Completion Report. Project Number: SRI Loan / Grant Number: L2167 / G0006 November Sri Lanka: Tsunami-Affected Areas Rebuilding Project

Completion Report Project Number: SRI 39144 Loan / Grant Number: L2167 / G0006 November 2012 Sri Lanka: Tsunami-Affected Areas Rebuilding Project

CURRENCY EQUIVALENTS Currency Unit Sri Lanka rupee/s (SLRe/SLRs) At Appraisal At Project Completion 28 January 2005 30 March 2011 SLRe1.00 = $0.0100 $0.0091 $1.00 = SLRs99.1100 SLRs110.3200 ABBREVIATIONS ADB Asian Development Bank CBO community-based organization km kilometer NDTF National Development Trust Fund NECCDEP North East Coastal Community Development Project NECORD North East Community Restoration and Development Project NGO nongovernment organization PMU project management unit TAARP Tsunami-Affected Areas Rehabilitation Project WRDS women s rural development society NOTE In this report, $ refers to US dollars. Vice President X. Zhao, Operations 1 Director General J. Miranda, Director General, South Asia Department (SARD) Director R. O Sullivan, Country Director, Sri Lanka Resident Mission (SLRM), SARD Team leader Team members C. McDeigan, Unit Head, Portfolio Management Unit, SLRM, SARD N. Amarasekera, Associate Project Analyst, SLRM, SARD N. Gunasekera, Senior Social Development Officer (Gender), SARD N. Salgado, Portfolio Management Officer, SARD K. Tilakaratne, Senior Project Officer, SARD 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.

iii CONTENTS Page BASIC DATA I. PROJECT DESCRIPTION 1 II. EVALUATION OF DESIGN AND IMPLEMENTATION 2 A. Relevance of Design and Formulation 2 B. Project Outputs 4 C. Project Costs 6 D. Disbursements 7 E. Project Schedule 7 F. Implementation Arrangements 7 G. Conditions and Covenants 8 H. Consultant Recruitment and Procurement 8 I. Performance of Consultants, Contractors, and Suppliers 9 J. Performance of the Borrower and the Executing Agencies 9 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 Outputs 12 D. Preliminary Assessment of Sustainability 12 E. Impact 13 IV. OVERALL ASSESSMENT AND RECOMMENDATIONS 14 A. Overall Assessment 14 B. Lessons 14 C. Recommendations 14 iv APPENDIXES Appendix 1: Design and Monitoring Framework with Achievements 16 Appendix 2: Description of the Tsunami-Affected Area Reconstruction Project 27 Appendix 3: Project Component Costs by Category and Donor 37 Appendix 4: S Curve for Actual Disbursements 39 Appendix 5: Implementation Schedule 40 Appendix 6: Project Organizational Chart 42 Appendix 7: Implementing Arrangements 43 Appendix 8: Status of Compliance with Loan Covenants 44 Appendix 9: Subprojects That Were Not Properly Used as of 31 December 2009 52 Appendix 10: Performance of Gender Action Plan 53

iv BASIC DATA A. Loan Identification 1. Country 2. Loan / grant number 3. Project title 4. Borrower 5. Executing agencies 6. Amount of loan / grant 7. Project completion report number Sri Lanka L2167 SRI (SF) / G0006 Tsunami-Affected Areas Rebuilding Project Government of Sri Lanka Ministry of Justice and Judicial Reform Ministry of Relief, Rehabilitation and Reconciliation Ministry of Highways Ministry of Urban Development and Water Supply Ministry of Regional Infrastructure and Development Ministry of Fisheries and Aquatic Resources Ministry of Finance and Planning Ministry of Relief, Rehabilitation and Reconciliation L2167: SDR4,553,000 G0006: $150,000,000 PCR: SRI 1366 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 and grant Interest rate Maturity (number of years) Grace period (number of years) No appraisal mission was conducted. The factfinding mission was conducted 15 28 February 2005. 30 March 2005 31 March 2005 14 April 2005 28 April 2005 28 July 2005 28 July 2005 0 30 June 2009 L2167: 31 January 2011 G0006: 1 April 2011 1 L2167: 1% per annum G0006: not applicable L2167: 40 G0006: not applicable 10

