Document of The World Bank FOR OFFICIAL USE ONLY Report No: ICR00004809 IMPLEMENTATION COMPLETION AND RESULTS REPORT IDA-55290 ON A CREDIT IN THE AMOUNT OF SDR 65 MILLION (US$ 100 MILLION EQUIVALENT) TO BOSNIA AND HERZEGOVINA FOR THE BIH FLOODS EMERGENCY RECOVERY PROJECT December 16, 2019 Urban, Resilience and Land Global Practice Europe and Central Asia Region CURRENCY EQUIVALENTS (Exchange Rate Effective Nov 04, 2019) Currency Unit = SDR SDR = US$1 US$ = SDR 1 FISCAL YEAR July 1 - June 30 Regional Vice President: Cyril E. Muller Country Director: Linda Van Gelder Regional Director: Steven N. Schonberger Practice Manager: David N. Sislen Task Team Leader(s): Ko Takeuchi, Goran Tinjic ICR Main Contributor: Carina Fonseca Ferreira ABBREVIATIONS AND ACRONYMS APCU Agriculture Project Coordination Unit BiH Bosnia and Herzegovina BD Brčko District CPS Country Partnership Strategy CRW Crisis Response Window DRM Disaster Risk Management EIB European Investment Bank EMP Environmental Management Plan ERR Economic Rate of Return ESMF Environmental and Social Safeguards Framework EU European Union FBiH Federation of Bosnia and Herzegovina FERP Floods Emergency Recovery Project GFDRR Global Facility for Disaster Reduction and Recovery NRA Recovery Needs Assessment PDO Project Development Objective PIT Project Implementation Team PIU Project Implementation Unit RS Republika Srpska TTL Task Team Leader UN United Nations TABLE OF CONTENTS DATA SHEET .......................................................................................................................... 1 I. PROJECT CONTEXT AND DEVELOPMENT OBJECTIVES ....................................................... 5 A. CONTEXT AT APPRAISAL .........................................................................................................5 B. SIGNIFICANT CHANGES DURING IMPLEMENTATION (IF APPLICABLE) ..................................... 11 II. OUTCOME .................................................................................................................... 12 A. RELEVANCE OF PDOs ............................................................................................................ 12 B. ACHIEVEMENT OF PDOs (EFFICACY) ...................................................................................... 13 C. EFFICIENCY ........................................................................................................................... 19 D. JUSTIFICATION OF OVERALL OUTCOME RATING .................................................................... 22 E. OTHER OUTCOMES AND IMPACTS (IF ANY) ............................................................................ 22 III. KEY FACTORS THAT AFFECTED IMPLEMENTATION AND OUTCOME ................................ 24 A. KEY FACTORS DURING PREPARATION ................................................................................... 24 B. KEY FACTORS DURING IMPLEMENTATION ............................................................................. 24 IV. BANK PERFORMANCE, COMPLIANCE ISSUES, AND RISK TO DEVELOPMENT OUTCOME .. 25 A. QUALITY OF MONITORING AND EVALUATION (M&E) ............................................................ 25 B. ENVIRONMENTAL, SOCIAL, AND FIDUCIARY COMPLIANCE ..................................................... 26 C. BANK PERFORMANCE ........................................................................................................... 27 D. RISK TO DEVELOPMENT OUTCOME ....................................................................................... 28 V. LESSONS AND RECOMMENDATIONS ............................................................................. 29 ANNEX 1. RESULTS FRAMEWORK AND KEY OUTPUTS ........................................................... 32 ANNEX 2. BANK LENDING AND IMPLEMENTATION SUPPORT/SUPERVISION ......................... 44 ANNEX 3. PROJECT COST BY COMPONENT ........................................................................... 46 ANNEX 4. EFFICIENCY ANALYSIS ........................................................................................... 47 ANNEX 5. BORROWER, CO-FINANCIER AND OTHER PARTNER/STAKEHOLDER COMMENTS ... 50 ANNEX 6. SUPPORTING DOCUMENTS (IF ANY) ..................................................................... 51 ANNEX 7. PROJECT PARTS AND COMPOSITION OF RELATED SUB-COMPONENTS .................. 53 ANNEX 8. PROJECT FUNDS ALLOCATION .............................................................................. 55 The World Bank BiH Floods Emergency Recovery Project (P151157) DATA SHEET BASIC INFORMATION Product Information Project ID Project Name P151157 BiH Floods Emergency Recovery Project Country Financing Instrument Bosnia and Herzegovina Investment Project Financing Original EA Category Revised EA Category Partial Assessment (B) Partial Assessment (B) Organizations Borrower Implementing Agency FBiH Ministry of Agriculture, Water Management and BiH Ministry of Finance and Treasury Forestry, RS Ministry of Agriculture, Water Management and Forestry Project Development Objective (PDO) Original PDO The Project Development Objective (PDO) is to meet critical needs and restore functionality of infrastructure essential for public services and economic recovery in floods affected areas. Page 1 of 55 The World Bank BiH Floods Emergency Recovery Project (P151157) FINANCING Original Amount (US$) Revised Amount (US$) Actual Disbursed (US$) World Bank Financing 100,000,000 100,000,000 87,527,164 IDA-55290 Total 100,000,000 100,000,000 87,527,164 Non-World Bank Financing 0 0 0 Borrower/Recipient 0 0 0 Total 0 0 0 Total Project Cost 100,000,000 100,000,000 87,527,164 KEY DATES FIN_TABLE_DAT Approval Effectiveness MTR Review Original Closing Actual Closing A 30-Jun-2014 15-Sep-2014 03-Jun-2016 31-Dec-2018 30-Jun-2019 RESTRUCTURING AND/OR ADDITIONAL FINANCING Date(s) Amount Disbursed (US$M) Key Revisions 09-Jan-2016 27.46 Change in Results Framework Change in Components and Cost Reallocation between Disbursement Categories 12-Dec-2018 76.73 Change in Loan Closing Date(s) KEY RATINGS Outcome Bank Performance M&E Quality Highly Satisfactory Satisfactory Substantial Page 2 of 55 The World Bank BiH Floods Emergency Recovery Project (P151157) RATINGS OF PROJECT PERFORMANCE IN ISRs Actual No. Date ISR Archived DO Rating IP Rating Disbursements (US$M) 01 20-Oct-2014 Satisfactory Satisfactory 4.41 02 29-May-2015 Satisfactory Moderately Satisfactory 11.14 03 17-Dec-2015 Satisfactory Moderately Satisfactory 23.42 04 22-Jun-2016 Satisfactory Moderately Satisfactory 36.19 05 16-Dec-2016 Satisfactory Moderately Satisfactory 50.13 06 14-Jun-2017 Satisfactory Satisfactory 57.73 07 15-Dec-2017 Satisfactory Satisfactory 66.14 08 20-Jun-2018 Satisfactory Satisfactory 70.43 09 29-Nov-2018 Satisfactory Satisfactory 75.83 10 24-Apr-2019 Satisfactory Satisfactory 80.94 SECTORS AND THEMES Sectors Major Sector/Sector (%) Agriculture, Fishing and Forestry 25 Other Agriculture, Fishing and Forestry 25 Public Administration 10 Other Public Administration 10 Transportation 40 Other Transportation 40 Page 3 of 55 The World Bank BiH Floods Emergency Recovery Project (P151157) Water, Sanitation and Waste Management 25 Other Water Supply, Sanitation and Waste 25 Management Themes Major Theme/ Theme (Level 2)/ Theme (Level 3) (%) Finance 25 Finance for Development 25 Disaster Risk Finance 25 Urban and Rural Development 75 Disaster Risk Management 75 Disaster Response and Recovery 25 Disaster Risk Reduction 25 Disaster Preparedness 25 Environment and Natural Resource Management 90 Climate change 90 Adaptation 90 ADM STAFF Role At Approval At ICR Regional Vice President: Laura Tuck Cyril E Muller Country Director: Ellen A. Goldstein Linda Van Gelder Director: Laszlo Lovei Steven N. Schonberger Practice Manager: Sumila Gulyani David N. Sislen Task Team Leader(s): David Michaud Ko Takeuchi, Goran Tinjic ICR Contributing Author: Carina Fonseca Ferreira Page 4 of 55 The World Bank BiH Floods Emergency Recovery Project (P151157) I. PROJECT CONTEXT AND DEVELOPMENT OBJECTIVES A. CONTEXT AT APPRAISAL Context In May 2014, Bosnia and Herzegovina (BiH) experienced heavy rainfall and catastrophic floods which would severely impact economic growth. Negative growth was projected to be -1.1 per cent for 2014, as opposed to projections made prior to the floods which estimated increases in economic growth from 1.8 percent in 2013 to 2.0 per cent in 2014 and 3.5 per cent in 20151. A state of emergency was declared on May 14, 2014 in affected municipalities of the Federation of Bosnia and Herzegovina (FBiH) and Republika Srpska (RS) as well as in Brčko District (BD)2. By the end of May 2014, at least 25 people were confirmed dead, 90,000 people had been evacuated from their homes, and about 1 million (more than a quarter of the total BiH population) had been directly affected in approximately 60 municipalities (nearly half of the 142 municipalities in BiH). The floods affected an area covering more than one-third of the country’s territory and caused over 3,000 landslides. According to the Recovery Needs Assessment (RNA)3 conducted by the government with support from the European Union (EU), United Nations (UN) and World Bank between May 25 and June 17, 2014, the floods caused an overall impact of around US$ 2.8 billion (US$ 1.7 billion in physical damages and US$ 1.1 billion in other losses) which was equivalent to nearly 15 percent of GDP in 2014. More than 51 percent of damages and losses occurred in the FBiH, 48 percent in the RS and around 1 percent in the BD (see Table 1 and Table 2). Table 1: Damages and losses caused by the 2014 floods in BiH (in US$ million)1,4 Damages Losses Total (%) Agriculture 141.9 112.7 254.6 9% Education 10.9 0.9 11.8 0.4%% Energy 67.6 71.1 138.7 5% Floods protection 67.0 0 67.0 2% Health 7.9 64.1 72.0 3% Housing and household items 577.7 38.7 616.4 22% Livelihoods and Employment 471.3 607.2 1,078.5 39% Public Services and Facilities 25.0 12.3 37.3 1% Transport and Communications 356.0 116.9 472.9 17% Water and Sanitation 7.4 2.8 10.2 0.4% Gender 0 11.5 11.5 0.4% Total 1,732.6 1,038.2 2,770.8 1 World Bank. 2014. Project Appraisal Document on a Proposed Credit in the Amount of SDR 65 Million in IDA Crisis Response Window Resources to Bosnia and Herzegovina for a Floods Emergency Recovery Project. Washington, DC: World Bank. 2 The decentralized institutional structure of BiH dates to the 1995 Dayton Peace Accords and assigns most power to the two entities: Federation of Bosnia and Herzegovina (the Federation or FBiH) and Republika Srpska, plus the special district of Brcko. 3 European Union, United Nations, World Bank, Global Facility for Disaster Reduction and Recovery. 2014. Recovery Needs Assessment 2014: Bosnia and Herzegovina Floods. 4 Damages include physical damage caused to property, infrastructure and goods while losses include the effects on livelihoods, income and production, among other factors. Page 5 of 55 The World Bank BiH Floods Emergency Recovery Project (P151157) Table 2: Damages and losses across BiH (in US$ million)4 Damages and losses (%) FBiH 1,414.4 51% RS 1,316.9 48% BD 40.3 1% Total 2,770.8 World Bank’s prompt response to the disaster included different instruments and focused on short to long -term objectives. First, the Bank, in partnership with the EU and UN, and with support of the Global Facility for Disaster Reduction and Recovery (GFDRR) provided assistance to the Government in preparation of the RNA that formed a basis for development of effective infrastructure and services rehabilitation measures in the affected areas and estimated the recovery and reconstruction needs (see Table 3). Second, the Bank prepared and promptly approved the emergency financing operation - Floods Emergency Recovery Project (FERP) with US$100 million allocated from the IDA Crisis Response Window (CRW). Third, the Bank conducted a portfolio review to identify projects that could be restructured to enable efficient recovery’s support. Finally, ongoing and planned projects (including Technical Assistance) were considered to contribute to building resilience in the long-term by supporting efforts to move from a reactive response to a proactive disaster risk management approach. Table 3: Recovery and reconstruction needs after the 2014 floods in BiH (in US$ million)5 Recovery Reconstruction Total (%) Agriculture 47.4 14.4 61.8 3% Education 7.2 12.0 19.3 1% Energy 4.6 82.6 87.2 4% Flood protection 0 249.5 249.5 10% Health 10.5 570.8 581.3 24% Housing and household items 127.7 601.5 729.2 30% Livelihoods and employment 13.8 28.1 41.9 2% Public Services and Facilities 0 480.6 480.6 20% Transport 12.2 0 12.2 1% Water and Sanitation 54.0 0 54.0 2% Gender 64.5 7.9 72.3 3% Security 5.3 7.4 12.7 1% Total 347.2 2,054.7 2,401.9 FERP sought to address an urgent crisis and to restore BiH to a growth path following the 2014 floods. The emergency response project was prepared in a short period of time with the use of streamlined procedures for operations in situations of urgent need for assistance6. The credit was approved on June 30, 2014, one and a half 5 European Union, United Nations, World Bank, Global Facility for Disaster Reduction and Recovery. 2014. Recovery Needs Assessment 2014: Bosnia and Herzegovina Floods. 6 Bank Procedure – Preparation of Investment Project Financing – Situations of Urgent Need of Assistance or Capacity Constraints Page 6 of 55 The World Bank BiH Floods Emergency Recovery Project (P151157) months after the declaration of a state of emergency, to rapidly support entity-level and local governments during the recovery phase. The project followed the administrative-territorial organization of the country, and it consisted accordingly of three parts - one for each entity FBiH and RS as well as the BD. Under each part, the activities were organized into similar components and implemented by each entity through their own project implementation units (PIUs) and steering committees. To enable prompt emergency response, the project’s implementation arrangements were built upon existing units and institutional capacities of ongoing World Bank projects in both the FBiH and RS. It sought to follow international experience suggesting that economic growth was likely to resume once recovery gets fully underway, as well as World Bank policies and best practices in the aftermath of disasters which follow the principles of “building back better”7. These principles advocate the need to improve resilience, while implementing recovery efforts to reduce the risk of future disasters. To maximize the project outcomes, FERP focused on three most disaster affected sectors that delivered public services and were essential for economic recovery - Agriculture and Rural Development, Transport, and Water and Sanitation. At the time of project appraisal, preliminary evidence showed that the largest impact from the floods was on infrastructure related to housing, local services, transport, water and sanitation and agriculture. Later findings from the RNA confirmed the initial evidence showing that these sectors were significantly affected by the disaster concentrating at least 35 percent of the total damages and losses and nearly 40 percent of the total recovery and reconstruction needs (as evidenced by RNA data presented in Tables 1 and 3 above). Although FERP planned to focus on the three identified priority sectors, other public infrastructure such as educational and health facilities could also be considered on a case-by-case basis. FERP’s component on rehabilitation of public infrastructure was designed under a framework approach to allow for prioritization of activities during project implementation based on the results of the RNA. Given the nature of the operation prepared in a situation of urgent need for assistance, the project priorities were tentatively identified during the project’s preparation based on preliminary information on the impact of the floods available shortly after the event. Where information was insufficient to define specific tasks, the project followed a framework approach which was tailored to facilitate prioritization during project implementation. Since the RNA was still ongoing at the time the PAD was finalized, the project framework at appraisal included the possibility to accommodate other priorities that could emerge based on the RNA results. 