terms community and village are used interchangeably in this report, as they refer to the same entity. 10 The district project managers were staff of the district department of public works.

16 6 water reservoirs, 21 shallow wells, 54 deep wells, 50 km of water supply pipelines, and 1,820 communal sanitation facilities were constructed, rehabilitated and improved under the project. The detailed distribution of types of infrastructure improvement in the four provinces is in Appendix About $51.3 million $45.3 million from loan proceeds, $0.2 million from government counterpart funds, and $5.8 million from beneficiaries 11 were used for the upgrading or construction of basic village infrastructure. The following chart shows the distribution of block grants by type of infrastructure. Distribution of Block Grants, by Type of Infrastructure Distribution of Investments $ (million) Water Supply and Sanitation Facilities, $5.48 Transport Facilities, $38.85 Irrigation Facilities, 0$.91 Drainage and Land Protection, $0.46 million Type of Infrastructure Sources: Ministry of Public Works Project Completion Report: Rural Infrastructure Support to the PNPM Mandiri Project. Jakarta and Asian Development Bank estimates. 27. According to the chart, the block grants were mostly used to finance the construction of transport facilities (roads and bridges, and water transport facilities), and the construction of water supply and sanitation facilities accounted for the second-largest investment. Observations and discussions with community members during the project completion and earlier review missions suggest that the built facilities were consistent with the infrastructure priorities set in the MTPRPs. During the discussions, many community members also suggested that the size of block grants needed to be increased. The community members stated that in some cases they had to settle for lower priority investments as the higher priority investments could not be covered by the existing block grant size. In other cases, some community members do not benefit directly from the project investments because the block grants are not enough to cover the whole village. 28. Community members made voluntary contributions to their village investment projects. The contributions involved donating land for the required facilities, 12 transporting materials from 11 In cash and inkind contributions. 12 The project required limited land acquisition (land donation) but no resettlement of people or reallocation of houses. Findings during the project completion and earlier review missions and data from reports prepared by the Directorate General of Human Settlements of the MPW indicate that the land acquisition had no significant impact on productive land and other productive assets. Safeguards related to land acquisition were built into the community decision-making process.

17 7 the main street to the project sites, sourcing locally available materials, clearing land, and providing food and refreshments during meetings and construction work. The community contributions constituted up to $5.8 million, or about 9% of the total project cost. 29. The majority of the community members (more than 75%) were employed as workers during construction. About Rp83 million per village, or 33% of the total grant, went to wages for village workers. On average, community members received Rp40,000 per day for construction work, thereby gaining additional income and increased motivation to participate in civil works. The civil works financed through the block grants generated about 162,600 person-months of short-term employment. 30. Observations made during the project completion and earlier review missions suggest that the quality of civil works, specifically those related to roads, pathways, bridges, and drainage channels, was generally satisfactory. But in some villages, the O&M arrangements for water and sanitation facilities in particular were weak. According to discussions with community members, CIOs, district and provincial administrations, consultants, and staff of the Directorate General of Human Settlements (DGHS) of the MPW, the establishment of user groups (KPPs) for the O&M of the built facilities was often only a formality preceding the transfer of block grants to the CIOs. In these villages, the KPPs and the O&M arrangements are not functioning as expected. While stating that training for CIOs and KPPs has been conducted several times, the DGHS agrees that more support (training and other capacity-building activities) will have to be provided to further strengthen CIOs and KPPs, given the above findings. The training is intended not only to equip the community members to operate, maintain and manage the built facilities but also to raise awareness of the need to implement the proper O&M arrangements. C. Project Cost 31. At appraisal, the project cost was estimated at $62.5 million, comprising $7.7 million (12%) for output 1 (community facilitation and mobilization), $51.8 million (83%) for output 2 (rural infrastructure improvements), and $3.0 million (5%) for financing charges during construction, project management, and other expenses (audits, surveys, taxes). ADB would contribute 80% of the project cost by providing a loan of SDR million (about $50.0 million) from the Asian Development Fund. 32. In 2010, the appreciation of the rupiah against major currencies, including the US dollar, increased the overall contract amounts (in US dollar equivalents) since all contracts were expressed in rupiah, and the original allocation of loan proceeds, particularly for the civil works and consulting services categories, was no longer sufficient to cover the contract amounts. The shortage of funds for civil works and consulting services was estimated at SDR1.96 million ($3.0 million). To cover this shortage, the government requested a reallocation of funds. Unallocated loan proceeds and funds allocated for audits, surveys, and studies, as well as undisbursed parts of the interest charge category, were reallocated to the civil works and consulting services categories. The Ministry of Finance also requested a reduction in ADB s share of financing for the civil works category from the original 100% to 95% for civil works contracts where disbursements had not yet been made. The government financed the remaining 5% under this new arrangement. 33. At project completion, the total project cost had been reduced to $60.40 million, comprising $6.9 million (11%) for output 1, $51.3 million (85%) for output 2, and $2.2 million (4%) for financing charges during construction, project management and other expenses. The ADB portion of the project cost had been reduced from $50.0 million to $48.1 million. The

