Document of The World Bank Report No: ICR00003876 IMPLEMENTATION COMPLETION AND RESULTS REPORT (IDA-47570) ON A IDA-47570 IN THE AMOUNT OF SDR 76.2 MILLION (US$ 115 MILLION EQUIVALENT) TO THE PEOPLE’S REPUBLIC OF BANGLADESH FOR A EMPOWERMENT AND LIVELIHOOD IMPROVEMENT “NUTON JIBON� PROJECT (ALSO CALLED SOCIAL INVESTMENT PROGRAM PROJECT - SIPP II) October 24, 2016 Global Practice Agriculture Bangladesh South Asia CURRENCY EQUIVALENTS (Exchange Rate Effective December 31, 2015) Currency Unit = BDT US$ 1 = 78.48 BDT FISCAL YEAR July 1 – June 30 ABBREVIATIONS AND ACRONYMS Nuton Jibon District Community CAS Country assistance strategy NJDCS Society Nuton Jibon Groups (formerly CDD Community driven development NJG referred to as Jibikayan Groups) CPF Country Partnership Framework OP Operational policy CPS Country partnership strategy PC Procurement Committee CTL Cluster Team Leader PDO Project development objective Environment and Climate Change Risk ECCRR PRSP Poverty reduction strategy paper Reduction FGD Focus group discussion SAC Social Audit Committee GOB Government of Bangladesh SDF Social Development Foundation Small ethnic and vulnerable GP Gram Parishad SECVP community people GS Gram Samiti SF Shabolombi fund KII Key informant interviews SSC Sanchay Sangrakhan Committee LAC Loan Approval Committee VCO Village Credit Organization Village Development and Risk NJCS Nuton Jibon Community Society VDRRF Reduction Fund Nuton Jibon Cluster Community NJCCS Society Senior Global Practice Director: Juergen Voegele Sector Manager: Shobha Shetty Project Team Leader: Seenithamby Manoharan, Pushina Kunda Ng’Andwe ICR Team Leader: Melissa Williams 2 BANGLADESH Empowerment and Livelihood Improvement “Nuton Jibon� Project (Also Called Social Investment Program Project - SIPP II) CONTENTS Data Sheet A. Basic Information...................................................................................................... iv B. Key Dates .................................................................................................................. iv C. Ratings Summary ...................................................................................................... iv D. Sector and Theme Codes ........................................................................................... v E. Bank Staff ................................................................................................................... v F. Results Framework Analysis ..................................................................................... vi G. Ratings of Project Performance in ISRs ................................................................. xiii H. Restructuring (if any) .............................................................................................. xiv I. Disbursement Profile ................................................................................................ xv 1. Project Context, Development Objectives and Design ............................................... 1 2. Key Factors Affecting Implementation and Outcomes .............................................. 6 3. Assessment of Outcomes .......................................................................................... 13 4. Assessment of Risk to Development Outcome......................................................... 27 5. Assessment of Bank and Borrower Performance ..................................................... 28 6. Lessons Learned ....................................................................................................... 31 7. Comments on Issues Raised by Borrower/Implementing Agencies/Partners .......... 32 Annex 1. Overview of Split Rating for Restructuring .................................................. 34 Annex 2. Project Costs and Financing .......................................................................... 36 Annex 3. Outputs by Component ................................................................................. 38 Annex 4. Economic and Financial Analysis ................................................................. 42 Annex 5. Bank Lending and Implementation Support/Supervision Processes ............ 54 Annex 6. Summary of Borrower's ICR and/or Comments on Draft ICR ..................... 56 Annex 7. List of Supporting Documents ...................................................................... 