v 8. Disbursements a. Dates Initial Disbursement 31 October 2005 Effective Date 28 July 2005 Final Disbursement 30 March 2011 Original Closing Date 30 June 2009 Time Interval 65 months Time Interval 47 months b. Amount (i) G0006 ($) Category Original Allocation Last Revised Allocation Amount Disbursed Undisbursed Balance a 11-Legal assistance and governance 1,900,000 2,857,000 3,144,969 (287,969) 02B-Equipment (NECCDEP) 1,000,000 2,074,282 1,730,398 343,884 03B-Community infrastructure development (NECCDEP) 17,400,000 17,555,087 17,508,733 46,354 04B-Livelihood grants (NECCDEP) 2,700,000 3,699,977 3,782,083 (82,106) 05B-Training (NECCDEP) 500,000 495,034 495,034 0 06B-Consulting (NECCDEP) 300,000 366,878 366,878 0 07B-Increase in operating cost (NECCDEP) 1,700,000 1,672,006 1,672,006 0 10B-Performance audit (NECCDEP) 100,000 0 0 0 01C-Civil works (road rehabilitation) 51,600,000 54,045,868 52,631,327 1,414,541 06C-Consulting (road rehabilitation) 2,000,000 3,200,000 4,841,803 (1,641,803) 07C-Increase in operating cost (road rehabilitation) 1,500,000 3,000,000 1,929,396 1,070,604 10C-Performance audit (road rehabilitation) 100,000 500,000 406,033 93,967 01D-Civil works (water supply & sanitation) 8,000,000 3,645,378 3,785,820 (140,442) 02D-Equipment (water supply & sanitation) 1,000,000 3,291,771 3,114,380 177,391 06D-Consulting (water supply & sanitation) 300,000 64,669 64,669 0 07D-Increase in operating cost (water supply) 700,000 518,182 221,652 296,530 10D-Performance audit (water supply) 100,000 0 0 0 01E-Civil works (Southern Province reconstruction) 15,000,000 19,928,533 19,535,250 393,283 02E-Equipment (Southern Province reconstruction) 2,000,000 94,219 202,197 (107,978) 03E-Community infrastructure development (Southern Province reconstruction) 2,000,000 1,330,980 1,328,147 2,833 06E-Consulting (Southern Province reconstruction) 400,000 774,539 914,043 (139,504) 07E- Increase in operating cost (Southern Province reconstruction) 700,000 1,171,729 1,328,572 (156,843) 10E-Performance audit (Southern Province reconstruction) 100,000 0 0 0 01F-civil works (coastal resources management) 10,000,000 9,744,904 9,228,483 516,421 02F-Equipment (coastal resources management) 1,000,000 475,000 453,063 21,937 03F-Community infrastructure development (coastal resources management) 2,000,000 450,000 529,573 (79,573) 06F-Consulting (coastal resources management) 400,000 566,799 564,187 2,612

vi 07F-Increase in operating cost (coastal resources management) 600,000 630,000 731,789 (101,789) 08F-Coastal resources management (Southern) 1,000,000 13,297 13,297 0 10F-Performance audit (coastal resources management) 100,000 0 0 0 09G-Rural finance 5,000,000 817,768 817,768 0 10G-Performance audit (rural finance) 100,000 0 0 0 01H-Civil works (NECORD-Tsunami) 5,000,000 9,243,041 9,226,941 16,100 02H-Equipment (NECORD-Tsunami) 1,100,000 950,970 968,576 (17,606) 03H-Community infrastructure development (NECORD-Tsunami) 11,000,000 5,822,785 5,803,159 19,626 06H-Consulting (NECORD) 200,000 0 0 0 07H-Increase in operating cost (NECORD-Tsunami) 1,000,000 999,304 991,138 8,166 10H-Performance audit (NECORD- Tsunami) 100,000 0 0 0 Financial audit 300,000 0 0 0 Total 150,000,000 150,000,000 148,331,364 1,668,639 NECCDEP = North East Coastal Community Development Project, NECORD = North East Community Restoration and Development Project. a The undisbursed loan amount in G0006 was cancelled on 01 April 2011. Category Amount (i) L2167 (SDR) Original Allocation Last Revised Allocation Amount Disbursed Undisbursed Balance a Credit line for microfinance 4,033,000 1,669,409 1,680,416 (11,007) Consulting services 65,000 0 0 0 Incremental operating cost 260,000 0 0 0 Interest charge 195,000 195,000 91,195 103,805 Civil works for roads, Component C 0 2,688,591 2,379,358 309,233 Total 4,553,000 4,553,000 4,150,969 402,031 Total ($ equivalent) 7,000,000 7,000,000 6,272,195 727,805 a The undisbursed loan amount in L2167 was cancelled on 31 January 2011. 9. Local costs (financed) - Amount a (SDR) 2,379,358 - Amount ($ million) 3.7 - Percent of local costs 57.32% - Percent of total cost 1.48% a Grant financing does not distinguish between foreign exchange and local currency. C. Project Data 1. Project cost ($ million) Cost Appraisal Estimate Actual Foreign exchange cost 213.5 237.3 Local currency cost 45.8 24.4 Total 259.3 261.7

vii 2. Financing plan ($ million) Cost Appraisal Estimate Actual Implementation costs Borrower financed 39.1 20.7 ADB financed 156.7 154.6 Other external financing European Commission Government of the Netherlands Agence Française de Développement 53.2 0.0 10.0 67.0 8.1 11.3 Total 259.0 261.6 IDC costs Borrower financed 0.0 0.0 ADB financed 0.3 0.1 Other external financing European Commission Government of the Netherlands Total 259.3 261.7 ADB = Asian Development Bank, IDC = interest during construction. 0.0 0.0 0.0 0.0