18 8 reduction in project cost and ADB s contribution was due to fluctuations in the exchange rate and the cancellation of about $0.8 million of the loan proceeds under the civil works and consulting services categories. The 0.3% reduction in ADB s share of the overall project cost, to 79.7%, was due to the change in ADB financing for the civil works category. Appendix 5 shows the project cost details and financing plans at appraisal and at completion. D. Disbursements 34. The loan proceeds were disbursed according to ADB s Loan Disbursement Handbook (2007, as amended from time to time). The funds flow mechanism used in the Rural Infrastructure Support Project was adopted, in view of that project s successful completion and its quick disbursement of loan proceeds. 13 Its main feature was a simple funds flow arrangement with direct transfer from the project imprest account to the CIO account. 35. In total, $48.1 million (SDR31.3 million) of the loan proceeds, or about 98.4% of the total loan amount of SDR million, was disbursed. The RRP did not include a disbursement schedule. An imprest account was opened at Bank Indonesia to facilitate the disbursement of loan proceeds. 36. Communities implemented all the civil works contracts (averaging about $27,700 per contract). The imprest account was established, managed, replenished, and liquidated in accordance with ADB s Loan Disbursement Handbook. The DGHS managed the imprest account efficiently, with no significant issues during project implementation. Appendix 6 gives the disbursement details, while Appendix 7 contains an illustration of the fund-channeling mechanism. E. Project Schedule 37. The project was implemented according to the project schedule in the RRP. Soon after the project was declared effective in January 2009, the DGHS started recruiting, training, and mobilizing community facilitators. The implementation of civil works started in December 2009 and was completed in late Project management consultants were mobilized in late 2009 and early The national project implementation unit, as well as the PPIUs and DPIUs, were established before project effectiveness. Socialization program works, CIO formation, and disbursement of advance payments to the majority of the CIOs (1,500 out of 1,724) were completed in the first year of project implementation (2009). The implementation stayed on schedule because of the considerable work done beforehand. Appendix 8 compares the original schedule with actual implementation. F. Implementation Arrangements 38. The implementation arrangements outlined in the RRP were followed. The DGHS, under the MPW, was the executing agency. The project was implemented over a period of 2 years as part of PNPM Mandiri and used the program s implementation mechanisms. The national project coordination and monitoring unit (PCMU) managed, coordinated, and monitored the project activities; the existing PPIUs and DPIUs provided social and technical facilitation to the communities in their infrastructure rehabilitation and improvement activities; and the PNPM Mandiri oversight body supplied policy directives, guidance, monitoring, and managerial 13 ADB Report and Recommendation of the President to the Board of Directors: Proposed Loan to the Republic of Indonesia for the Rural Infrastructure Support Project. Manila.

19 9 oversight. The project implementation arrangements were considered appropriate for efficient project implementation. No change in the implementation arrangements was required during project implementation. Appendix 9 gives details of the project implementation arrangements. 39. Facilitators were selected by the PPIUs to support the CIOs in physical planning, budgeting and procurement, the maintenance of village bank accounts, inventory control in villages, and the recording of work progress and expenditures. 40. To carry out the civil works, the district project manager (satker) entered into a contract with the CIO. The contract specified the scope of the work, the materials to be purchased, and the payments to be made directly to the CIO from the national treasury. The district satker then authorized a series of payments to the CIO as provided in the contract, with each payment being conditional on CIO performance. The implementation arrangements were carefully designed and successfully executed. G. Conditions and Covenants 41. Thirty-seven loan covenants were fully complied with; two were partly complied with. The partial compliance reflected the absence of O&M plans in about one-third of the villages and the failure to post financial information in public in all villages. The loan covenants and their compliance status are listed in Appendix 10. H. Consultant Recruitment and Procurement 42. National consultants were hired for 1,080 person-months (against 960 person-months estimated at appraisal). Facilitators were also hired, for 1,179 person-months, 14 with government funds. All facilitators were mobilized in early 2009, while the national consultants were mobilized in May Civil works financed through block grants were contracted out to CIOs in compliance with ADB s Procurement Guidelines (2007). The CIOs procured materials and arranged the rental of equipment. Most of the participating villages followed the regulation governing the purchase of goods and services by choosing the supplier that offered the lowest price from among three or more competing suppliers. In a few instances, however, the CIOs obtained only one quotation, arguing that the prices were within DGHS standards or that only one supplier was available nearby. 44. The consultants were recruited according to ADB s Guidelines on the Use of Consultants (2007) and within a reasonable time (no significant delays). Civil works for rural infrastructure rehabilitation and expansion were contracted out to the communities on the basis of agreements signed by CIOs and DPIUs, and according to agreed procedures for community participation set out in the procurement plan. Community contracts followed the standard format of community contracts under PNPM Mandiri, and included evidence of community facilitation, the design of village infrastructure facilities to be improved under the contracts, and community O&M plans for the facilities. The first two community contracts in each participating province were submitted to ADB for prior approval. 14 Every group of five villages was served by two facilitators, each one being employed for an average of 5 person-months.