7 Further information on the World Bank’s policies and best practices in the aftermath of disasters based on the principles of “building back better” can be found for example in: (i) World Bank. 2012. Acting today for tomorrow: a policy and practice note for climate and disaster resilient development in the Pacific Islands region: Main report (English). Washington, DC: World Bank. http://documents.worldbank.org/curated/en/231221468288339561/Main-report (ii) Urbano M De Bettencourt, Teresa Sofia; Ebinger, Jane Olga; Fay, Marianne; Ghesquiere, Francis; Gitay, Habiba; Krausing, Jarl; Kull, Daniel Werner; Mccall, Kevin; Reid, Robert Curle Jesse; Simpson, Alanna Leigh. 2013. Building resilience: integrating climate and disaster risk into development - the World Bank Group experience: Main report (English). Washington DC; World Bank. http://documents.worldbank.org/curated/en/762871468148506173/Main-report, and (iii) Jha, Abhas K.; Barenstein, Jennifer Duyne; Phelps, Priscilla M.; Pittet, Daniel; Sena, Stephen; Jha, Abhas K.; Barenstein, Jennifer Duyne; Phelps, Priscilla M.; Pittet, Daniel; Sena, Stephen. 2010. Safer homes, stronger communities: a handbook for reconstruction after natural disaster: Safer homes, stronger communities: a handbook for reconstructing after natural disasters (English). Washington, DC: World Bank. http://documents.worldbank.org/curated/en/290301468159328458/Safer-homes-stronger- communities-a-handbook-for-reconstructing-after-natural-disasters Page 7 of 55 The World Bank BiH Floods Emergency Recovery Project (P151157) In addition to rehabilitation of public infrastructure, the project envisaged financing the purchase of goods for priority sectors using a positive list for fast disbursement against government expenditures for emergency goods during and shortly after the emergency period. Although at the time of the project appraisal the RNA was still ongoing, the governments had tentatively defined their priorities in emergency goods, namely (i) the need to replenish and maintain their respective strategic reserves of food and fuel, which were being quickly depleted through the disaster recovery effort; (ii) the need to rapidly support housing reconstruction in flooded areas, and the agriculture sector by replacing livestock and providing seeds and other agricultural goods to farmers in time for the growing season; and (iii) the need to support civil protection agencies with the purchase of additional emergency goods. FERP supported the governments of BiH to address these priorities by financing the purchase of goods including food, fuel, construction materials, agricultural goods, among other emergency goods. Theory of Change (Results Chain) FERP was designed to support entity-level and local governments in BiH to respond to the crisis and restore the country growth path through two main outcomes: (i) critical needs met in flood affected areas, and (ii) restored functionality of infrastructure essential for public services and economic recovery in flood affected areas. The first outcome was planned to be achieved through financing the purchase of goods urgently needed for disaster recovery, as per positive list developed under FERP in line with government’s priorities. The project’s positive list included: (i) logistical goods, (ii) reconstruction goods, (iii) emergency goods, (iv) energy and power sector goods, (v) agricultural recovery goods, and (vi) any other goods or commodities essential to the Recipient’s recovery from flooding, as agreed in advance with the World Bank. For the second outcome, the project financed rehabilitation of both local and regional disaster-affected public infrastructure essential for public services and economic recovery such as roads and railways. In the long-term, the project sought to restore BiH to a growth path while improving climate adaptation through more resilient critical infrastructure. Although the aim was to restore public services disrupted by the floods to pre- floods condition, rehabilitation of infrastructure sought to follow the principles of “building back better”. Following these principles, rehabilitated infrastructure would be designed to be more resilient to floods than before the 2014 disaster, thus reducing the potential impact of future floods and improving adaptation to climate change in the long- term. Two important underlying assumptions for achieving the project’s outcomes were as follow: (i) the governments had adequate capacity to assess critical needs and distribute emergency goods to disaster affected populations, and (ii) municipalities had sufficient capacity to prioritize and implement rehabilitation works in order to restore functionality of infrastructure essential for public services and economic recovery . The Governments’ capacity to distribute eligible goods to the populations in need in the aftermath of the floods was essential to reach the first project outcome on critical needs met in flood affected areas, as there were multiple institutions including line ministries in both entities and the BD responsible for preparing the final list of goods to be financed by the project and ensuring successful disbursement of these funds. The local governments’ capacity to define priorities for rehabilitation of local public infrastructure, provide any necessary building permits, participate in procurement processes, and oversee implementation of civil works was essential to achieve the second project objective on restoring functionality of infrastructure. The Results Chain below (Figure 1) depicts the links between the project’s activities, outputs and outcomes. Page 8 of 55 The World Bank BiH Floods Emergency Recovery Project (P151157) Figure 1: Theory of change – Floods Emergency Recovery Project Project Development Objectives (PDOs) The Project Development Objective (PDO) is to meet critical needs and restore functionality of infrastructure essential for public services and economic recovery in floods affected areas. Key Expected Outcomes and Outcome Indicators The expected outcomes and respective PDO indicators to be used to assess achievement of each component of the project’s outcomes are shown in Table 4. Table 4: Unpacked Project Outcomes and Corresponding Outcome indicators Outcomes PDO Indicators A. Critical needs met in floods affected Indicator 1: Emergency goods distributed by entity-level areas strategic reserves (US$) B. Restored functionality of Indicator 2: Public service infrastructure facilities recovered to infrastructure essential for public pre-floods condition with project support (Number) services and economic recovery in floods affected areas A&B Indicator 3: Direct project beneficiaries (Number), of which female beneficiaries (Percent) Page 9 of 55 The World Bank BiH Floods Emergency Recovery Project (P151157) Components The project consists of three parts. Part A includes activities for FBiH, Part B supports activities for the RS and Part C for the BD. Each of the project’s parts include three similar components, as follows: Component 1: Emergency Disaster Recovery Goods (estimated cost US$40.0 million/actual cost US$32.6 million) financed, through the use of a positive list, the purchase of goods urgently needed during the recovery phase. The positive list included logistical goods, reconstruction goods, emergency goods, energy and power sector goods as well as agricultural recovery goods, with an initial priority set of fuel and food, agricultural and emergency goods. Expenditures under this Component included reimbursement of goods purchased during the emergency phase, emergency goods that the governments intended to purchase to continue responding to the emergency, or replenishment of stocks depleted during the emergency. Component 2: Rehabilitation of Key Public Infrastructure (estimated cost US$57.0 million/actual cost US$64.4 million) supported the rehabilitation of both local and regional public infrastructure based on the RNA. It financed the rehabilitation or reconstruction of high priority, public service delivery infrastructure in the worst-affected areas and was designed under a framework approach in which activities were identified in a demand-driven manner based on the RNA. The Component was divided in two Subcomponents that followed different project cycles, namely (a) Regional Infrastructure Rehabilitation Subcomponent to support the rehabilitation of strategic, regional-level public infrastructure such as transport links, distribution and transmission infrastructure, dikes and other strategic public infrastructure, with overall sector financial envelopes set at the entity level by the Steering Committees and the list of concrete activities defined by entity ministries in charge of particular sectors for further implementation by the PIUs, and (b) Local Infrastructure Rehabilitation Subcomponent with a focus on local-level infrastructure, such as water supply schemes, local roads and bridges, where the implementation process was to be based on a principle of allocating resources across the most affected municipalities, allowing municipalities to prioritize activities, and bundling those activities into larger packages procured by the PIUs. Component 3: Project Implementation Support and Capacity Building (estimated cost USD3.0 million/actual cost US$3.0 million): financed the management of the project and support needed for the identification and implementation of the sub-activities. The component also provided technical assistance and capacity building to help ensure future improved disaster resilience. The amounts allocated for each of the project’s components were revised through the formal project restructuring in 2016 when US$7.4 million were reallocated from Component 1 to Component 2. This restructuring is discussed in more detail in section B of the present report. Annex 7 outlines the structure of the project. The respective allocation of funds is shown in Annex 8. Page 10 of 55 The World Bank BiH Floods Emergency Recovery Project (P151157) B. SIGNIFICANT CHANGES DURING IMPLEMENTATION (IF APPLICABLE) Revised PDOs and Outcome Targets Overall, the project was formally restructured twice – in January 2016 and in December 2018 - with main changes approved at an early stage of implementation (around fourteen months after effectiveness) as part of restructuring conducted in 2016. The original PDO was not revised during the project. Revised PDO Indicators The main PDO indicators were not revised during the project. At the same time, since FERP was prepared in a situation of urgent need for assistance, target values of the PDO outcome indicators were determined during the first year of implementation after the RNA was completed, and actual investments and projected achievements were finally confirmed. The finalized results framework was stipulated through the first project restructuring approved on January 9, 2016. Main changes related to PDO indicators include:  the end target value for PDO indicator #1 “Emergency goods distributed by entity-level strategic reserves (US$)” was reduced from US$ 40 million to US$ 30 million to reflect reallocation of funds from component 1 to component 2,  the end target value for PDO indicator #2 “Public service infrastructure facilities recovered to pre-floods condition with project support (number)” was determined as 200 facilities, and  the end target value for PDO indicator #3 “Number of direct project beneficiaries” was set as 300,000. This result indicator was further broken down by the number of direct beneficiaries under project components 1, 2A and 2B with corresponding target values assigned. A supplemental sub type indicator to account for the percentage of female beneficiaries remained unchanged with an end target of 51% of total beneficiaries. Revised Components The composition of the project’s intermediate results indicators was slightly adjusted as part of January 2016 restructuring to better track the full scope of FERP’s supported activities. Main changes included: (i) deletion of indicator on “Distribution lines constructed or rehabilitated”, since the Governments have decided they would not support electricity investments under the project, because the related companies were profit-making, and (ii) inclusion of three new intermediate indicators on “Number of bridges rehabilitated under the project”, “Length of flood protection infrastructure rehabilitated or reconstructed”, and “Number of landslides rehabilitation activities”. Other Changes Reallocation of Funds: The first project restructuring in January 2016 supported reallocation of US$7.4 million from Component 1 “Emergency Disaster Recovery Goods” to Component 2 “Rehabilitation of Key Public Infrastructure”. These funds remained unused under Component 1 after completion of the emergency phase, and this reallocation allowed to assign them to finance rehabilitation of key public infrastructure under Component 2. Adjustment of Description of Component 3B “Capacity Building for Disaster Resilience”: this sub-component was initially designed as a pure technical assistance component. During project restructuring in January 2016, this component was adjusted to allow it to finance the purchase of disaster preparedness goods and equipment. Page 11 of 55 The World Bank BiH Floods Emergency Recovery Project (P151157) Extension of the Closing Date: In December 2018, the project was extended by six months, from December 31, 2018 to June 30, 2019, to allow the project to complete the remaining activities and fully utilize the remaining credit proceeds. Rationale for Changes and Their Implication on the Original Theory of Change The above described changes did not affect the Theory of Change and the originally expected project outcomes. The changes related to finalization of the Results Framework were necessary, as the FERP was prepared in a situation of urgent need for assistance and followed the streamlined procedures to enable the client to get immediate financial support. Due to the nature of this operation as an emergency recovery loan, the end targets for most indicators were not known at appraisal, because the RNA was still ongoing at that time, and were determined during the implementation stage only. It should be noted that the PDO indicators remained the same with changes made only in their end targets and introduction of the breakdown of one of the indicators to better track and report the number of direct beneficiaries under the particular project’s components. The changes made to the intermediate results indicators sought to align the Results Framework with the full scope of the confirmed activities, and better capture and monitor the project outputs. The reallocation of funds from Component 1 to Component 2 was justified by the fact that funds originally allocated to the purchase of emergency goods exceeded the needs, while the needs for rehabilitation of disaster-affected infrastructure exceeded funds allocated. Thus, this change sought to adjust the project to better respond to the recovery needs, as these needs were made increasingly evident from the RNA and the progress in the field captured by local institutions. Finally, the project closing date extension was needed to ensure enough time for the completion of all ongoing activities and utilization of remaining credit proceeds to enable the project to fully achieve its objectives and targets. At the time of the second restructuring, the project had disbursed 76 percent of the funds, and the remaining credit proceeds were either contracted or almost fully committed. While the extension of the project closing date slightly affected the project’s efficiency, it did not have any significant implication for the original theory of change. II. OUTCOME A. RELEVANCE OF PDOs Assessment of Relevance of PDOs and Rating Relevance Rating: High The objectives of FERP are fully relevant to the current World Bank Country Partnership Framework (CPF) for BiH and aligned with governments’ strategic objectives, and therefore the relevance of PDOs is rated as “High”. Building resilience to natural shocks is one of the three focus areas of the CPF for FY16-20. In particular, CPF’s objective 3b - build resilience to floods – refers to the Bank’s support to build resilience to natural shocks by capitalizing on its existing multi-sector engagements in Disaster Risk Management (DRM), and by strengthening its