viii 3. Cost breakdown by project component ($ million) Component Appraisal Estimate Actual Legal Assistance, Governance and Anticorruption 2.2 3.1 North East Coastal Community Development 28.0 26.6 Road Rehabilitation and Reconstruction 130.2 139.6 Water Supply and Sanitation 12.3 8.6 Southern Province Reconstruction 24.8 28.0 Coastal Resources Management 17.9 14.3 Rural Finance 11.9 3.3 NECORD (Tsunami) 31.7 38.2 Interest during construction 0.3 0.1 Total 259.3 261.7 4. Project schedule Item Appraisal Estimate Actual First Contract Awarded for Each Component Component A: Supply of Digitized Birth, Marriage, and Death Certificates with Database Package 1 Component B: Construction of Market Building at Valachchenai in Batticaloa Component C: Consultancy Services for Preparation of Outline Designs & Bid Documents for Civil Works Contracts in South & South East Region Component D: Supply & Installation of Mechanical Electrical Equipment & Accessories for Intake and Treatment for Ranna and Ambalantota Water Supply Scheme (Augmentation) Component E: Repairs to Damaged Roads in Thissamaharama Pradeshiya Saba Area Component F: Emergency Work at Hikkaduwa, South Coast, Bentota River Mouth, Kelani River Mouth, Bentota Barrier Beach Part I Component H: Procurement of Twenty-One 10 Kilovolt-Amp Generators Q1 2005 Q4 2010 Q2 2005 Q2 2008 Q2 2005 Q3 2005 Q3 2005 Q2 2005 Q3 2005 Q3 2006 Q1 2005 Q2 2005 Q3 2005 Q3 2005 5. Project performance report ratings Implementation Period Development Objectives Ratings Implementation Progress From 28 July 2005 to 28 November 2006 S S From 29 November 2006 to 29 May 2007 S S From 30 May 2007 to 30 July 2007 S U From 31 July 2007 to 29 May 2010 S S From 30 May 2010 to 30 July 2010 From 31 July 2010 to 31 December 2010 S S U S

ix D. Data on Asian Development Bank Missions Name of Mission a Date No. of Persons No. of Person-Days Specialization of Members b Fact-finding 15 28 Feb 2005 17 63 a, b, c, d, d, d, e, f, g, h, i, j, k, l, m, n, o Appraisal Not conducted Inception 5 18 May 2005 4 52 d, d, h, p Review 10 23 Dec 2006 2 15 h, p Review 5 22 Mar 2007 8 28 d, d, d, e, q, r, s, v Review 15 31 Oct 2007 14 67 d, d, d, d, e, h, p, t, q, u, x Review 7 8 Feb 2008 2 4 d, d Midterm review 21 30 Apr 2008 7 37 d, d, d, d, e, q, u, v, x Review 24 27 Jun 2009 1 3 d Review 27 Jul 4 Aug 2009 1 7 d Review 21 24 Nov & 11 19 Dec 2008 3 33 d, d, d, d, e, q, u, v, x Review 4 22 Jan 2010 9 22 d, d, d, d, e, p, w, v, x Review 20 23 Sep 2010 2 4 d, v Review 28 Nov 01 Dec 2010 2 6 d, v Project completion review 18 20 Jun 2012 & 08 10 Jul 2012 3 2 6 4 d, w a = senior transport specialist, b = country director, c = financial control specialist, d = project implementation officer, e = staff consultant, f = post-conflict implementation consultant, g = urban development specialist, h = Office of the General Counsel staff member, i = resettlement specialist consultant, j = environmental specialist, k = senior project economist, l = post-conflict management consultant, m = senior transport specialist, n = resettlement specialist consultant, o = project analyst, p = project administration unit head, q = senior project officer, r = economics officer, s = programs officer, t = legal/governance consultant, u = financial management/disbursements officer, v = assistant project analyst, w = senior portfolio management specialist, x = gender specialist.

I. PROJECT DESCRIPTION 1. On 26 December 2004, a 9.0 magnitude earthquake off the Sumatran coast sent several tsunami waves radiating into the Bay of Bengal, Andaman Sea, and Indian Ocean. Striking Sri Lanka s east and south coasts, the waves caused widespread damage and destruction to dwellings, buildings, roads, fishing boats, and other infrastructure. More than 35,000 lives were lost, with thousands more missing or injured. More than 500,000 people lost their homes and livelihoods, about 75% of whom were from the poor and lower income sectors working in the informal economy. 1 In addition to these direct victims, more than 1 million people living in the vicinity were indirectly affected through loss of economic activity and access to basic services. 2. The Government of Sri Lanka created a Special Task Force for Rebuilding the Nation later renamed the Reconstruction and Development Agency to assess the damage to personal lives, livelihoods, and basic services of the tsunami-affected people. 2 Asian Development Bank (ADB), the World Bank, Japan Bank for International Cooperation, and the United Nations assisted the government in developing a comprehensive assessment to determine the reconstruction needs of the damaged coastal areas. ADB provided technical assistance to support the first phase of the joint needs assessment and a report was presented to the government on 28 January 2005. The total costs estimated for the tsunami-related reconstruction came to over $1.5 billion. 3 3. Subsequent to the joint assessment, ADB conducted a short mission to prepare the Tsunami-Affected Areas Rebuilding Project (TAARP) to support needs agreed with the government. 4 ADB s intervention was designed to accelerate economic recovery of the affected areas through (i) establishing legal identities and rights of the affected people, (ii) reestablishing livelihoods, (iii) restoring access to basic services, and (iv) reconstructing damaged infrastructure. TAARP funding included $150 million from the Asian Tsunami Fund; 5 $6.7 million reallocated from ADB s ongoing loan portfolio; a $53.2 million grant from the European Commission, later increased to $67 million; and an $8.1 million grant from the Government of the Netherlands. TAARP s contribution consisted of 8 components as follow: (i) Component A: Legal Assistance, Governance, and Anticorruption. To assist victims of the tsunami to restore their legal identities, including (i) legal assistance to address issues arising from loss of documentation and identity papers; (ii) establishment of a mechanism for tsunami victims for dispute resolution and grievance review and mediation; (iii) restoration of birth, death, and marriage certificates; property titles, inheritance rights and identification cards; as well as identification of affected children, especially orphans; and (iv) financial support and cash grants to orphans and injured victims. (ii) Component B: North East Coastal Community Development. To help reconstruct basic infrastructure in the coastal communities of Ampara, Batticaloa, Jaffna, 1 ADB, World Bank, Japan Bank for International Cooperation. 2005. Post-Tsunami Recovery Program, Preliminary Damage and Needs Assessment. Colombo, Sri Lanka. 2 The Task Force focused on longer-term solutions, as immediate needs and support to the affected persons were addressed through other programs. The oversight responsibility of Reconstruction and Development Agency was eventually transferred to the External Resources Department of the Ministry of Finance and Planning. 3 ADB. 2005. Technical Assistance to the Democratic Socialist Republic of Sri Lanka for the Sri Lanka Post-Tsunami Needs Assessment and Preparation for Emergency Assistance Implementation Strategies. Manila. 4 ADB. 2005. Proposed Loans and Grants to the Democratic Socialist Republic of Sri Lanka for the Tsunami- Affected Areas Rebuilding Project (TAARP) and the North East Community and Restoration Project II (NECORD II). Manila. While both projects were prepared through the same technical assistance, NECORD is a separate project to continue support for people internally displaced by war. 5 ADB. 2005. Asian Tsunami Fund. Manila.