20 10 I. Performance of Consultants, Contractors, and Suppliers 45. In general, the performance of consultants, facilitators, and suppliers was satisfactory. However, the substandard quality of some facilities built under the project and the lack of attention to the maintenance of more complex infrastructure investments (such as water supply and sanitation facilities) indicate that consultants and technical facilitators provided less-than-satisfactory guidance and technical supervision during project implementation. Many of the facilitators were young and inexperienced recent graduates from university. Community feedback pointed to the benefits to be gained from a more robust process of selection and additional training and support for the community facilitators, most especially the inexperienced ones. Regular meetings at which community facilitators from different districts could share experiences and learn from successful community empowerment approaches would have been useful. The role and performance of consulting teams in assisting the facilitators and strengthening the community empowerment process should also be reviewed and improved, as should communication between consulting teams and facilitators. J. Performance of the Borrower and the Executing Agency 46. The performance of the government (borrower) and the DGHS (executing agency) was satisfactory. Both demonstrated strong commitment to the success of the project. Counterpart funds were provided on time and in the amounts required for the project. Rapid progress was possible as considerable work had been completed before implementation. The decrees and other legal instruments that were needed to launch the project were endorsed on time by all participating provinces and districts. The Minister of Public Works appointed by decree the national, provincial, and district project managers, and the provincial governors and heads of local governments formed management teams (coordination teams, PPIUs, and DPIUs). 15 Village heads legalized the authorized CIOs and confirmed the appointment of the chairman, secretary, treasurer, and ordinary members of CIO executive committees. The kecamatan heads registered the CIOs. No significant issues were raised during the process and the CIOs were therefore able to open bank accounts and receive the community block grants without difficulty. K. Performance of the Asian Development Bank 47. ADB s performance was satisfactory. During the 2-year implementation period, ADB fielded five project administration missions (one inception mission and four review missions) for a total of about 65 staff person-days, or about 33 staff person-days per year. The mission teams mainly comprised project implementation specialists and project officers. The frequency and number of staff-person days per mission appeared sufficient for effective supervision. ADB assisted the DGHS in the detailed calculation of the reallocated loan proceeds. With regard to consultant management, ADB assisted the DGHS in evaluating the consultants performance and advised on remedial actions, such as efficient consultant management, replacement of nonperforming personnel, and scrapping of unnecessary positions. All of the recommended actions were followed in full by the DGHS. As a result, consultant management by the DGHS improved. 16 ADB also provided the DGHS with access to its consultant database to enable it to find suitable replacements for the nonperforming consultants. 15 These teams were supported by project management consultants (at national, provincial and district levels). 16 Consultant performance was discussed during all project review missions and noted in the missions memorandums of understanding.