support in water resources management for better protection and preparedness against floods. Among others, the CPF includes that the Bank will continue to support disaster risk management via the FERP, possibly complementing it with further DRM Page 12 of 55 The World Bank BiH Floods Emergency Recovery Project (P151157) policy dialogue. In the 2015 Strategic Framework for BiH, which has been serving as a guiding portfolio of strategic objectives for preparation of medium-term work programs by the BiH Council of Ministers, a strategic objective related to the resilience and sustainability of infrastructure was included - improving environmental management and development of environmental infrastructure while increasing resilience to climate change. Continued relevance is demonstrated by the Sava and Drina Rivers Corridors Integrated Development Program (P168862) in the pipeline whose objective is to improve flood protection and enable transboundary water cooperation in the Sava and Drina Rivers Corridors. B. ACHIEVEMENT OF PDOs (EFFICACY) Assessment of Achievement of Each Objective/Outcome Efficacy Rating: High The project’s PDO statement specifies two outcomes: (A) meet critical needs in flood affected areas; and (B) restore functionality of infrastructure essential for public services and economic recovery in flood affected areas These outcomes are linked to the PDO indicators described in Table 5. Table 5: Achievement of PDO indicators Outcome PDO Indicator Achievement (A) Critical needs met in Emergency goods distributed by entity- US$ 27 million emergency goods flood affected areas level strategic reserves (US$) distributed versus target value of US$ 30 million Direct project beneficiaries under 104,537 beneficiaries versus target value Component 1 of 75,000 (B) Restored Public service infrastructure facilities 446 facilities recovered versus target value functionality of recovered to pre-flood condition with of 200 facilities infrastructure essential project support (Number) for public services and economic recovery in Direct project beneficiaries under 173,765 beneficiaries of rehabilitated flood affected areas Component 2 regional infrastructure versus target value of 120,000 840,995 beneficiaries of rehabilitated local infrastructure versus target value of 300,000 (A)+(B) Direct project beneficiaries, and percentage 868,895 direct beneficiaries versus target of female value of 300,000 51% female benefitted Page 13 of 55 The World Bank BiH Floods Emergency Recovery Project (P151157) The achievement of each outcome is further discussed in the following. The achievement of Outcome A - critical needs met in flood affected areas – depends on two main factors: firstly, the contribution of the project to making available eligible emergency goods in flood affected areas, and, secondly, the extent to which this output contributed to meet the overall critical needs in those areas8. According to the RNA, emergency needs in BiH were estimated at about US$153 million, with US$120 million made already available by early June 2014 through international cooperation, UN agencies, and private contributions and remittances. Under component 1, FERP planned to finance US$30 million emergency goods which would cover most of the remaining needs. Although the PDO indicator related to “Emergency goods distributed by entity-level strategic reserves (US$)” was not fully achieved (US$27 million emergency goods distributed versus target value of US$30 million), the efficacy of outcome A is considered high. During the implementation of the project, new evidence from the ground showed that the amount initially allocated for Component 1 exceeded the needs, and therefore the unused funds were reallocated to Component 2 for rehabilitation of infrastructure as part of the formal restructuring processed in January 2016. The project procured a wide range of needed emergency goods following a positive list that included the items described in Table 6 below and allowed 40 percent more people than initially estimated to benefit from activities under this component as per the PDO indicator addressing the number of direct project beneficiaries (104,537 beneficiaries versus target value of 75,000). According to the RNA results, nearly 90,000 people were displaced as a result of the 2014 floods, and the total population in disaster affected municipalities was about 2,500,000. These data show that the amount of emergency goods financed under FERP had the potential to reach a significant part of the most severely and directly affected population. However, to evaluate whether the project met critical needs, it is necessary to verify whether the emergency goods reached the intended beneficiaries. External audits conducted in the FBiH, RS and BD in 2017 by a third party9 concluded that the emergency goods purchased with FERP’s funds had reached the final intended beneficiaries. These intended beneficiaries included registered farmers and agriculture producers that had been affected by the floods (26 percent of affected farms benefitted from project’s support), emergency staff in flooded cities who undertook the distribution of goods to the population and/or were engaged in search and rescue operations, and other populations affected by the floods including displaced people. This evidence demonstrates that the project contributed to meet the overall critical needs in flood affected areas by making eligible emergency goods available to the populations in need. There were delays in the delivery of some goods in the agriculture sector as explained below in the Efficiency section. However, according to the RNA, in the immediate aftermath of the flooding, the supply chain operated with difficulty and interruptions but quickly recovered. Agriculture related inputs and services were available in many cases through delayed payment terms which helped farmers to recover. Under Component 1, 46 percent of FERP’s expenditures corresponded to the reimbursement of goods purchased during the emergency phase, while the remaining expenditure corresponded to emergency goods that the Governments purchased to replenish stocks depleted during the emergency or to continue helping affected people. This indicates that the delays incurred under Component 1 affected the project’s efficiency but did not jeopardize the achievement of the project’s outcome. 8Given the reduction in the PDO indicator target as part of the first project restructuring, a split evaluation was considered. However, since only 33% was disbursed at the time of restructuring, it will not have any impact on the outcome rating. 9 Grant Thornton S.A. Athens, Greece, in consortium with Grant Thornton OOD, Bulgaria, and Grant Thornton d.o.o. Banja Luka Page 14 of 55 The World Bank BiH Floods Emergency Recovery Project (P151157) Table 6: Emergency goods procured under component 1 Part Responsible Institutions Procured Emergency Goods Intermediate Results Part A FBiH Ministry of - Construction materials Total for Parts A, B and C: Federation of Displaced Persons and - Agriculture equipment 100% of emergency disaster Bosnia and Refugees - Reimbursement of eligible goods recovery goods purchased in Herzegovina line with the established (FBiH) FBiH Ministry of - Agriculture mechanization and equipment schedule Agriculture, Water - Corn, soy, tobacco, onion, wheat, potato, beans 26% of affected farms Resources and Forestry seeds benefitted from Government - Tomato, pepper saplings support versus 20% target - Eggplant, melon, watermelon, strawberry cucumber seedlings - Fruits seedlings (plum, apple, pear, cherry, walnut, hazelnut, raspberry) - Beehives - Greenhouses FBiH Directorate for - Diesel fuel Strategic Reserves - Wheat seed - Mineral fertilizer FBiH Administration for - Search and rescue equipment Civil Protection Part B Solidarity Fund - Construction materials, furniture, house Republika Srpska appliances, among other goods needed for (RS), and reconstruction of damaged property Ministry of Interior - Logistical equipment Part C Brčko District Civil Protection - Ambulances (BD) Administration - Rescue equipment - Terrain vehicles - Rescue boats RS Strategic Reserves - Refinancing of commodity reserves (seeds, animal feed, diesel fuel, food, water, among other goods) Ministry of Agriculture, - Fuel Forestry and Water - Livestock Management - Seeds, seedlings Department for Health - Hot meals and Other Services - Medicines, vaccines Department of Public - Hygiene products Security - Boots - Milk, hot meals - Raincoats - Boat engines - Among others Page 15 of 55 The World Bank BiH Floods Emergency Recovery Project (P151157) The achievement of Outcome B – restored functionality of infrastructure essential for public services and economic recovery in flood affected areas – is demonstrated by the actual number of infrastructure facilities recovered with project support, and the appropriateness of the criteria used to allocate funds and prioritize facilities in flooded areas. These considerations are instrumental to make sure that infrastructure rehabilitated under the project is essential for public services and economic recovery. The actual number of infrastructure facilities recovered with project support exceeded initial estimates. Under Component 2, the number of public service infrastructure facilities recovered to pre-flood condition with FERP support more than doubled the estimated end target (446 facilities were recovered as compared to the initial target of 200 facilities), and thus enabled a larger number of people to benefit from the restored infrastructure. The number of direct beneficiaries from the rehabilitated regional infrastructure reached 173,765 people (almost 30 percent more than target value of 120,000), while the number of people who benefited from rehabilitated local infrastructure reached 840,995 (more than 2.5 times the target value of 300,000). Table 7 lists the types of infrastructure rehabilitated under the project both at regional and local levels as well as the values of intermediate result indicators for Component 2. Final values of most intermediate result indicators exceeded the planned targets, including indicators on the number of piped household water connections that are benefitting from rehabilitation works (end value is almost four times higher than the planned end target), kilometers of roads rehabilitated (final value is three times higher than the projected end target in rural areas and two times higher in non-rural areas), number of landslide rehabilitation activities (final value exceeded almost four times the end target), and the number of public infrastructure facilities rehabilitated (exceeded 50 percent of the end target). Thus, the project helped to rehabilitate more infrastructure than initially planned. Page 16 of 55 The World Bank BiH Floods Emergency Recovery Project (P151157) Table 7: Infrastructure rehabilitated under component 2 Part Sub-component Infrastructure Rehabilitated Intermediate and Other Results Part A 2A. Rehabilitating or - Main roads in JP Ceste FBiH, including 39 municipalities benefitted Federation of reconstructing high rehabilitation of landslides, retaining Bosnia and priority, regional-level walls and bridges Co-financing from 16 municipalities Herzegovina public infrastructure - Railways in JP Željeznice FBiH, in the total amount of US$ 330,000 (FBiH) through Sub-projects in including rehabilitation of landslides, Flood Affected Areas bridges, power plant, construction of 100% of targeted municipalities culverts have identified, consulted and - Flood protection works (Sava River prepared final lists of priority local Basin Agency), including service activities for funding under reconstruction of dykes the project according to project criteria 2B. Rehabilitating or - Rehabilitation or reconstruction of reconstructing high local roads, sewerage systems, Ministries have presented final lists priority, local-level public bridges, landslides, retaining walls, of priority activities according to service delivery - Flood protection works, including project criteria for 93% of regional infrastructure through reconstruction of dykes activities budgeted Sub-projects in Flood - Among others Affected Areas 58,109 piped household water Part B 2A. Rehabilitating or - Rehabilitation of roads, including connections benefitted from Republika Srpska reconstructing high bridges, landslides rehabilitation works versus target (RS), and priority, regional-level - Flood protection works, including value of 15,000 public infrastructure rehabilitation of dykes, cleaning of Part C through Sub-projects in riverbeds 263.58 km of rural roads Brčko District Floods Affected Areas - Rehabilitation of Doboj Penitentiary rehabilitated versus target value of (BD) - Construction of Doboj Police Station 80 km 2B. Rehabilitating or - Rehabilitation or reconstruction of 63.45 km of non-rural roads reconstructing high local roads, bridges, landslides, water rehabilitated versus target value of priority, local-level public supply systems, sewage systems 25 km service delivery - Flood protection works, including infrastructure through rehabilitation and reconstruction of 35 bridges rehabilitated versus Sub-projects in Floods riverbanks, cleaning and widening of target value of 15 Affected Areas riverbeds - Among others 53,460 m of flood protection infrastructure rehabilitated or reconstructed versus target value of 50,000 m 56 landslides rehabilitated versus target value of 15 19 public facilities rehabilitated (schools, health facilities, priority public buildings, draining ditches) versus target value of 10 Page 17 of 55 The World Bank BiH Floods Emergency Recovery Project (P151157) To evaluate whether this rehabilitated infrastructure was essential for public services and economic recovery, it is important to discuss the eligibility criteria introduced under Component 2 as well as the importance of this infrastructure for the economy in BiH. The project’s eligibility criteria excluded any activity involving rehabilitation, repair or replacement of infrastructure not directly affected by the 2014 floods. Moreover, all activities financed under Component 2 had to be aligned with the RNA’s conclusions. To enable prompt implementation of civil works, there were other criteria that excluded sub- projects requiring long preparation processes, resolving ownership related issues or complex infrastructure projects with significant environmental impacts. All activities to be financed had to benefit from proper technical designs based on “build back better” principles, to secure resilience to future similar events and sustainability of the rehabilitated infrastructure in the long-term. Under sub-component 2B, the selection of most affected municipalities was done based on the RNA’s economic loss per capita estimation, and the funds were assigned proportionally to the overall recovery needs, with a correction factor for municipalities’ poverty level and capacity to generate revenues. All project activities were to comply with these conditions as they were screened against the eligibility criteria by the PIUs. The transport sector in BiH is recognized as a key sector for the country’s economic growth. The transport and flood management sectors benefitted the most from the activities under Component 2. Even before the 2014 floods, transport infrastructure in BiH required upgrades in line with the agreed comprehensive network of the South East Europe Transport Observatory Network (SEETO)10. An important objective for BiH was to improve connectivity as a means for emerging markets to integrate into the global economy. This would expand opportunities for business in BiH and promote job creation. Road transport in BiH accounts for 95% of all goods and passenger movement. The poor condition of local level roads and related structures such as bridges impacts, for instance, the agriculture sector by increasing transportation costs and affecting the quality of the products transported such as fruits, vegetables and milk. This subsequently affects production and sales prices throughout the supply chain. Considering that the agricultural sector represents nearly 10 percent of GDP and 20 percent of overall employment and knowing that severely affected flooded areas across central and Northeastern BiH had been historically oriented towards agriculture and agribusiness, it is reasonable to conclude that rehabilitated infrastructure contributed to economic recovery and growth. Railways, on the other hand, have historically functioned as the predominant transport mode supporting heavy industry including mining and quarrying industries. Based on the above, it is possible to conclude that, through investments on railways and roads, including both rural and non-rural road network, FERP targeted key economic sectors, while through investments on flood management, FERP built resilience to future floods. Rural connectivity in particular, benefitted from rehabilitation of more than 260 Km of rural roads. In terms of flood management, 53,460 m of flood protection infrastructure was rehabilitated or reconstructed under FERP (as compared with the target value of 50,000 m). Therefore, this suggests that FERP achieved its intended outcome of restoring functionality of infrastructure essential for public services and economic recovery in flood affected areas. 