2 Kilinochchi, Mullaitivu, and Trincomalee districts that were affected by the tsunami. Interventions under the component included (i) construction of small-scale community infrastructure, (ii) livelihood development activities, and (iii) community development planning and coordination at the local authority level. The scope also included support for restoration and sustainable management of natural habitats damaged during the tsunami. (iii) Component C: Road Rehabilitation and Reconstruction. To rehabilitate and improve damaged coastal highways and roads, including emergency access and egress links to improve the overall transport network in tsunami-affected areas. (iv) Component D: Water Supply and Sanitation. To repair, expand, and improve water and sanitation systems in coastal communities, villages, and towns where the water supply systems were damaged by the tsunami in Batticaloa, Hambantota, Muttur, and Trincomalee districts. This component expanded supply capacity and constructed new distribution systems to service new resettlement areas built for displaced persons. (v) Component E: Southern Province Reconstruction. To restore access to basic services, including rehabilitation of local roads, culverts and drains, small-scale water supply, sanitation and solid waste management facilities, village markets, medical clinics and dispensaries, children s playgrounds, and community buildings. The Government of the Netherlands provided a $1.5 million grant to support an owner-driven housing program. (vi) Component F: Coastal Resources Management. To repair damage done to coastal stabilization structures, breakwaters, fishing boat anchorages, fisheries-related markets, and buildings used by fishermen societies for equipment storage and repair. (vii) Component G: Rural Finance. To provide noncommercial credit to restore livelihoods, support microenterprise training, and provide support for operational costs and rebuilding physical assets of microfinance institutions. (viii) Component H: North East Community Reconstruction and Development Project (NECORD). To reconstruct physical infrastructure, support community development, and help regenerate economic activity in the coastal communities damaged by the tsunami in Ampara, Batticaloa, Jaffna, Kilinochchi, Mullaitivu, and Trincomalee districts, following on from the objectives of NECORD. 6 The component included reconstruction of health clinics, special needs schools, water supply schemes, irrigation, rural roads, village and community infrastructure, local government facilities, and damaged and destroyed housing in the North East. 4. Components proposed under TAARP had a scope similar to those of ongoing projects. The design and monitoring framework is in Appendix 1. II. EVALUATION OF DESIGN AND IMPLEMENTATION A. Relevance of Design and Formulation 5. The project design and formulation were in response to the government s strategy to have a reconstruction plan to address the tsunami damage based on the joint needs assessment prepared in coordination with other donors. This ensured that all needs were addressed in an efficient manner without gaps or overlaps between the government and other donor programs. The project was consistent with ADB s disaster policy, which emphasizes a 6 ADB. 2001. Report and Recommendation of the President to the Board of Directors on a Proposed Loan to the Socialist Democratic Republic of Sri Lanka for the North East Community Restoration and Development Project. Loan 1846-SRI(SF). Manila.