21 11 III. EVALUATION OF PERFORMANCE A. Relevance 48. Overall, the project remained relevant to the government and ADB, and consistent with their policies and priorities for reducing poverty in poor rural areas. No changes in the project scope and design were required during implementation. 49. The design of the project (i) conformed to the RPJMN for , which targets a reduction in the poverty rate from 14.2% in 2009 to 8% 10% by 2014; and (ii) was aligned with the government s core poverty reduction program, PNPM Mandiri, which provides block grants to poor rural and urban communities nationwide to improve essential social services and basic infrastructure. The RPJMN for includes PNPM Mandiri as a development tool to accelerate poverty reduction, with an indicative budget allocation of $6,754 million over 5 years for seven PNPM Mandiri programs The project derived its strength from its community-driven approach: community members were involved in investment planning and decision making, and block grants were transferred directly and transparently to communities. The CDD approach provides good-quality community infrastructure at significantly lower cost than technically comparable projects using traditional (non-cdd) modes of government contracting. Impact analyses of similar CDD projects in Indonesia demonstrate that the projects have a major impact on short-term employment generation and are highly cost-effective in improving community infrastructure. 51. Encouraged by the success of the project, the government requested ADB to provide additional assistance using the CDD approach to improve basic rural infrastructure under PNPM Mandiri. In November 2009, the Rural Infrastructure Support to the PNPM Mandiri Project II (footnote 4), aimed at rehabilitating and improving rural infrastructure in 1,500 villages, was approved, continuing the support provided under the first project. In August 2011, a third project contributing to the overall PNPM Mandiri scheme and also using the CDD approach was approved for $100 million Lessons from the Rural Infrastructure Support to the PNPM Mandiri Project I, relating in particular to the mobilization of community facilitators, have been integrated into the two new projects. The lessons include the need to reorganize monitoring arrangements and provide better guidance to community facilitators through the stronger involvement of DPIUs and PPIUs, and more careful selection and training of community facilitators PNPM Mandiri programs included in the RPJMN are: PNPM Urban ($664 million); PNPM Rural ($5,420 million); PNPM Regional Infrastructure for Social and Economic Development ($132 million); PNPM Rural Infrastructure ($335 million); PNPM for Sanitation (SANIMAS, $37 million); PNPM for Water Supply and Sanitation (PANSIMAS, $469 million); and PNPM Support for Poor and Disadvantaged Areas (SPADA, $277 million). In 2010, the government increased the allocation for PNPM Mandiri to $1.3 billion from $1.1 billion in ADB Report and Recommendation of the President to the Board of Directors: Proposed Loan to the Republic of Indonesia for the Urban Sanitation and Rural Infrastructure Support to the PNPM Mandiri Project. Manila. The project supports about 600 rural communities in the rehabilitation and improvement of basic infrastructure, and 1,350 poor urban neighborhoods in 34 cities in the improvement of sanitation services. 19 For example, the project administration manual for Rural Infrastructure Support to the PNPM Mandiri Project II includes a supplementary appendix, which contains recommendations (pertaining to workload, recruitment and selection, compensation, training program, etc.) based on the implementation of the first project, for improving the quality of community facilitators.

22 12 B. Effectiveness in Achieving Outcome 53. The project was effective in facilitating access to basic infrastructure services among the poor and near poor in the project districts. It gave the poor significantly better access to nearby markets where they could trade their farm produce, improved the supply of water for domestic use, provided better sanitation facilities, improved irrigation facilities, and provided permanent boat-landing amenities. The infrastructure improvements have facilitated access to public health facilities and good-quality education services among poor communities. At completion, the project had helped improve the livelihood of about 2 million poor people in 1,724 villages by constructing or upgrading economic as well as social infrastructure. The project covered 1,724 villages, exceeding the appraisal target of 1,650 villages. C. Efficiency in Achieving Outcome and Output 54. In general, the project was efficient in attaining its outcome and output in a cost-effective manner. Various surveys indicate that infrastructure built using the CDD approach cost much less (by about 40%) than similar works built by contractors. The infrastructure-upgrading works carried out under the project did not require a high level of expertise and could be performed by untrained rural labor. 55. Efficient implementation arrangements, including a competent PCMU at the DGHS and committed PPIUs and DPIUs in all participating provinces and districts, contributed to the successful implementation of the project. In general, the district and provincial staff involved in project administration and facilitation were dedicated and managed their technical project responsibilities well. Project activities were implemented as scheduled. The economic and financial reevaluation (Appendix 11) indicates that the project benefits have already offset the costs during the 5 years of project implementation. 56. The project had a twofold output: (i) an increase in community capacity for planning and development, and (ii) improvements in village services and infrastructure with the help of community development grants. 1. Community Facilitation and Mobilization 57. Under this component, the project provided support to communities to empower them and strengthen their capacity to select and implement infrastructure upgrading programs financed through block grants. Community facilitators (1,179 in all) were contracted to assist communities in planning, designing, and implementing investment projects in their villages. Committed and knowledgeable community facilitators were the backbone of the community empowerment process, ensuring that poor, disadvantaged, and low-educated community members were involved in planning and decision making for the block grants. The DGHS, realizing the vital role of community facilitators, gradually improved the mechanisms for their support and supervision, by deepening the involvement of district administrations. The DGHS organized additional training for the facilitators and introduced regular meetings at which they could share experiences and learn from successful community empowerment approaches. 58. However, observations made during project review missions indicate that the level of community participation largely varied from village to village, depending on the success of the empowerment process and the homogeneity of the communities. In some villages, the socialization activities appeared to have been rushed by the community facilitators such that many community members were not fully involved in planning and decision making. In other