10 The SEETO Comprehensive Network is a multimodal regional transport network defined under the MoU, supported by the European Commission. It contemplates main and ancillary transport infrastructure in the South East Europe (SEE) as the base for the implementation of the transport investment programs. Within the revision process of the TEN-T network, the SEETO Comprehensive Network maps are included in the TEN-T Guidelines, where they appeared as indicative, and moreover, the SEETO Comprehensive Network was defined as TEN-T Comprehensive Network in the South East Europe and interlinks were determined. In this way, a clear perspective is given to become fully a part of the wider TEN-T network through the individual accession processes of Regional Participants. Page 18 of 55 The World Bank BiH Floods Emergency Recovery Project (P151157) Justification of Overall Efficacy Rating The project’s overall efficacy is rated “High” since the project fully achieved its intended outcomes 11. Activities financed by the project addressed the needs of public sectors that concentrated nearly 60% of the recovery and reconstruction needs as per the RNA, including agriculture, flood protection, transport, and public services and facilities. Agriculture and the transport sectors are key sectors of the economy in BiH. The project has used a positive list of goods and robust eligibility criteria to make sure that it would address critical emergency needs and public infrastructure damaged by the floods. An external audit conducted by a third party confirmed that emergency goods reached the intended beneficiaries. These beneficiaries included farmers and agriculture producers whose farms were damaged by the floods. They received seeds, seedlings, agriculture equipment and other goods needed to resume production. Displaced and other people directly affected by the floods received emergency goods such as hot meals, vaccines, and hygiene products. Reconstruction goods such as construction materials were provided to people whose houses had been damaged by the floods. Equipment damaged in the aftermath of the floods during search and rescue operations was replaced by new equipment. The activities related to rehabilitation of public infrastructure were prioritized based on the detailed damage assessment conducted after the floods and reported in the RNA which ensured only damaged infrastructure would be financed. Rehabilitated infrastructure focused on the transport sector which is a key sector for public services delivery and BiH’s economy. Designs were based on the principles of “build back better” which along with the financed flood protection works promotes long-term sustainability of the project’s outcomes. This evidence demonstrates that the project met critical needs and restored functionality of infrastructure essential for public services and economic recovery in flood affected areas which are the project’s intended outcomes. C. EFFICIENCY Assessment of Efficiency and Rating Efficiency Rating: Substantial Although the ERR calculated at completion allows for evaluating the efficiency of one of the major activities supported by FERP (road rehabilitation), it is insufficient to evaluate the overall efficiency of the project. In this context, aspects of design and implementation made an instrumental contribution to the overall project efficiency. In the following, the efficiency of the project is analyzed per component considering how efficient each component was in contributing to the overall achievement of the project outcomes and longer-term objectives. Specific aspects of design and implementation that either contributed to or reduced efficiency are also considered. Under Component 2 rehabilitation of roads was one of the main outputs. The respective ERR estimated at completion indicates that the efficiency of some activities exceeded expectations . Considering that specific activities to be financed by the project were not defined at appraisal, the Economic Rate of Return (ERR) was not calculated for the project. Nonetheless, an ERR of 22 percent over 20 years was estimated for rehabilitation/reconstruction of secondary roads. The economic analysis conducted at completion (Annex 4) indicates that the annual benefit from reduced travel time on rehabilitated roads is US$30 million (as opposed to USD$6 million estimated at appraisal) with an ERR of 15 percent in five years (29 percent in fifteen years). 11Photos and additional information about the infrastructure rehabilitated under FERP can be found in the following link; https://www.worldbank.org/en/news/video/2019/12/06/building-back-better-in-bosnia-and-herzegovina Page 19 of 55 The World Bank BiH Floods Emergency Recovery Project (P151157) Investing in affected infrastructure through Component 2 is likely to have contributed to long-term growth in BiH. The contribution of infrastructure to growth is typically around 8 to 20 percent12. In the context of BiH, a conservative estimate of around 8 percent seems to be adequate. Investing in infrastructure seems therefore to be an efficient investment when compared to alternative investments. Anecdotal evidence from Sandski Most in FBiH further demonstrates Component 2’s contribution to long-term growth. An important foreign investor from the aeronautic industry told local authorities that the quality of their infrastructure had been instrumental for the selection of a site in this municipality for investment. The application of the principles of building back better during the implementation of Component 2 contributed to improve resilience in the affected areas by reducing the expected average annual losses due to floods in the future. The flood protection works financed by the project improved resilience of infrastructure in those areas (flood protection works were designed for a 100-year return period flood while pre-disaster floods were typically designed for shorter return periods). The Net Present Value (NPV) associated to the improvement in resilience in the affected areas is estimated at US$50 million over 20 years discounted at 5 percent (see Annex 4 for details). Although improving resilience in the affected areas was not a project outcome, this is another valid indicator of the project’s efficiency. Ownership and co-financing at the local level for rehabilitation of local level public infrastructure proved to be an efficient solution. Within the context of an emergency recovery project with many concurrent priorities which go beyond the financial capacity of the project, the condition imposed by FERP to local municipalities in which financing would be provided only for projects with prior completion of project designs had a positive contribution to the project’s efficacy and efficiency. In terms of the project’s efficacy, project funds were not spent on feasibility studies and detailed designs which saved resources for execution of a larger number of works and therefore benefitted more people. By financing works with completed designs, FERP also ensured that only works deemed as feasible would be financed which accelerated the implementation of Component 2B. Moreover, accountability at the local level also fostered action and mechanisms to overcome barriers and advance the implementation. Activities managed at entity-level for rehabilitation of key infrastructure reduced the overall efficiency of the project as these activities experienced a low disbursement pace that led to the need to extend the project closing date for 6 months. Activities under Component 2A aimed primarily to restore connectivity at the regional level. This sub-component experienced the most challenging implementation arrangements since the project relied on entity- level ministries responsible in each sector to reach out to beneficiaries. The delays in this sub-component led to the need for the second project restructuring approved on 12-Dec-2018 in which the loan closing date was extended by six months, from December 31, 2018 to June 30, 2019. These delays are not directly linked to the project design and implementation, as the main cause is related to governance aspects. Another factor that caused delays in the implementation of Component 2 related to changes in credit conditions for municipalities in the RS. The project funds for implementation of Component 2B were initially planned to be provided to municipalities in the form of a grant in both the FBiH and RS. This decision changed during implementation to a credit line in the RS. Although this change resulted in approximately a 1-year delay in implementation of activities under Component 2B in the RS, the PIU was thereafter able to accelerate the implementation and complete local level activities by the project’s initial closing date. 12 Bom, PRD and JE Lighthart (2009): How productive is Public Capital? A meta-analysis, Georgia State University, Andrew Young School of Policy studies, International Studies program working paper 09-12, quoted in: The impact of infrastructures on growth in developing countries, IFC economic notes, April 2012. Page 20 of 55 The World Bank BiH Floods Emergency Recovery Project (P151157) Considering the economic structure in BiH, transactions provided under Component 1 are likely to have stimulated the economy while limiting disruption in productive sectors. While there is not enough evidence available to evaluate whether the transfers made to beneficiaries through a positive list (vouchers) rendered more efficient outcomes than other types of transactions, the positive list is likely to have significantly contributed to ensuring that the project finances the right priorities. The overall efficiency of Component 1 in meeting critical needs in flood affected areas and contributing to restore BiH to a growth path can be discussed by considering the macroeconomic value of an early resumption of consumption. Considering the economic structure in BiH, the use of voucher transactions under this component is likely to have stimulated the economy and limited disruption in productive sectors. This, on the other hand, made available the necessary workforce to support the recovery and reconstruction. Delivery of goods in the agriculture sector under Component 1 underwent noticeable delays. One year after the floods, a significant part of the agricultural support had not reached the final beneficiaries. The causes for these delays concern poor coordination among line ministries, agencies and municipalities as well as difficulties in devising and implementing mechanisms to ensure goods would reach the intended beneficiaries. These delays reduced efficiency even though they could have prevented inadequate allocation of funds since without those quality assurance mechanisms in place the goods could have reached the wrong beneficiaries. Allocation of funds among entities was consistent with the results of the RNA which indicates that the project’s resources were efficiently used throughout flood affected areas independently of the administrative-territorial boundaries. Furthermore, in the aftermath of the floods, prices were homogeneous throughout entities and BD without price distortions leading to high nominal prices. FBiH and RS received the same share of the project funds (47.5 percent in FBiH and 47.5 percent in RS) while BD received the remaining 5 percent of funds. This fund allocation is aligned with the needs assessed through the RNA which was completed shortly after appraisal. According to the RNA, 51 percent of the total damages and losses occurred in FBiH while 48 percent occurred in RS and 1 percent in BD. However, data on the relative impact within each entity indicates that the floods had the greatest impact on RS with more than 5 percent of the population displaced (against 2.4 percent in FBiH and 1.3 percent in BD). At the sectoral level, this impact on RS is even more evident. For instance, in the agricultural sector, which was one of the most affected by the floods, almost 52 percent of damages and 80 percent of losses were concentrated in RS. In terms of value for money, prices were homogeneous throughout entities, as expected in a connected market, and no price distortion leading to high nominal prices was recorded. These considerations are valid indicators of efficiency. However, they are insufficient to demonstrate the efficiency of the project in achieving its outcomes and long-term objectives as per the theory of change. The extent to which a need-based fund allocation fosters growth does not depend only on the actual needs, since for instance least affected areas (hence with minor needs) might have higher capacity to convert project inputs (transfers or infrastructure investments) into growth. Timely reallocation of funds from Component 1 to Component 2 allowed to more efficiently respond to the actual needs in BiH during the reconstruction phase. As the post-disaster situation evolved from the recovery to the reconstruction phase, the project’s funds were used to cover critical needs through Component 1 at an early stage of the project, and during the reconstruction phase the project financed rehabilitation of infrastructure through Component 2. Reallocation of US$7.4 million from Component 1 “Emergency Disaster Recovery Goods” to Component 2 “Rehabilitation of Key Public Infrastructure” helped to improve efficiency because it made available additional funds for rehabilitation of infrastructure which was the most pressing need in the country at the time of funds’ reallocation. By this time, the project had already achieved its first intended outcome of meeting critical needs in the flood affected areas with the number of direct beneficiaries exceeding the end target. The reallocated funds Page 21 of 55 The World Bank BiH Floods Emergency Recovery Project (P151157) therefore allowed to mobilize additional financing to make progress towards achieving the second intended outcome of recovering infrastructure essential to public services and economic recovery. As a result, the project helped to rehabilitate more infrastructure than initially planned as discussed under the Efficacy section. Implementation arrangements for the BD were adjusted to accelerate implementation and, thus, contributed to efficiency. When procurement tasks for the BD began to be managed directly from the BD PIT with the hiring of one dedicated staff for procurement (initially procurement for BD was managed in the RS), the implementation in the BD immediately accelerated. This fact had also a positive impact in the RS with the two dedicated procurement staff being able to focus on the works implemented in the RS. With this arrangement in place, paperwork no longer had to be sent by car to Banja Luka in the RS and a more streamlined clearance process was in place. The project’s efficiency is rated “Substantial” since the project demonstrated efficiency that would be expected in a post-disaster context and in sectors involving participation of line ministries, municipalities and other authorities managing both regional and local infrastructure in BiH. The overall rating on implementation progress was modified during the initial year of implementation from satisfactory to moderately satisfactory and changed again to satisfactory in early 2017 to address the fact that the implementation was slower than anticipated between certain periods of time but