3 holistic and systematic approach to disaster and emergency assistance that extends beyond mere reconstruction to rehabilitation. 7 The project and components are described in Appendix 2. 6. Due to the need for rapid response, the assessment and design were completed in four weeks. Project design relied heavily on input of the provincial government and line ministries. ADB s emergency policy allows for a general estimation of scope, followed by more detailed analysis, planning, and design during implementation. This flexibility enabled the implementing agencies to adjust the scope to reflect actual needs and demands during implementation. The subsequent selection of works and programs was based on established eligibility criteria while following agreed ADB safeguards and gender development plans. 8 Community development was driven by the participating villages community-based organizations (CBOs) with the assistance of implementing partner nongovernment organizations through a consultative process with local beneficiaries. This process was well established in the NECORD and NECCDEP projects. 7. Component A was formulated in close consultation with the Ministry of Finance and Planning and the Ministry of Justice, the latter being central to linking a number of legal and governance institutions. Component G followed the government s established procedures for microfinancing through the Ministry of Finance and Planning and the National Development Trust Fund (NDTF). NDTF disbursed funds to local microfinance institutions through Sri Lanka Savings Bank. 8. Under components B, E, and H, the tsunami-affected villages were mobilized and supported by implementing-partner NGOs. Village development plans, designs, and simplified works contracts were prepared with support from the district-level project management units (PMUs). Under Component B, the investment in each village was limited to $50,000, and implementing-partner NGOs mobilized communities and prepared participatory village development plans with the communities. The prioritized works in the village development plans were then approved by the divisional coordinating committee, which assigned the contracts to the respective CBOs such as rural development societies, women s rural development societies, and fisheries societies for implementation. This approach improved ownership and capacity of local governments and organizations to operate and maintain the new facilities. For components E and H, the investments were larger and had to be approved by the provincial project coordination committees, after which the work contracts were tendered by the PMUs. Components C, D, and F involved larger investments and followed normal ADB loan implementation procedures as established under past and ongoing projects. 9. There were no major changes to the project. Due to the ground conditions as explained in paras. 15 and 16, however, the 3-year project design implementation period was an ambitious target. At completion, the project s support to the government s strategic initiatives for economic, services, and infrastructure development remained highly relevant to government s poverty reduction policy framework. 9 Key elements of the project continued to conform to ADB s country partnership strategy for the country for 2009 2011. 10 These had focused on socially inclusive economic growth through greater access to economic and social services for the poor and those living in conflict-affected areas. 7 ADB.2004. Disaster and Emergency Assistance Policy. Manila. 8 No land acquisition, resettlement, or major environmental impact were allowed under TAARP. 9 Government of Sir Lanka, Department of National Planning, Ministry of Finance and Planning. 2010. The Emerging Wonder of Asia: Mahinda Chintana Vision for the Future. Colombo. 10 ADB. 2008. Country Partnership Strategy: Sri Lanka, 2009 2011. Manila.

4 B. Project Outputs 10. The project s physical outputs 11 and estimated beneficiaries 12 for each component are briefly summarized below. The achievements versus original targets are summarized in the Design and Monitoring Framework in Appendix 1. (i) Component A: Legal Assistance, Governance, and Anticorruption. The project provided legal facilitation and assistance for more than 23,000 beneficiaries from the 11 regional legal aid centers and special tsunami mediation boards established to address various legal and personal issues and complaints. 13 Overall, 6,800 persons received new or replacement identity documents while 5,600 children were brought under the protection of the National Child Protection Authority and child-based court procedures. The capacity of judicial and enforcement officers to respond to the tsunami s impacts was enhanced through a training program for magistrates and judges focused on legal aspects of property and child custodian issues arising from the tsunami. Some 35,000 persons received information, training, and advice on awareness of corruption and bribery as well as legal rights and entitlements from the special tsunami programs. More than 200 staff assisted the Department of Registration of Persons and the Registrar General s Department of Sri Lanka to decentralize their services to regional centers, particularly the digitization and computerization of birth, marriage, and death certificates. Sri Lanka s Commission to Investigate Allegations of Bribery and Corruption was strengthened through public awareness campaigns and outreach. Local district registration departments were trained in digitization and computerization of legal documents. The Government of Sri Lanka enacted legislation with special provisions for persons and property affected by the tsunami and incidental matters. The Tsunami (Special Provisions) Act No. 16 of 2005 was enacted by Parliament in June 2005. A NGO Secretariat was established under the Social Services Ministry to guide and monitor NGOs registration, action plans, progress and audit reports. However, there was no special regulation, rules or guidelines specifically laid down in relation to NGOs dealing with tsunami affected people and properties. (ii) Component B: North East Coastal Community Development. The project improved the provision of basic needs for 322 villages and created temporary employment for over 7,500 people. Twenty-two markets, 20 agricultural and handicrafts training facilities, and 85 storage and repair facilities for fishermen s societies were constructed to help restore livelihoods; 2,466 women (widows and other vulnerable women) received skills training and support in developing enterprises; 120 women s rural development societies (WRDSs) with a total of 7,800 members were established (WRDSs received a SLRs1.5 million [$15,000] start-up grant and many have increased this to SLRs3 4 million ($30,000 40,000) in savings which they use to provide microfinance to their members). Some 117,000 residents were indirect beneficiaries. The restoration and sustainable management of natural habitats damaged during the tsunami originally included in the scope of Component B were also completed under NECCDEP. 14 11 The achievements are as reported by the implementing agencies, in government project completion reports, confirmed by ADB review missions, reported via the external audit, and observed in spot checks by the project completion review mission. 12 The implementing agencies reported the number of beneficiaries for each subcomponent or program on a cumulative basis. Inasmuch as many beneficiaries under each component may benefit from several subcomponents, however, this approach may lead to double counting. 13 The judges, magistrates, and 34 consultants who conducted the workshops through the mobile clinics were either specialists in their respective fields or received training and guidance from the Ministry of Justice, which was the executing agency. 14 ADB. 2003. North Eastern Province Coastal Community Development Project. Manila.