61 MAP iii A. Basic Information Empowerment and Livelihood Country: Bangladesh Project Name: Improvement "Nuton Jibon" Project Project ID: P073886 L/C/TF Number(s): IDA-47570 ICR Date: 06/24/2016 ICR Type: Core ICR People's Republic Of Lending Instrument: SIL Borrower: Bangladesh Original Total XDR 76.20M Disbursed Amount: XDR 76.18M Commitment: Revised Amount: XDR 76.20M Environmental Category: B Implementing Agencies: Social Development Foundation (SDF) Cofinanciers and Other External Partners: B. Key Dates Revised / Actual Process Date Process Original Date Date(s) Concept Review: 01/15/2008 Effectiveness: 09/24/2010 11/16/2011 Appraisal: 04/05/2010 Restructuring(s): 02/19/2014 12/24/2015 Approval: 06/23/2010 Mid-term Review: 09/30/2013 10/31/2013 Closing: 06/30/2016 12/31/2015 C. Ratings Summary C.1 Performance Rating by ICR Outcomes: Moderately Satisfactory Risk to Development Outcome: Moderate Bank Performance: Moderately Satisfactory Borrower Performance: Moderately Satisfactory C.2 Detailed Ratings of Bank and Borrower Performance (by ICR) Bank Ratings Borrower Ratings Moderately Quality at Entry: Government: Satisfactory Unsatisfactory Implementing Quality of Supervision: Moderately Satisfactory Moderately Satisfactory Agency/Agencies: Overall Bank Overall Borrower Moderately Satisfactory Moderately Satisfactory Performance: Performance: iv C.3 Quality at Entry and Implementation Performance Indicators Implementation Performance Indicators QAG Assessments (if any) Rating Potential Problem Project at any time No Quality at Entry (QEA): None (Yes/No): Problem Project at any time (Yes/No): No Quality of Supervision (QSA): None DO rating before Closing/Inactive Moderately status: Satisfactory D. Sector and Theme Codes Original Actual Sector Code (as % of total Bank financing) Agro-industry, marketing, and trade 1 Animal production 35 Crops 10 Microfinance 32 40 Other social services 44 10 Public administration- Other social services 23 5 Theme Code (as % of total Bank financing) Participation and civic engagement 10 25 Rural markets 2 3 Rural non-farm income generation 34 30 Rural policies and institutions 22 12 Rural services and infrastructure 32 Social Inclusion 30 E. Bank Staff Positions At ICR At Approval Vice President: Annette Dixon Isabel M. Guerrero Country Director: Qimiao Fan Ellen A. Goldstein Practice Manager/Manager: Shobha Shetty Simeon Kacou Ehui Project Team Leader: Seenithamby Manoharan Meena M. Munshi ICR Team Leader: Melissa Williams ICR Primary Author: Melissa Williams v F. Results Framework Analysis Project Development Objectives (from Project Appraisal Document) The original Project Development Objective was to improve the livelihoods, quality of life and resilience to climate variability, natural hazards and other shocks of the rural poor, especially the left-out poor and vulnerable households. Revised Project Development Objectives (as approved by original approving authority) The revised PDO was to improve the livelihoods of extreme poor communities and to strengthen the community institutions in selected districts. (a) PDO Indicator(s) Formally Actual Value Original Target Revised Achieved at Indicator Baseline Value Values (from Target Completion or approval documents) Values Target Years Village institutions supported by the project are functioning in a Indicator 1 : transparent, inclusive and accountable manner, as per project guidelines Value quantitative or 1224 75% 2753 92% Qualitative) Date achieved 12/21/2011 06/30/2016 02/19/2014 12/31/2015 Achieved 123% of PAD target and 110% of the stated target from the restructuring in February 2014.1 Original PAD target was 75%. Of the 658,661 Comments households benefitting by the project, 651,480 were mobilized into Nuton Jibon (incl. % Groups (NJGs), which are clustered into 6957 Sanchay Sangrakham Committees achievement) (SSCs). Baseline is from SIPP-1 ICRR report, where 1224 out of 1407 total villages had institutions meeting the target. Indicator 2 : Direct project household beneficiaries Value New quantitative or 258,113 652,000 658,661 Indicator Qualitative) Date achieved 02/19/2014 6/30/2016 12/31/2015 Achieved 104% of target. This indicator was added during restructuring a core Comments indicator. Using average 5.5 people/household as per PAD, the total number of (incl. % direct beneficiaries was 3,622,635. Of the poor, hard-core poor, and vulnerable achievement) poor beneficiary households, 582,000 were represented in the community institutions by women and 36,000 were represented by men. Indicator 3 : Direct beneficiary women as a share of direct project HHs beneficiaries 1 At restructuring, the revised results framework gave the EOP target as a number (2753) instead of a percent. Even though the paper indicates that no change was made to the indicator, the number given