later recovered to a satisfactory pace. Nonetheless, the discussion above demonstrates that factors contributing to reduce efficiency are mostly related to governance aspects in BiH. The project was designed and implemented in a way that contributed to improve efficiency despite the large number of institutions involved and all the difficulties that arise during the aftermath of disasters. The need to extend the closing date of the project was mainly related to implementation of regional activities under Component 2A which required coordination among line ministries and institutions managing this infrastructure. D. JUSTIFICATION OF OVERALL OUTCOME RATING The overall outcome rating is considered “Highly Satisfactory”. The evidence analyzed for the ICR demonstrates that the project relevance and efficacy are both rated “High”, while there were only minor shortcomings in the operation’s efficiency that is rated “Substantial”. Although emergency recovery projects should deliver faster than non-emergency projects, FERP included activities that are typical of a non-emergency project. These activities included infrastructure projects following the principles of “Building Back Better” and other activities contributing to capacity building in disaster risk management. The main activity which required the 6-month extension of FERP represented a very minor share out of the total project amount. Moreover, part of the emergency goods financed by FERP were paid for retroactively, and therefore the delay (which was shorter than similar projects in Bosnia and Herzegovina) had minor impacts. As a result, the shortcomings identified in the project’s efficiency were considered minor, and the overall Outcome rating is considered as Highly Satisfactory. E. OTHER OUTCOMES AND IMPACTS (IF ANY) Gender The project benefitted the entire population affected by the 2014 floods equally, including both women and men. The percentage of female beneficiaries was similar to the percentage of benefited men (about 50 percent). The 51 percent target was therefore not achieved as initially intended. This notwithstanding, the project benefitted more women than initially estimated. A total of 435,000 women benefitted from the project as opposed to the initial estimate of 153,000 (i.e. 51 percent of 300,000 direct project beneficiaries). There is not enough evidence of the Page 22 of 55 The World Bank BiH Floods Emergency Recovery Project (P151157) project’s positive impact in closing relevant gaps between males and females or contributing to increasing women’s assets, income earning and employment opportunities. Institutional Strengthening Under Component 3 of the project, 137 public officials were trained on disaster risk management aspects, including search and rescue activities. Component 3B also supported a multi-hazard risk assessment for the BD which was adopted by the respective government in 2018. This risk assessment aims to support integration of risk reduction considerations into land use plans and other instruments as well as inform policy reforms to mainstream disaster risk management in the BD. Mobilizing Private Sector Financing Although there was not direct mobilization of private sector financing for implementation of the project’s activities, the investments made in rehabilitation of infrastructure are likely to promote private sector investments. As pointed out in the Efficiency section, a foreign investor from the aeronautic industry for instance told local authorities that the quality of their infrastructure was instrumental for the selection of a site for investment in one of the beneficiary municipalities under FERP. Poverty Reduction and Shared Prosperity Under Subcomponent 2B on local infrastructure, the funds were assigned proportionally to the overall recovery needs with a correction factor for municipalities’ poverty level and capacity to generate revenues. By prioritizing the poorest flood affected municipalities and therefore targeting the poor, the project might have had positive effect on poverty. This positive effect could have been superseded by the possible leakage of benefits to other groups which had incurred more losses because of the floods. Other Unintended Outcomes and Impacts Building resilience to floods in BiH. The application of the principles of building back better contributed to raising awareness on disaster risk management for the first time in the disaster affected areas. Feedback received by project stakeholders and beneficiaries during consultations held in March 2019 showed that there is an overall sense that public infrastructure is more resilient than its condition prior to the May 2014 floods. A sufficient bulk of activities executed under Component 2 aimed to rehabilitate or reconstruct flood protection structures such as dykes. These flood protection structures were rehabilitated/reconstructed based on designs that were envisaged to operate for floods with higher return periods as compared to the original structures that were in place before the floods. Moreover, the construction of culverts and cleaning and widening of riverbeds is further expected to contribute to improve resilience in the areas impacted by the 2014 floods. In this context, the project contributed to improve adaptation to climate-change which may increase the frequency of occurrence of flood events. Page 23 of 55 The World Bank BiH Floods Emergency Recovery Project (P151157) III. KEY FACTORS THAT AFFECTED IMPLEMENTATION AND OUTCOME A. KEY FACTORS DURING PREPARATION The following key factors during preparation affected the project implementation and outcome:  Simple design and readiness for implementation had a positive impact on the project outcomes: given the importance of delivering quick results, the implementation was based on existing structures, mechanisms and capacities built up through ongoing Bank projects at the time of appraisal.  Lack of technical and procurement capacity of PIUs to simultaneously manage a large number of sub- projects reduced project efficiency. Both the PIU in the FBiH and the PIU in the RS had experience with Bank financed projects. However, their capacity to manage hundreds of flood management and infrastructure rehabilitation sub-projects was low compared to the demand. In the case of the BD, it was the first time a Bank financed project was implemented in the District. PIUs were strengthened later during implementation. B. KEY FACTORS DURING IMPLEMENTATION The following key factors during implementation affected the project implementation and outcome: (a) Factors subject to the control of the government and/or implementing entities  Effective coordination and engagement between PIUs and local municipalities. Especially for implementation of Component 2B on rehabilitation of key public infrastructure at the local level, effective communication between the PIUs and municipalities resulted in an effective cooperation between project stakeholders which proved to be fundamental for timely processing of project applications, supervision and procurement and payment of works.  Co-financing and ownership at local level facilitated implementation of Component 2B. Local governments were key stakeholders throughout implementation of FERP, from co-financing of rehabilitation works and selection of Component 2B sub-projects to participation in evaluation committees, communication with beneficiaries and issuance of permits.  Lack of human resources and slow procurement processes. Design preparation and tendering progress advanced slower than anticipated in both entities and the District. Procurement tasks for the BD for instance were initially managed in RS which caused delays and administrative burden in RS. Once procurement for BD was managed directly from the BD PIT, the implementation in the BD immediately accelerated.  Changes in credit conditions for municipalities in the RS caused significant delays in the implementation of Component 2. The RS Ministry of Finance decided to transfer FERP resources to municipalities as IDA term credits rather than grants. The fact that initially the funds under Component 2 were planned to be provided as grants in the RS and later this decision was changed to a credit line, resulted in approximately a 1-year delay in implementation of activities under Component 2B in the RS. The sub-credit agreements were signed towards the end of 2015. Page 24 of 55 The World Bank BiH Floods Emergency Recovery Project (P151157)  Technical quality assurance mechanisms strengthened during implementation. PIUs hired civil engineers who provided support to participating municipalities, reviewed designs of sub-projects submitted by municipalities, and oversaw the implementation. Moreover, PIUs have outsourced the design and/or supervision of works to third parties when municipalities did not have sufficient capacity to conduct these activities.  The requirement for riparian notification for works along the Sava River explains some of the delays in the implementation of flood protection works in the FBiH. Although an Exception to Riparian Notification under OP 7.50 on International Waterways was granted by the World Bank’s Vice President during the preparation of the project, riparian countries were notified by the FBiH about the sub-project that involved raising the dyke’s height of the Sava River due to the potential implications those works could have over the entire river basin. The RS assigned funds towards other activities which did not require notifying riparian countries. (b) Factors subject to the control of the World Bank  Adequacy of supervision: Bank TTL-ship was kept consistent throughout the project implementation. The task team was proactive in providing adequate following-up and technical advice through a multi- disciplinary team of specialists from the relevant fields.  Mobilization of additional grant support to build institutional capacity in disaster risk management : The Bank mobilized grant support to complement activities under Component 3B aimed at enhancing institutional capacity in disaster risk management. This allowed funds under Component 3B to be used for equipment for post-disaster response situations. IV. BANK PERFORMANCE, COMPLIANCE ISSUES, AND RISK TO DEVELOPMENT OUTCOME A. QUALITY OF MONITORING AND EVALUATION (M&E) M&E Design M&E design was aligned with the PDO and effective arrangements were in place to monitor progress towards the PDOs through adequate indicators. The PDO was clearly specified and well-aligned with the project design. The results framework was adequately used to monitor progress towards the project’s outcomes. The M&E mechanisms could have included indicators to track the project’s contribution to the overall economic recovery. Considering that the project followed the principles of “building back better,” the indicator on public service infrastructure facilities could have quantified not only the facilities recovered to pre-flood condition but the ones recovered with improved resilience. Finally, M&E data could have systematically measured women’s participation and involvement. For instance, intermediate indicators such as number of public officials trained in disaster risk management could have been disaggregated by gender. M&E Implementation M&E data was adequately collected and analyzed. Although end target values of the PDO indicators were only determined during the first year of the project’s implementation with the results framework finalized through the first project restructuring in January 2016, M&E data was adequately collected and systematically analyzed starting Page 25 of 55 The World Bank BiH Floods Emergency Recovery Project (P151157) from the second year (2015) until project completion. The methodology used to calculate the indicators addressed the issue of double counting. For instance, in cases where a settlement benefitted from both a bridge repair and flood protection works, the number of direct beneficiaries was not counted more than once. Both the PIU at FBiH and APCU at RS have assigned dedicated staff for the M&E of the project. However, activity forms have not always been systematically provided by line ministries and municipalities. It should be noted that the PIUs have adequately reported on the progress achieved in each of the sub-projects in their semi-annual implementation status reports, even though not all semi-annual reports were submitted in a timely manner. M&E Utilization M&E was utilized to track project implementation progress, demonstrate results and inform reallocation of funds. Instruments such as ISRs reported adequately the Project’s progress, and consistently illustrated this progress through the overall Project ratings. M&E data was used to inform reallocation of funds for instance from Component 1 to Component 2 to ensure a more efficient use of Project’s resources. Justification of Overall Rating of Quality of M&E The overall rating of M&E quality is considered “Substantial”. The M&E system as designed and implemented was robust and sufficient to assess the achievement of the objectives and ensure a more efficient use of Project’s resources, but there are moderate weaknesses in a few areas as described above. B. ENVIRONMENTAL, SOCIAL, AND FIDUCIARY COMPLIANCE  Environmental and Social Safeguards. The project was classified as a Category B project, triggering the Operational Policies OP 4.01 on Environmental Assessment and OP 4.12 on Involuntary Resettlement. The project triggered the International Waterways policy OP/BP 7.50, but was processed under an Exception to Riparian Notification requirement under paragraph 7(a) of OP 7.50, Projects on International Waterways. Despite this exception, the FBiH notified riparian countries through the International Sava River Commission on the proposed works along the Sava River aimed at raising the river’s embankments, and another notification was issued for works along the Bosna river. Responses were received from Slovenia, Croatia and Serbia with no objections or issues noted. The project had an Environmental and Social Management Framework (ESMF) which were applied to all sub-projects identified within the FERP. Each of the sub-projects financed under the project underwent an environmental screening as per the checklist within the ESMF and an EMP where applicable. Environmental and social safeguards compliance remained satisfactory throughout implementation.  Environmental issues. There were reported issues related to solid waste management of the project’s sites as well as cases where the requirements of ESMF have not been fully addressed, in particular the requirements related to workplace safety. These issues were addressed by strengthening the mechanisms in place to ensure that ESMPs were included in bidding documents for civil works and supervision, as well as by supporting municipalities to check compliance with all provisions including disclosure of ESMPs and screening checklists.  Social issues. No cases of permanent acquisition of private land were reported. The site-specific checklists have been generally followed adequately. For civil works requiring temporary expropriation of private land plots for the duration of the corresponding contract, the respective implementing authorities were required to submit proof that land acquisition had been adequately addressed along with the respective signed Page 26 of 55 The World Bank BiH Floods Emergency Recovery Project (P151157) agreements. PIUs kept copies of all agreements signed by respective landowners and authorities. The contracts for execution of civil works could only be initiated after the PIUs received the evidence that land acquisition issues had been adequately addressed.  Grievance Redress Mechanism (GRM). Project-affected people were aware of the GRM mechanism. Given the positive impact related to local and regional flood protection investments and with no permanent land acquisition taking place, citizens have mostly expressed satisfaction with the rehabilitation works carried out. Only one complaint was received to date and it was timely and successfully resolved. The World Bank received a complaint concerning construction of a bridge in the RS in which community members expressed their concern with the design of the draining system. There were concerns that the proposed design would increase the likelihood of floods. The PIU together with representatives of the city of Banja Luka, designer, constructor and supervisor teams met with the concerned community members. The solution found to solve the issue consisted of construction of manually operated gates to regulate rainwaters flow. This was planned to be implemented through the EIB-financed Vrbanja River embankment works.  