5 (iii) Component C: Road Rehabilitation and Reconstruction. The targets at appraisal included rehabilitation of 300 kilometers (km) of national roads and connectors (classes A and B) and 200 km of provincial roads. A total of 330 km of classes A and B national roads were reconstructed and improved to current standards in remaining sections; 34 km of classes C, D, and E provincial roads were rebuilt. A population of around 5 million in the Eastern, Southern and Uva provinces is considered to consist of direct beneficiaries of the project. All residents in the communities along the rebuilt roads were indirect beneficiaries. The scope of the provincial roads was later reduced, as damage was limited to roads within a few meters of the coast (except in bays, where the waves were much higher and reached further inland). Also some of these roads were included into other programs by other donors. In their place, an additional 30 km of national roads was approved for rehabilitation and was completed. (iv) Component D: Water Supply and Sanitation. The appraised scope was to expand 3 small and medium schemes, as well as to construct 10 community-based schemes and over 500 common wells. At completion, 13 small and medium schemes had been expanded while 9 community-based schemes and 571 common wells had been constructed. In addition, 52 rainwater tanks were constructed in Trincomalee District. (v) Component E: Southern Province Reconstruction. The target outputs were to reconstruct community infrastructure in about 100 villages and provide shelter to about 2,000 families. At completion, the project supported reconstructing infrastructure in 125 villages and provided shelters for about 3,000 families. Through the project, 46 local government buildings and/or secretariats were rehabilitated; 306 sections of damaged roads repaired; plus 7 community centers, 4 bus stations, and 10 bus stands built. Eleven WRDSs were established for vulnerable groups (women based-initiatives). Also constructed were 14 markets, 5 cooperative society buildings, 18 institutional buildings (including 5 hospital wings), 3 education offices, 4 preschool buildings, 2 sports centers, and 4 staff quarters. (vi) Component F: Coastal Resources Management. The target outputs were to repair and extend coastal stabilization structures and 3 fisheries harbors, as well as to support restoration of the livelihoods of coastal communities. Twelve major coastal stabilization structures were repaired, heightened, and lengthened; 2 fisheries harbors were reconstructed (8 had been identified in the original scope but 6 were rebuilt using funds from other donors); 6 fishing villages received new storage, repair, and office facilities; 9 km of local access roads were repaired; and the Coastal Conservation Department office in Matara was constructed. (vii) Component G: Rural Finance. At completion, microfinance was provided to 7,750 beneficiaries (4,650 female and 3,100 male recipients). That compares to the appraisal target of 8,000. Total disbursement was $3.1 million through 31 local microfinance institutions. Activities under this component were halted for a year during implementation due to an ADB audit of the implementing agency (NDTF). Only 28% of the funds were utilized under this component, and the resulting savings were reallocated to Component C. (viii) Component H: NECORD (Tsunami). The basic community and institutional infrastructure were improved. Project interventions included repair and/or reconstruction of 87 institutional buildings (15 health clinics, 32 schools, 5 special needs schools, and 35 local government buildings); 16 road sections totaling 40 km; 10 community development schemes including water supply, drainage, and community centers; 24 WRDSs; 2 fisheries markets and equipment buildings; and 15 agricultural schemes including 5 agricultural extension centers, 3 milk-chilling plants, 3 food processing training centers, and 4 livestock breeding centers. Cofinancing from the Government of

6 the Netherlands used for the owner-driven housing repair and reconstruction program supported 6,500 families. The amount allocated for the shelter subcomponent was reallocated to the community infrastructure subcomponent once the cofinancing from the Government of the Netherlands had been confirmed. ADB administered this cofinancing and provided shelter for 6,500 families. C. Project Costs 11. The detailed cost estimates at appraisal and actual disbursements are in Appendix 3. The original estimated contribution from the government and beneficiaries at appraisal was $39.1 million equivalent, proportionately distributed for each component, based on estimates for taxes, duties, other fees, and labor input from beneficiaries. The government subsequently agreed, however, that grant funds would not be taxed and the government s contribution was reduced to $21.1 million equivalent. A summary of costs by component, net of the changes in government contribution and additional grants, is in the table below. Component Cost Breakdown by Project Components ($ million) Appraisal Estimate A: Legal Assistance, Governance, and Anticorruption 2.2 3.1 B: North East Coastal Community Development 28.0 26.6 C: Road Rehabilitation and Reconstruction 130.2 139.6 D: Water Supply and Sanitation 12.3 8.6 E: Southern Province Reconstruction 24.8 28.0 F: Coastal Resources Management 17.9 14.3 G: Rural Finance 12.2 3.3 H: NECORD (Tsunami) 21.7 38.2 Total 249.3 261.7 Actual 12. For Component A there was a 40% increase in cost, mainly due to the 1-year extension to support the National Child Protection Authority program for destitute children and to continue support for mediation and dispute resolution. It was originally planned to strengthen the 11 regional legal aid centers to process the new legal documents. However, people from remote villages could not travel to the regional centers and the government developed mobile clinics to reach tsunami victims. This required more legal consultants and related equipment resulting in higher cost. Component B experienced slow disbursements initially due to weak capacity of the implementing-partner NGOs and the resumption of hostilities. As a result, the scope was reduced by about 5%. Component C was able to rehabilitate an additional 30 km of national highway due to a $13.8 million increase in the European Commission grant. 13. During implementation Austria and South Korea showed interest to fund some of the water supply needs identified from the project. Therefore from component D, ADB reallocated the savings of $3.7 million to other components. The actual disbursements for components E and H were higher, as grants of $1.5 million and $6.6 million for owner-driven housing from the Government of the Netherlands and additional $11.3 million in cofinancing from Agence Française de Développement for community infrastructure development was provided for Component H. Component F originally included 8 fisheries anchorages, but only 2 were built as 6 were funded from other donors, and therefore actual costs were less than allocated. Component G disbursed about 28% of the allocated amount due to a suspension of activities at NDTF (the implementing agency) pending completion of an audit (see para. 40 of Appendix 2).

7 D. Disbursements 14. Disbursement projections were not undertaken initially, but these were completed by the end of 2005, when the scope of the components was clearer. Actual disbursements in the first 2 years were much higher than projected due to the quick start of major works under components C, D, and F. Disbursements in years 3 and 4 were low due to a slow start of the community works under components B, E, and H. Second generation imprest accounts were established by the executing agency of each component. The small community works under components B, E, and H followed statement of expenditure procedures for small works. The projected versus actual disbursements are summarized in Appendix 3 and the cumulative TAARP expenditures are depicted by an S-curve in Appendix 4. E. Project Schedule 15. Overall implementation took 19 months longer than the 36 months originally envisaged, and the project loan closing date of 30 June 2009 was extended to the end of January 2011. The implementation schedule was too ambitious. Preparation of TAARP did not benefit from the 8 12 months for project preparatory technical assistance as do regular loan projects. As a result, the planning and preliminary design carried out under such project preparatory technical assistance had to be completed during the early stages of TAARP implementation. This shortcoming was exacerbated by the fact that many of the implementing-partner NGOs were new to the project area and not familiar with planning and design of community infrastructure and with ADB procedures. Similarly, the district secretariats had to be trained in the simplified approval and contract award procedures. The resumption of hostilities in 2006 interrupted the implementation of activities in some components B and H villages, and it slowed some works due to lack of access to quarries and restricted movement of equipment and materials. 16. Procurement under Component B was initially slow, as the CBOs had to be trained on basic construction techniques, documentation, and billing procedures. Implementation was also delayed due to time needed for identifying and prioritizing subprojects. Small contractors under components E and H were not familiar with externally funded procurement procedures, and, because many were initially disqualified, this led to a high percentage of rebidding. CBOs and small contractors were often slow to mobilize and were manual-labor intensive. The majority of contracts had to be extended by 6 12 months. The larger contracts for roads and coastal protection structures had periodic delays due to temporary shortages of construction material, particularly gravel and asphalt. F. Implementation Arrangements 17. The initial implementation arrangements as planned at appraisal were utilized throughout the project. The project implementation schedule is in Appendix 5 and the organization chart in Appendix 6. The executing and implementing agencies are detailed in Appendix 7. The project had been designed such that components B, C, D, E, F, G, and H utilized the existing implementing agencies already responsible for similar ongoing projects in tsunami-affected areas. This facilitated project start-up and overall implementation efficiency, thus resulting in lower consulting and incremental costs as compared to regular ADB loan projects in Sri Lanka. Component A was implemented by a newly formed PMU at the Ministry of Justice. The PMU (after initial trial and error) developed a new mechanism using mobile clinics to reach the tsunami survivors in small villages. Other programs under this component were implemented by strengthening regional legal aid offices and through training local government

8 staff in legal documentation. In general, the implementation arrangements were adequate. However, delays were caused by the time it took for the government to appoint incremental staff to some PMUs. Due to the heavy burden of numerous projects arising from the $1.5 billion tsunami reconstruction effort, the government struggled to find experienced staff and, for the same reason, it was also difficult to find consultants, especially for the more remote PMUs near the conflict areas. 18. Community subprojects were approved by divisional coordinating committees and cluster projects by district coordinating committees. This arrangement was agreed during fact-finding to help the northeastern coastal communities which were victims of both conflict and, subsequently, tsunami damage. G. Conditions and Covenants 19. Project-specific covenants in the financing agreement were complied with as summarized in Appendix 8. However, Schedule 6, para. 25 of the financing agreement, which stated that the government shall ensure reasonable access to the project areas, could not be complied with fully, as the resumption of hostilities during implementation prevented access to some villages in the north and east. Schedule 6, para. 35 of the financing agreement, which required the design of all new structures under the project to include strengthening against high winds, earthquakes, and flooding, was not achieved for some local government buildings in the Southern Province because the local authorities used standard provincial building department designs. The time frame of this covenant was not realistic, as it takes time to develop new standards. The government did develop new building standards, but these were not promulgated until 2009. However, the design of such major structures as highway bridges, water treatment plants, hospital additions, and coastal revetment structures included hardening against earthquakes and tsunamis. 20. Covenants related to safeguards were complied with. Though an initial environmental examination and resettlement framework were prepared at appraisal, TAARP had no environmental impacts, land acquisition, or resettlement. A gender action plan was also completed at appraisal and the proposed initiatives were achieved. Physical improvements achieved through TAARP benefitted women equally with men, and the support for WRDSs, vocational training, and microcredit specifically targeted vulnerable women. H. Consultant Recruitment and Procurement 21. In accordance with the financing agreement, the implementing agencies engaged all international and domestic consultants following ADB s Guidelines on the Use of Consultants (2001, as amended from time to time) to help government agencies plan, design, and implement the project. Streamlined procedures for tsunami operations introduced in February 2005 under the Disaster and Emergency Assistance Policy facilitated early engagement and mobilization of consultants. 15 These procedures introduced several flexible, simplified modalities for consultant selection. These included direct selection of all individual consultants, singlesource selection of firms already active in the area, adjustment in thresholds and reduction in proposal submission periods for different types of contracts, and delegation of authority to the resident mission. The various TAARP implementing agencies engaged 5 firms through 5 contracts and a total of 90 individual consultants to assist in planning, design, capacity building, 15 ADB. 2004. Disaster and Emergency Assistance Policy. Operations Manual D7/BP and DT/OP: Disaster and Emergency Assistance. Manila.