Procurement. FERP was able to reimburse goods purchased during the emergency phase, purchase goods to continue responding to the disaster, and replenish stocks that were depleted during the emergency using a positive list13. The PIUs in both entities were responsible for procurement in the respective entities. The PIU in the RS was responsible for procurement in the BD at the beginning of the project’s implementation. At a later stage, procurement in the BD was directly managed by a local unit. The two PIUs were under the Ministry of Agriculture, Forestry and Water Management which had procurement officers with experience in World Bank financed projects. There were reported procurement issues both in the FBiH and RS. In the FBiH, one of the main issues concerned the results of a bid evaluation for a railway sub-project under Component 2A. The procurement procedure was cancelled, and the funds reallocated in a process that took about one-and-half years. In the RS, there were complaints issued by a prospective bidder on a situation of conflict of interest which was promptly solved.  Financial management. Overall, both PIUs had adequate financial management systems, including accounting, budgeting, staffing, internal controls, counterpart funding, audit and financial reporting. Audit reports were timely submitted, including the interim un-audited financial reports (IFRs).  Consultations with relevant stakeholders. The project applied an overall highly participatory approach to prioritize regional and local infrastructure sub-projects. Consultations with affected municipalities were held since the beginning of the project. Joint meetings with relevant authorities were also held to inform them on the World Bank’s environmental and social safeguards provisions as well as to plan mitigation measures. C. BANK PERFORMANCE Quality at Entry To support the governments of BiH in the aftermath of the 2014 floods, the Bank prepared FERP in a short time period with the use of streamlined procedures for operations in situations of urgent need for assistance. This prompt 13A positive list is a list of eligible essential goods that can be retroactively financed. These goods could have been procured through national emergency procedures as set forth in the relevant laws of the Borrower or commercial practices of the private sector as set forth in the respective Project Operation Manuals. A technical and financial review of the goods procured were conducted for each expenditure which determined their acceptability to the World Bank. Page 27 of 55 The World Bank BiH Floods Emergency Recovery Project (P151157) response from the Bank was instrumental to achieve the project’s outcomes in meeting critical needs and restoring functionality of disaster affected infrastructure. As a standard practice for such type of emergency response operations, it was possible to adequately define specific activities, technical aspects and verify M&E arrangements at the stage of the project’s implementation. Environmental and social aspects however were adequately addressed at entry, which contributed to a smooth implementation overall. The Bank adequately assessed the main risks, including the need to strengthen the PIUs in their fiduciary and monitoring capacities. The Bank’s experience in emergency response projects around the world and in particular in the region was instrumental in incorporating the principles of “build back better” as part of FERP’s approach, so that the project would contribute to build resilience and strengthening disaster risk management’s capacity. Resilience aspects however could have been more explicitly incorporated into M&E arrangements. It should be noted that the Bank mobilized technical assistance funds during implementation to complement activities financed under Component 3B. Quality of Supervision The Bank mobilized and ensured continuity of a strong task team with different expertise, including technical, environmental and social safeguards and procurement. The Bank had environmental and social experts and engineers conducting missions and checking issues on the ground. The team leader role was handed over in mid- 2016 keeping the co-team leader from the beginning of the operation to the end of the project. Regular implementation support missions were undertaken throughout the project and the issues identified have been promptly addressed and adequately reported in aide memoires, ISRs and management letters. In general, the Bank acted proactively to identify issues and discuss solutions related to fiduciary and safeguard aspects, as well as funds reallocation when needed to achieve the intended project’s outcomes and facilitate implementation. The Bank assisted the PIU in the RS and BD in procurement aspects through the Bank’s local procurement consultant and staff. Justification of Overall Rating of Bank Performance The overall rating of the Bank’s performance is considered “Satisfactory” since there were only minor shortcomings in quality at entry and quality of supervision. At entry, the Bank could have better incorporated international experience building resilience in post-disaster contexts into the project’s design. During implementation, the Bank could have had a more proactive role at an early stage of the project in strengthening the implementation units in both entities and the District. D. RISK TO DEVELOPMENT OUTCOME The risk to FERP’s development outcomes is mainly related to the potential impact of future floods on communities and infrastructure benefitted under the project. BiH is prone to floods and the likelihood that floods of similar intensity to the 2014 will hit the country again is significant. The sustainability of the project’s outcomes as well as its long-term impact will depend on how resilient the infrastructure rehabilitated and reconstructed under the project is. Moreover, it will depend on building capacity of local institutions in disaster risk management. Although FERP did some important steps towards strengthening the institutional capacity, planning instruments in the country still have not incorporated these aspects and a strong effort is needed in the years to come to build this capacity in BiH. Building resilience to natural shocks is one of the focus areas of the CPF for FY16-20 and the Bank continues an active policy dialogue in the country on DRM. Page 28 of 55 The World Bank BiH Floods Emergency Recovery Project (P151157) V. LESSONS AND RECOMMENDATIONS Rapid project preparation with the use of streamlined procedures does not compromise the project’s implementation if a strong task team with experience in the country and in similar projects is mobilized and implementation issues are anticipated and addressed. FERP was prepared and implemented by an experienced task team which had a deep knowledge of the country context and policy environment. The team also included technical experts on flood management, landslides and rehabilitation of infrastructure that could effectively identify quality issues in specific sub-projects and support the implementation units in solving these issues. Efforts need to be made in emergency recovery projects to expedite project effectiveness for prompt assistance to emergency affected areas. FERP reached effectiveness in record time for projects in BiH (project became effective in September 2014, three months after signing). In other infrastructure projects in the country, the process towards effectiveness typically takes more than six months. The Borrower was able to timely obtain the approvals necessary for signing of the project agreements and subsidiary credit agreements. Moreover, the World Bank organized the signing promptly after receiving adequate authorizations from the Council of Ministers and the entity governments. After signing of the project agreement, the World Bank was proactive in requesting the Borrower to expedite project effectiveness through management letters. Flexible project design with strong M&E and the use of a positive list are effective ways to implement emergency response projects. However, having overall analysis of the impact of the disaster and the needs in the flood affected areas available during the design and early implementation stages of emergency projects promotes transparency in the allocation of funds and prioritization of activities among relevant stakeholders and decision makers. Using a positive list, FERP was able to reimburse goods purchased during the emergency phase, purchase goods to continue responding to the disaster, and replenish stocks that were depleted during the emergency through national emergency procurement procedures as set forth in the relevant laws of the Borrower or commercial practices of the private sector. While retroactive financing under Component 1 (emergency disaster recovery goods) and simplified procurement procedures during the early phase of implementation contributed to quickly delivery of emergency response goods to vast beneficiaries, other factors also significantly played an important role in expediting procurement of hundreds of packages under FERP. This includes utilization of existing project PIUs which had long experience in implementing projects financed by World Bank and other financiers, strengthening of the PIUs as needed, and strong and clear guidance through World Bank supervision by providing target dates to finish delivery of emergency recovery goods and to move onto reconstruction, among other factors. Through a framework approach and robust eligibility criteria with evidence-based prioritization, FERP could rehabilitate infrastructure essential for public services and economic recovery. However, strong M&E mechanisms had to be in place to ensure the project would reach the intended outcomes. In emergency response projects, decentralized implementation with participation of municipalities empowers authorities who work closely with the recipient beneficiaries to act and builds capacity on disaster risk management at the local level. Local authorities with proper support by local dedicated personal were well prepared to take an active role during FERP implementation and to communicate with beneficiaries and foster citizen engagement. Communication with local beneficiaries was effective under FERP (Component 2B) due to the proximity between the places where decisions on sub-projects were made (municipal assemblies) and project beneficiaries. In addition, local experienced staff directly reached out to the project beneficiaries explaining the Page 29 of 55 The World Bank BiH Floods Emergency Recovery Project (P151157) project objectives and decisions. These facts fostered engagement of beneficiaries which ultimately could have potentially contributed to a better prioritization of investments under FERP and strengthened capacity of municipalities on the principles of “building back better”. Accountability and municipal co-financing are efficient mechanisms to reach a larger number of beneficiaries, ensure feasibility of investments at an early stage of the project and contribute to effective project implementation. However, potential lack of capacity at the local level should be addressed by providing the necessary timely support to local authorities. Local governments under Component 2B “Rehabilitation or reconstruction of high priority, local-level public infrastructure” played a role of key stakeholders throughout implementation of FERP, from co-financing of rehabilitation works and selection of Component 2B sub-projects to participation in evaluation committees, communication with beneficiaries and implementation of works. Within the context of an emergency recovery project with many concurrent priorities which go beyond the financial capacity of the project, the FERP envisaged that the local municipalities would finance preparation of feasibility studies and detailed designs for priority infrastructure by themselves, and that had a positive impact on the project’s efficiency. This allowed for execution of a larger number of investments in rehabilitation of infrastructure and therefore, expanding the project coverage and benefitting more people. By financing works with completed designs, FERP also ensured that only works deemed as feasible would be financed which contributed to accelerate the implementation of Component 2B. Moreover, accountability at the local level also fostered action and mechanisms to overcome barriers and advance the implementation. Participation of local governments in project decisions promotes stakeholders buy-in and ownership as well as accountability at the local level which is critical in post-disaster situations when decisions must be rapidly made and enforced. Credit conditions under Bank financed operations are in general attractive to municipalities in BiH which have difficulties to access credit in the market. However, in the context of an emergency project a grant-based approach can be more efficient. Different entities had an opportunity to execute different approaches in terms of credit financing lines to the local authorities. As such, project funds for implementation of Component 2B were provided to municipalities in the RS under a credit basis as opposed to the FBiH in which grants were provided to municipalities. Initially the funds under Component 2 were planned to be provided as grants in the RS as well, and later the RS government decided to change this approach to a credit line mechanism. As a result of this decision, some municipalities decided not to participate in the project. During consultations with local municipalities from the RS held in March 4, 2019, in Banja Luka, municipalities affirmed their satisfaction with the credit conditions provided by their government under FERP, since for them it is typically difficult to access credit in the market (typically municipalities cannot borrow more than 18 percent of their annual budget). At the same time, change from the initially planned grant approach to a credit line approach that happened during project implementation resulted in approximately a 1-year delay in implementation of activities under Component 2B in the RS. The application of the principles of “build back better” during emergency response operations is not only a tool to improve the resilience of the rehabilitated infrastructure but also an effective way to raise awareness on disaster risk management. Although improving resilience in the affected areas was not a primary project outcome, the flood protection works financed by the project improved resilience of infrastructure in those areas (flood protection works were designed for a 100-year return period flood while pre-disaster floods were typically designed for shorter return periods). Feedback received by project stakeholders and beneficiaries during consultations held in March 2019 showed that there is an overall sense that public infrastructure intervened is more resilient than its condition prior to the May 2014 floods. Page 30 of 55 The World Bank BiH Floods Emergency Recovery Project (P151157) Implementation arrangements should seek to reduce administrative burden by providing all key implementation units with necessary mechanisms and executed power to implement activities. When procurement tasks for the BD began to be managed directly from the BD PIT with the hiring of one dedicated staff for procurement (initially procurement for BD was managed in the RS), the implementation in the BD immediately accelerated. This fact also had a positive impact in the RS with the two dedicated procurement staff being able to focus on the works implemented in the RS. With this arrangement in place, paperwork no longer had to be sent by car to Banja Luka in the RS and a more streamlined clearance process was in place. To improve efficiency in emergency recovery projects, procurement procedures can include cut-offs which would serve as a reference for re-allocation of funds from delayed activities to other priorities . Periodic cut-offs could be applied at