9 training, and project management. Procurement for major contracts over $1 million followed ADB s international competitive bidding procedures, smaller contracts less than $1 million followed national competitive bidding procedures, and small village-level contracts under $50,000 could be procured through local procedures acceptable to ADB. 16 I. Performance of Consultants, Contractors, and Suppliers 22. Consultants recruited under TAARP to supplement existing PMUs generally performed satisfactorily. With the exception of Component F, consulting and incremental costs were less than half those for regular loan projects administered by the Sri Lanka Resident Mission. 23. The PMUs under components B, E, and H pointed to the inexperience of many implementing-partner NGOs that helped mobilize village populations and support the villagelevel committees in developing proposed subprojects. Many of these organizations were new to development initiatives and needed training by the PMUs to function within TAARP. PMUs needed to adjust a number of the initial village plans as work progressed until the training enhanced the capacity of these implementing-partner NGOs. 24. A number of small contractors across components demonstrated weak capacity and required extension of their contracts in order to complete the works. Initially, there were delays as the CBOs had to learn ADB s requirements for quality control, progress claims, and completion certificates. There were also delays caused by various suppliers of materials, especially granular base for roads, sand and gravel for cement, and oil for asphalt. The huge tsunami reconstruction effort had resulted in high demand for such materials. Nevertheless, major engineering works related to the national highways and water systems were completed satisfactorily to international standards. The contractor for 158 km of the coastal highway from Matara to Hambantota performed particularly well, with good quality work completed on schedule and on budget. The performance of the contractors was partly satisfactory. 25. The combination of weak implementing-partner NGOs which slowed project planning and design, inexperienced rural development societies and contractors, and shortage of materials were the main reasons for the 19-month project extension. The quality of the completed works generally met specifications, although a number of substandard works had to be redone or improved to pass quality control. The implementing-partner NGOs were not funded under the project and the small contractors had lower markups than did larger contractors, so many of the works under components B, E, and H were done at costs lower than projected. This enabled use of the savings to expand the scope of other components. J. Performance of the Borrower and the Executing Agencies 26. Eight different ministries and executing agencies were involved in implementing TAARP. 17 With the exception of Component A, all the executing agencies had extensive experience with ADB-financed projects and implemented TAARP through PMUs that were already in place for ongoing loan projects. Additional PMUs were established in 3 districts in Northern Province under Component B, but these were managed by experienced staff and supplemented by additional consultants and incremental staff. Though the executing agencies and PMUs already had extensive institutional capacity, ADB conducted 34 training sessions to 16 ADB. Procurement Guidelines (as amended from time to time). Manila. 17 Several of the ministries were consolidated during the project s progress due to the installation of a new national government. However, this did not affect performance of the executing agencies or the PMUs.

10 introduce the special requirements of TAARP and train new staff. The independent project performance audit conducted every 6 months by PricewaterhouseCoopers pointed out a number of weaknesses, especially in accounting and maintenance of separate records for TAARP in the early stages of the project. These deficiencies were addressed through the training sessions conducted by ADB staff and the ongoing support from ADB missions. Quality control was another area of concern due to the use of small, inexperienced contractors and CBOs in components B, E, and H, under which more than 3,000 contracts were implemented. ADB helped the PMUs improve quality control procedures through the training sessions and recruited an independent consultant to visit project sites for spot checks on quality. Although the project completion review mission noted a number of quality control issues during site visits, the works generally met standards. 27. The Ministry of Justice had not previously implemented an emergency-based program to restore identity and legal documentation to tens of thousands of people within a short time frame. There were initial delays related to finding a successful mechanism to reach the tsunamiaffected people in hundreds of villages across a wide geographic region. Once the new mechanism of mobile clinics was put in place, however, implementation of this component proceeded satisfactorily. Component G was implemented by NDTF using existing mechanisms for microcredit. While this component had low disbursements, the actual implementation followed established microcredit procedures. Overall, the performance of the borrower and the executing agencies was satisfactory. K. Performance of the Asian Development Bank 28. ADB provided administrative support and 34 training sessions to TAARP executing agencies and PMUs, mostly through its resident mission staff which had also played a major role in shaping the TAARP program and its implementation procedures. Even as ADB s emergency procurement policies saved considerable time by allowing post facto approval of small contracts, ADB staff provided extensive assistance to the PMUs to ensure they met ADB requirements. The training sessions focused upon accounting procedures, quality control, and preparation of works contracts for post facto approvals. ADB conducted 14 official missions for TAARP involving more than 800 person-days over the 58 months of project implementation. Moreover, ADB staff members also were able to attend most of the monthly provincial project coordination committee meetings, which approved subprojects, through concurrent review missions of ongoing projects. In addition, ADB funded an independent audit of the PMUs accounts every 6 months. The issues and gaps raised through these audits were discussed with the PMUs, whose procedures were adjusted to meet requirements. 29. The executing agencies and PMUs all expressed appreciation for the support that ADB staff provided in assisting all implementing agencies to adjust to the scope of the various components as required to reflect realities on the ground, including delays and lack of access caused by the renewal of hostilities in the eastern and northern regions of the project area. This avoided overlap with other programs while ensuring there were no gaps left by other donors. The dozen representatives of WRDSs met by the project completion review mission also thanked ADB for its monetary support and especially for the visits by ADB staff which improved their status with local governments. ADB was able to administer TAARP without increasing staff. Overall, ADB s performance was satisfactory.