specific steps during the implementation of activities (e.g. design of works or procurement of goods). This would allow to identify barriers in specific activities at an early stage and if needed to re-allocate funds to other priorities in which implementation can promptly advance. Effective communication between project entity level and local stakeholders is instrumental to deliver timely results. As confirmed by the roundtable discussions with project stakeholders and beneficiaries, effective communication between the ACPU/FBiH PIU and municipalities resulted in an effective cooperation between project stakeholders which proved to be fundamental for timely processing of project applications, supervision and procurement and payment of works. For emergency recovery projects in general, it is recommended to include explicit objectives on improving resilience in disaster affected areas to promote change in disaster risk management policies and build long-term sustainability for project’s outcomes. Although FERP contributed to build resilience in the disaster affected areas of BiH by applying the principles of building back better, the actual progress on resilience could have been better promoted and quantified if the project had established measurable indicators of resilience. This would potentially promote reforms on public policies and leverage recovery investments towards mainstreaming disaster risk management in countries prone to natural hazards and climate change. In the long-term, raising awareness of key stakeholders and civil society on disaster risk management is key to continue building resilience of flood-prone areas and effectively move from a reactive response to a preventive risk management approach. The municipalities consulted under FERP in March 2019 reiterated the need to continue raising awareness, since no specific studies aimed at mapping areas prone to hazards and understanding risk were developed to inform investments and policy reforms. In the RS, for instance, land use plans have not been updated along with the updates made in flood maps after the 2014 floods. Since illegal construction in flood prone areas has been pointed out by project stakeholders as one of the main factors of risk, public policy reforms along with investments on infrastructure are necessary to improve resilience. Current budget allocation for disaster risk management at local level is typically used for maintenance of infrastructure. Future efforts should seek to put in . place mechanisms that trigger comprehensive reforms and ensure long-term sustainability of investments. Page 31 of 55 The World Bank BiH Floods Emergency Recovery Project (P151157) ANNEX 1. RESULTS FRAMEWORK AND KEY OUTPUTS A. RESULTS INDICATORS A.1 PDO Indicators Objective/Outcome: To meet critical needs and restore functionality of infrastructure Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Emergency goods distributed Amount(USD) 0.00 30000000.00 27000000.00 by entity-level strategic reserves 15-Jun-2014 31-Dec-2018 30-Jun-2019 Comments (achievements against targets): Mostly Achieved: 90% of targeted emergency disaster recovery goods distributed by entity-level authorities Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Public service infrastructure Number 0.00 200.00 446.00 facilities recovered to pre- floods condition with Project 15-Jun-2014 31-Dec-2018 30-Jun-2019 support Page 32 of 55 The World Bank BiH Floods Emergency Recovery Project (P151157) Comments (achievements against targets): Fully achieved: 446 facilities recovered versus target value of 200 facilities which corresponds to an achievement rate of 220% Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Direct project beneficiaries Number 0.00 300000.00 868895.00 15-Jun-2014 31-Dec-2018 30-Jun-2019 Of which, Component 1 Number 0.00 75000.00 104537.00 15-Jun-2014 31-Dec-2018 30-Jun-2019 Of which, Component 2A Number 0.00 120000.00 173765.00 15-Jun-2014 31-Dec-2018 30-Jun-2019 Of which, Component 2B Number 0.00 300000.00 840995.00 15-Jun-2014 31-Dec-2018 30-Jun-2019 Female beneficiaries Percentage 0.00 51.00 51.00 Page 33 of 55 The World Bank BiH Floods Emergency Recovery Project (P151157) Comments (achievements against targets): Fully achieved: 868,895 direct beneficiaries versus target value of 300,000 which corresponds to an achievement rate of 350%. A.2 Intermediate Results Indicators Component: Component 1: Emergency Disaster Recovery Goods Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Emergency disaster recovery Percentage 0.00 100.00 100.00 goods purchased in line with the established schedule, 15-Jun-2014 31-Dec-2018 30-Jun-2019 percentage Comments (achievements against targets): Fully achieved: 100% of emergency disaster recovery goods purchased in line with the established schedule Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Percentage of affected farms Percentage 0.00 20.00 26.00 benefiting from Government support 15-Jun-2014 31-Dec-2018 30-Jun-2019 Page 34 of 55 The World Bank BiH Floods Emergency Recovery Project (P151157) Comments (achievements against targets): Fully achieved: 26% of affected farms benefited from Government support versus 20% target Component: Component 2: Rehabilitation of Key Public Infrastructure Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Percentage of targeted Percentage 0.00 100.00 100.00 municipalities that have identified, consulted and 15-Jun-2014 31-Dec-2018 30-Jun-2019 prepared final lists of priority local service activities Comments (achievements against targets): Fully achieved: 100% of targeted municipalities have identified, consulted and prepared final lists of priority local service activities for funding under the project according to project criteria Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Percentage of regional Percentage 0.00 100.00 93.00 Page 35 of 55 The World Bank BiH Floods Emergency Recovery Project (P151157) activities budget for which 15-Jun-2014 31-Dec-2018 30-Jun-2019 line ministries have presented final lists of priority activities according to Comments (achievements against targets): Mostly achieved: Ministries have presented final lists of priority activities according to project criteria for 93% of regional activities budgeted Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Piped household water Number 0.00 15000.00 58109.00 connections that are benefiting from 15-Jun-2014 31-Dec-2018 30-Jun-2019 rehabilitation works undertaken by the project Comments (achievements against targets): Fully achieved: 58,109 piped household water connections benefited from rehabilitation works versus target value of 15,000 which corresponds to an achievement rate of 380%. Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Formally Revised Completion Page 36 of 55 The World Bank BiH Floods Emergency Recovery Project (P151157) Target Roads rehabilitated, Rural Kilometers 0.00 80.00 263.00 15-Jun-2014 31-Dec-2018 30-Jun-2019 Comments (achievements against targets): Fully achieved: 263 km of rural roads rehabilitated versus target value of 80 km which corresponds to an achievement rate of 320%. Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Roads rehabilitated, Non- Kilometers 0.00 25.00 63.00 rural 15-Jun-2014 31-Dec-2018 30-Jun-2019 Comments (achievements against targets): Fully achieved: 63 km of non-rural roads rehabilitated versus target value of 25 km which corresponds to an achievement rate of 250%. Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Number of bridges Number 0.00 15.00 35.00 rehabilitated under the 15-Jun-2014 31-Dec-2018 30-Jun-2019 Page 37 of 55 The World Bank BiH Floods Emergency Recovery Project (P151157) Project Comments (achievements against targets): Fully achieved: 35 bridges rehabilitated versus target value of 15 which corresponds to an achievement rate of 230% Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Length of flood protection Meter(m) 0.00 50000.00 53460.00 infrastructure rehabilitated or reconstructed 15-Jun-2014 31-Dec-2018 30-Jun-2019 Comments (achievements against targets): Fully achieved: 53,460 m of flood protection infrastructure rehabilitated or reconstructed versus target value of 50,000 m which corresponds to an achievement rate of 107%. Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Number of landslides Number 0.00 15.00 56.00 rehabilitation activities 15-Jun-2014 31-Dec-2018 30-Jun-2019 Comments (achievements against targets): Page 38 of 55 The World Bank BiH Floods Emergency Recovery Project (P151157) Fully achieved: 56 landslides rehabilitated versus target value of 15 which corresponds to an achievement rate of 370%. Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Number of public Number 0.00 10.00 19.00 infrastructure facilities rehabilitated (schools, health 15-Jun-2014 31-Dec-2018 30-Jun-2019 facilities, priority public buildings, drainage ditches) Comments (achievements against targets): Fully achieved: 19 public facilities rehabilitated (schools, health facilities, priority public buildings, draining ditches) versus target value of 10 which corresponds to an achievement rate of 190%. Component: Component 3: Project Implementation Support and Capacity Building Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Number of public officials Number 0.00 30.00 137.00 who successfully completed capacity building activity on 15-Jun-2014 31-Dec-2018 30-Jun-2019 disaster risk management Page 39 of 55 The World Bank BiH Floods Emergency Recovery Project (P151157) Comments (achievements against targets): Fully achieved: 137 public officials trained which corresponds to an achievement rate of 450%. Page 40 of 55 The World Bank BiH Floods Emergency Recovery Project (P151157) B. KEY OUTPUTS BY COMPONENT Project Development Objective 1: Critical needs met in floods affected areas 1. Emergency goods distributed by entity-level strategic reserves (US$) Outcome Indicators 2. Direct project beneficiaries under Component 1 3. Direct project beneficiaries in total, and percentage of female 1. Emergency disaster recovery goods purchased in line with the established schedule (Percentage) Intermediate Results Indicators 2. Percentage of affected farms benefitting from Government support (Percentage) 1. US$27 million emergency goods distributed versus target value of US$30 million 2. 104,537 beneficiaries under component 1 versus target value of 75,000 Key Outputs by Component 3. 868,895 direct beneficiaries in total versus target value of 300,000 (linked to the achievement of the Objective/Outcome 1) 4. 51% female targeted 5. 100% of emergency disaster recovery goods purchased in line with the established schedule 6. 26% of affected farms benefitted from Government support versus 20% target Project Development Indicator 2: Restored functionality of infrastructure essential for public services and economic recovery in floods affected areas 1. Public service infrastructure facilities recovered to pre-floods condition with Project support (Number) Outcome Indicators 2. Direct project beneficiaries under Component 2 3. Direct project beneficiaries in total, and percentage of female Page 41 of 55 The World Bank BiH Floods Emergency Recovery Project (P151157) 1. Percentage of targeted municipalities that have identified, consulted and prepared final lists of priority local service activities for funding under the project according to project criteria 2. Percentage of regional activities budget for which line ministries have presented final lists of priority activities according to project criteria 3. Piped household water connections that are benefiting from rehabilitation works undertaken by the project (Number) Intermediate Results Indicators 4. Roads rehabilitated, Rural (Kilometers) 5. Roads rehabilitated, Non-rural (Kilometers) 6. Number of bridges rehabilitated under the project (Number) 7. Length of flood protection infrastructure rehabilitated or reconstructed (Meter) 8. Number of landslides rehabilitation activities (Number) 9. Number of public infrastructure facilities rehabilitated (schools, health facilities, priority public buildings, drainage ditches) (Number) 1. 446 facilities recovered versus target value of 200 facilities 2. 173,765 beneficiaries of rehabilitated regional infrastructure versus target value of 120,000 3. 840,995 beneficiaries of rehabilitated local infrastructure versus target value of 300,000 4. 51% female targeted Key Outputs by Component 5. 100% of targeted municipalities have identified, consulted and (linked to the achievement of the Objective/Outcome 2) prepared final lists of priority local service activities for funding under the project according to project criteria 6.Ministries have presented final lists of priority activities according to project criteria for 93% of regional activities budgeted 7. 58,109 piped household water connections benefitted from rehabilitation works versus target value of 15,000 8. 263.58 km of rural roads rehabilitated versus target value of 80 km Page 42 of 55 The World Bank BiH Floods Emergency Recovery Project (P151157) 9. 63.45 km of non-rural roads rehabilitated versus target value of 25 km 10. 35 bridges rehabilitated versus target value of 15 11. 53,460 m of flood protection infrastructure rehabilitated or reconstructed versus target value of 50,000 m 12. 56 landslides rehabilitated versus target value of 15 Page 43 of 55 The World Bank BiH Floods Emergency Recovery Project (P151157) ANNEX 2. BANK LENDING AND IMPLEMENTATION SUPPORT/SUPERVISION A. TASK TEAM MEMBERS Name Role Preparation David Michaud Task Team Leader(s) Kashmira Hoshang Daruwalla Procurement Specialist(s) Lamija Marijanovic Financial Management Specialist Vera Dugandzic Social Specialist Esma Kreso Beslagic Social Specialist Supervision/ICR Ko Takeuchi, Goran Tinjic Task Team Leader(s) Sidy Diop Procurement Specialist(s) Lamija Marijanovic Financial Management Specialist Anne N. Ranasinghe Procurement Team Vera Dugandzic Social Specialist Sanja Tanic Team Member Esma Kreso Beslagic Environmental Specialist Marinos Skempas Team Member Darko Milutin Team Member Page 44 of 55 The World Bank BiH Floods Emergency Recovery Project (P151157) B. STAFF TIME AND COST Staff Time and Cost Stage of Project Cycle No. of staff weeks US$ (including travel and consultant costs) Preparation FY14 0 2,592.77 FY15 22.471 75,684.40 FY16 0 0.00 Total 22.47 78,277.17 Supervision/ICR FY15 14.156 78,225.50 FY16 20.766 98,252.66 FY17 17.930 106,012.18 FY18 19.550 216,779.05 FY19 24.786 244,406.26 FY20 0 22.50 Total 97.19 743,698.15 Page 45 of 55 The World Bank BiH Floods Emergency Recovery Project (P151157) ANNEX 3. PROJECT COST BY COMPONENT Amount at Approval Actual at Project Percentage of Approval Components (US$M) Closing (US$M) (US$M) Component 1: Emergency 40.00 27.00 67.50% Disaster Recovery Goods Component 2: Rehabilitation 57.00 57.23 100.40% of Key Public Infrastructure Component 3: Project Implementation Support and 3.00 3.30 110.00% Capacity Building Total 100.00 87.53 87.53% Page 46 of 55 The World Bank BiH Floods Emergency Recovery Project (P151157) ANNEX 4. EFFICIENCY ANALYSIS 1. Efficiency under Component 1 depends on how to best channel transfers: (i) through in-kind distributions of set items (e.g. fuel), (ii) choice among a restricted list of goods (vouchers), or (iii) just cash, which leaves the most freedom to recipients. Each of these options yields a different opportunity for growth, hence a different efficiency. According to literature14 and past operations, under certain situations, including emergencies, unconditional cash transfers can be the most efficient formula, which gives the maximum flexibility to beneficiaries to allocate resources. 2. However, this holds true if the market is efficient and goods are available, which in an emergency situation is not always granted. Price distortions may also occur, resulting in a drop of consumption. There was no report of such case in the aftermath of the 2014 floods in BiH, confirming the intuition that this somewhat small market connected to other abundant ones was well suited for cash transfers. This was probably true for most items with the notable exceptions of those with a limited distribution circuit, such as fuel. For staple food, in-kind transfers were likely more efficient as a result of economy of scale (wholesale vs. retail). 3. The efficiency of Component 1 is that of the effect of a transfer on economic recovery and growth (higher- level objective) This can be reformulated as the macroeconomic value of timely consumption in the aftermath of a disaster. A given transfer is more efficient than another whenever it has a bigger impact on growth. While it is intuitive that a marginal disbursement in a post disaster context has a superior impact than one in a stable situation, this is challenging to document and there is little hard evidence of it. One reason is that relevant datasets are not only few but most often not for open source, e.g. property of (re)insurers. To the best of the author’s knowledge, some publications15 offer a relevant proxy if one considers these transfers as a trigger of an insurance policy where the State would be the insurer and premiums proxied by some share of the taxes collected from citizens. Literature that provides empirical and quantitative evidence of the impact of insurance on growth is based on the analysis of a comprehensive disaster database where post-disaster growth is regressed against the share of insured and uninsured losses, controlling for past growth patterns (lag modelling), macroeconomic variables, and the level of economic development16. 4. For each percent of asset transferred from uninsured to insured loss, there is a difference in overall cumulated growth impact of 2.9% of the GDP for low to middle income countries. This suggests an order of 14 For example: Johannes Haushofer and Jeremy Shapiro, 2016, The Short-Term Impact of Unconditional Cash Transfers to The Poor: Experimental Evidence from Kenya. 15 BIS Working Papers No 394, Unmitigated disasters? New evidence on the macroeconomic cost of natural catastrophes by Goetz von Peter, Sebastian von Dahlen, Sweta Saxena; Monetary and Economic Department, Bank of International Settlements, December 2012. 16 Such thorough analytical work is required to ensure that the observed variation in growth can be specifically explained by a post disaster situation – put differently, avoiding the correlation – causation confusion, i.e. to ensure that conclusions are drawn from relevant comparisons only. The auto-regression model is the following: 2 4 = ∑ ∗ − + ∑ ∗ log(− ) + | | + =1 =0 Where i and t range respectively over countries of dataset and time (in years), Y is the variable to explain (growth) – thus an autoregression of length 2, L are the losses of a given disaster, and is the standard error. Page 47 of 55 The World Bank BiH Floods Emergency Recovery Project (P151157) magnitude of the effect of transfers comparable to this project, as the effect is similar to that of the disbursement of a disaster policy insurance. 5. The issue of external validity, i.e. whether conclusions drawn from other situations are valid in this specific instance, should be discussed. Indeed, the results yield a significant variance over the growth impact of post- disaster transfers. Notwithstanding the cautiousness to use these figures, the BiH context seems favorable to such interpretation. Indeed “BiH is one of the world’s most consumption-driven economies fueled by a large public sector, with recurrent spending over investments and is fostered by the prevalence of remittances”17. 6. In this context, meeting critical needs through cash transfers are likely to have stimulated the economy in a way it was already structured for. Indeed, the productive sector was not largely exposed to disruption, and there was an existing workforce to address the emergency and rehabilitation agenda. The stimulation of consumption could have been an important mechanism for recovery. 7. The efficiency of Component 2 is that of the effect of infrastructure on economic recovery and growth. A typical range for the contribution to growth from infrastructure is of 8 to 20%18. The road sector which is one of the most benefitted sectors under FERP has a very different impact on growth depending on the context, and generally the impact is higher when connectivity is part of a transnational scheme. These observations could lead us to consider the most conservative estimates regarding the range of impact on growth as BiH has one of the poorest network connectivity in the region. 8. Anecdotal evidence of this is Sandski Most, in FBiH, were authorities were told by a significant foreign investor in the aeronautic industry that the quality of their infrastructure had been critical in choosing their site for their investment.19 9. Due to the complex nature of the project with over 400 sub-projects, an economic analysis similar to the one conducted at appraisal is feasible for the activities conducted in the transport sector under Component 2, i.e. road rehabilitation, which is a significant part of the project’s activities in both entities and the District. According to the economic model at appraisal, a rehabilitation of 50 km of secondary road would yield a yearly economic value of US$6 million. Given that the project achieved about 250 km of rural road rehabilitation, using the same model yields US$30 million benefit every year. This leads to an economic rate of return (ERR) of 15% in five years (29% in 15 years), which is well above the estimated benefit at appraisal. 10. While the above ERR provides confidence about the overall efficiency level, a finer assessment is required to get deeper insights about the operation efficiency. To understand how well the project performed in terms of (i) meeting critical needs and (ii) restoring functionality of infrastructure essential for public services and economic recovery in flood affected areas, it is reasonable to evaluating whether there was any better way 17 Rebalancing Bosnia and Herzegovina, A systematic country diagnosis, World Bank Group, 2015. Also documented in Higher but Fragile Growth, Western Balkans Regular Economic Report N°14, Fall 2018, World Bank Group; The Western Balkans – Reviving up the engines of growth and prosperity, World Bank Group, 2017 (report N° ACS22690 18 Bom, PRD and JE Lighthart (2009): How productive is Public Capital? A meta-analysis, Georgia State University, Andrew Young School of Policy studies, International Studies program working paper 09-12, quoted in: The impact of infrastructures on growth in developing countries, IFC economic notes, April 2012. 19 This was reported during a workshop in Sarajevo on the 6th of March 2019 with FBiH PMU and their counterparts, i.e. representatives of local authorities of affected areas. Page 48 of 55 The World Bank BiH Floods Emergency Recovery Project (P151157) of proceeding that would have yielded a higher performance along these two metrics (critical needs and public service infrastructure), with growth maximization as the key expected outcome. It should be noted that a need-based allocation, which in general addresses the “most critical needs”, may not be the best approach in terms of fostering growth. 11. On a different note, one could wonder about the impact on efficiency of the types of projects that were eligible under Component 2. For instance, other flood management projects could have been more relevant and potentially very efficient, falling perhaps within a Category A operation. However, these activities were out of the scope of a typical reconstruction project since they follow a different pace and require detailed hydrological studies. 12. There may exist more effective initiatives in flood risk management than those undertaken under FERP, but this cannot be interpreted as a lack of efficiency of the project. It means that flood risk management should be approached not solely through reconstruction projects but also with other mechanisms allowing for a more preventive approach. 13. It is likely that the “Build Back Better” approach will lead to reduction of average annual losses in the future. A thorough assessment of this would require extensive technical input, but a few considerations can be made. The average annual losses of BiH related to floods is about US$636 million. Given that the 2014 flood affected 40% of the territory, a conservative estimate of potential damages in this area yearly is US$254 million. The overall rehabilitation works throughout the entities and District reached 64 km, which can be estimated to represent about 16% of the equivalent river length corresponding to an annual damage of US$ 254 million, hence addressing an equivalent base of US$ 40 million. An assumption of risk reduction of 10% would yield a yearly benefit of US$ 4 million, which is equivalent to a Net Present Value (NPV) of US$50 million over 20 years (discounted at 5%). This is another factor that contributed to project efficiency. Page 49 of 55 The World Bank BiH Floods Emergency Recovery Project (P151157) ANNEX 5. BORROWER, CO-FINANCIER AND OTHER PARTNER/STAKEHOLDER COMMENTS The BD had no comments on the ICR. No comments were sent by the FBiH and RS. Page 50 of 55 The World Bank BiH Floods Emergency Recovery Project (P151157) ANNEX 6. SUPPORTING DOCUMENTS (IF ANY) 1. European Union, United Nations, World Bank, Global Facility for Disaster Reduction and Recovery. 2014. Recovery Needs Assessment 2014: Bosnia and Herzegovina Floods 2. World Bank. 2014. Project Appraisal Document on a Proposed Credit in the Amount of SDR 65 Million in IDA Crisis Response Window Resources to Bosnia and Herzegovina for a Floods Emergency Recovery Project. Washington, DC: World Bank. 3. World Bank. 2011. Bosnia and Herzegovina - Country partnership strategy for FY2012-FY2015 (English). Washington, DC: World Bank. http://documents.worldbank.org/curated/en/768631468007532455/Bosnia- and-Herzegovina-Country-partnership-strategy-for-FY2012-FY2015 4. FBiH and RS Implementation Status Reports, various 2014-2019 5. World Bank, Implementation Status Report, various 2014-2019 6. World Bank, Aide Memoires and Management Letters, various 2014-2019 7. World Bank. 2012. Acting today for tomorrow: a policy and practice note for climate and disaster resilient development in the Pacific Islands region: Main report (English). Washington, DC: World Bank. http://documents.worldbank.org/curated/en/231221468288339561/Main-report 8. Urbano M De Bettencourt, Teresa Sofia; Ebinger, Jane Olga; Fay, Marianne; Ghesquiere, Francis; Gitay, Habiba; Krausing, Jarl; Kull, Daniel Werner; Mccall, Kevin; Reid, Robert Curle Jesse; Simpson, Alanna Leigh. 2013. Building resilience: integrating climate and disaster risk into development - the World Bank Group experience: Main report (English). Washington DC; World Bank. http://documents.worldbank.org/curated/en/762871468148506173/Main-report, and 9. Jha, Abhas K.; Barenstein, Jennifer Duyne; Phelps, Priscilla M.; Pittet, Daniel; Sena, Stephen; Jha, Abhas K.; Barenstein, Jennifer Duyne; Phelps, Priscilla M.; Pittet, Daniel; Sena, Stephen. 2010. Safer homes, stronger communities: a handbook for reconstruction after natural disaster: Safer homes, stronger communities: a handbook for reconstructing after natural disasters (English). Washington, DC: World Bank. http://documents.worldbank.org/curated/en/290301468159328458/Safer-homes-stronger-communities-a- handbook-for-reconstructing-after-natural-disasters 10. Rebalancing Bosnia and Herzegovina, A systematic country diagnosis, World Bank Group, 2015. Also documented in Higher but Fragile Growth, Western Balkans Regular Economic Report N°14, Fall 2018, World Bank Group; The Western Balkans – Reviving up the engines of growth and prosperity, World Bank Group, 2017 (report N° ACS22690 11. Bom, PRD and JE Lighthart (2009): How productive is Public Capital? A meta-analysis, Georgia State University, Andrew Young School of Policy studies, International Studies program working paper 09-12, quoted in: The impact of infrastructures on growth in developing countries, IFC economic notes, April 2012. Page 51 of 55 The World Bank BiH Floods Emergency Recovery Project (P151157) 12. Johannes Haushofer and Jeremy Shapiro, 2016, The Short-Term Impact of Unconditional Cash Transfers To The Poor: Experimental Evidence From Kenya. 13. BIS Working Papers No 394, Unmitigated disasters? New evidence on the macroeconomic cost of natural catastrophes by Goetz von Peter, Sebastian von Dahlen, Sweta Saxena; Monetary and Economic Department, Bank of International Settlements, December 2012. 14. Bosnia and Herzegovina Council of Ministers, Directorate for Economic Planning, Strategic Framework for BiH, August 2015. Page 52 of 55 The World Bank BiH Floods Emergency Recovery Project (P151157) ANNEX 7. PROJECT PARTS AND COMPOSITION OF RELATED SUB-COMPONENTS Part Component Description Part A 1. Emergency Provision of Eligible Essential Goods and Federation Disaster Commodities of Bosnia Recovery and Goods Herzegovina 2. Rehabilitation 2A. Rehabilitating or reconstructing high priority, (FBiH) of Key Public regional-level public infrastructure through Sub- Infrastructure projects in Flood Affected Areas 2B. Rehabilitating or reconstructing high priority, local-level public service delivery infrastructure through Sub-projects in Flood Affected Areas 3. Project 3A. Project implementation support through the Implementation provision of incremental operating costs of Support and Federation PIU and equipment for project Capacity implementation Building 3B. Strengthening the institutional capacity of the Federation and local authorities within the Federation to a proactive approach of risk management, through the provision of technical assistance and training Part B 1. Emergency Provision of Eligible Essential Goods and Republika Disaster Commodities Srpska Recovery (RS) Goods 2. Rehabilitation 2A. Rehabilitating or reconstructing high priority, of Key Public regional-level public infrastructure through Sub- Infrastructure projects in Flood Affected Areas 2B. Rehabilitating or reconstructing high priority, local-level public service delivery infrastructure through Sub-projects in Flood Affected Areas 3. Project 3A. Project implementation support through the Implementation provision of incremental operating costs of RS Support and APCU and equipment for project Capacity implementation Building 3B. Strengthening the institutional capacity of the RS and local authorities within the RS to a proactive approach of risk management, through the provision of technical assistance and training Page 53 of 55 The World Bank BiH Floods Emergency Recovery Project (P151157) Part C 1. Emergency Provision of Eligible Essential Goods and Brčko Disaster Commodities District Recovery (BD) Goods 2. Rehabilitation 2A. Rehabilitating or reconstructing high priority, of Key Public regional-level public infrastructure through Sub- Infrastructure20 projects in Flood Affected Areas 3. Project 3A. Project implementation support through the Implementation provision of incremental operating costs of RS Support and APCU and equipment for project Capacity implementation Building 3B. Strengthening the institutional capacity of the BD to a proactive approach of risk management, through the provision of technical assistance and training 20 Part C for the BD does not include component 2B on local infrastructure, since BD has only one governing administrative unit Page 54 of 55 The World Bank BiH Floods Emergency Recovery Project (P151157) ANNEX 8. PROJECT FUNDS ALLOCATION Part Component Allocation at Allocation after appraisal project (US$ restructuring millions)18 (US$ millions)18 Part A Eligible Essential Goods and 19.0 15.2 Federation Commodities for Component 1 of Bosnia Goods, works, non-consulting services, 28.5 32.3 and consultants’ services, training, and Herzegovina incremental operating costs for (FBiH) Components 2 and 3 Part B Eligible Essential Goods and 19.0 15.9 Republika Commodities for Component 1 Srpska Goods, works, non-consulting services, 28.5 31.6 (RS) consultants’ services, training, and incremental operating costs for Components 2 and 3 Part C Eligible Essential Goods and 2.0 1.4 Brčko Commodities for Component 1 District Goods, works, non-consulting services, 3.0 3.6 (BD) consultants’ services, training, and incremental operating costs for Components 2 and 3 Project amount (in US$ million)21 FBiH 47.5 RS 47.5 BD 5.0 Total 100.0 21The amount of the credit allocated to each part and component of the project is expressed in SDR in the Financing Agreement. The corresponding amounts in US$ million are estimated by considering an average exchange rate of 1.54 for June 2014, i.e. 1US$ = 1.54SDR. Page 55 of 55