was actually 84 percent of the 3262 target villages. Using an over-abundance of caution, achievement against both targets (the intentional and the inadvertent) was measured. vi Formally Actual Value Original Target Revised Achieved at Indicator Baseline Value Values (from Target Completion or approval documents) Values Target Years Value New quantitative or 90% 90% 94% Indicator Qualitative) Date achieved 02/19/2014 06/30/2016 12/31/2015 Achieved 104% of target. This indicator was added during restructuring as a core Comments indicator. All of the 39,255 decision-making positions in the village (incl. % organizations are held by members of the poor, HCP, and vulnerable (against a achievement) target of 80%) and 94% of these office holders are women. Increase in HHs income by at least 50 percent against base year for 50 Indicator 4 : percent of targeted households by the end of the project Value quantitative or 11% 50% No change 62% Qualitative) Date achieved 11/30/2012 06/30/2016 12/31/2015 Comments Achieved 124% of target. The final measurement is based on independent studies (incl. % and the findings of the independent impact evaluation of the project. achievement) At least 80% of targeted households benefit from improved community Indicator 5 : infrastructure or social services sub-projects Value quantitative or 258,113 80% No change 92% Qualitative) Date achieved 12/21/2011 06/30/2016 12/31/2015 Achieved 115% of target. 3259 villages have a Gram Samiti office building that include a sanitary latrine, and many also serve as safe spaces for villagers during floods. Most subprojects benefit a broad section (if not all) of the village, even non-target households. In addition, latrines installed on school grounds or serving Comments multiple families benefitted a larger number of people than the number of (incl. % subproject would indicator. Focus group discussions reveal that the Gram Samiti achievement) office buildings have raised the social status of the GS members. Some GS generate income by renting out the GS office, others use the GS office to host campaigns like immunization programs. Earthen roads have improved access to markets and to critical health care, and culverts have significantly reduced water pollution, water logging and erosion of roads. (b) Intermediate Outcome Indicator(s) Original Target Actual Value Formally Values (from Achieved at Indicator Baseline Value Revised approval Completion or Target Values documents) Target Years GSs accessing and managing project funds as per agreed rules and service Indicator 1 : standards as agreed in the COM vii Original Target Actual Value Formally Values (from Achieved at Indicator Baseline Value Revised approval Completion or Target Values documents) Target Years Value (quantitative 33% 80% No Change 98% or Qualitative) Date achieved 12/21/2011 06/24/2010 12/31/2015 Comments Target 123% achieved. Indicator revised at restructuring to say "project funds" (incl. % instead of "VDRRF"; targets were not changed. achievement) JG members having mobilized savings and accessed additional sources of Indicator 2 : finance Value (quantitative 17% 80% No Change 91% or Qualitative) Date achieved 12/21/2011 06/30/2016 02/19/2014 12/31/2015 Target 114% achieved. According to the community finance thematic report commissioned by SDF, NJG members disbursed at least BDT 469.83 million as Comments loans from their accumulated savings, and over 70% of them availed shabolombi, (incl. % vulnerability, or incentive related loans to embark on livelihood ventures or risk achievement) reduction measures with those members who borrowed over several cycles steadily increasing the amounts that they borrowed. At least 80% of completed community infrastructure subprojects are being Indicator 3 : operated and maintained by communities Value (quantitative 43% 80% No Change 89% or Qualitative) Date achieved 12/21/2011 06/30/2016 02/19/2014 12/31/2015 Target 111% achieved. According to SDF MIS, 3006 of the 3262 villages had an Comments O&M plan and 3005 had started an O&M fund, thus contributing to (incl. % sustainability. The major infrastructure investments were Gram Samiti office, achievement) earthen road, culvert (ring/box), tube-well, latrine / urinal, market development, school development, drain, other 50% of villages have 5 Community Professionals (CPs) identified, trained Indicator 4 : and offering services by EOYr 3; and 70% by EOP Value (quantitative 0 70% No Change 47% or Qualitative) Date achieved 05/28/2010 06/24/2010 02/19/2014 12/31/2015 While the project did identify and build the capacity of 16,082 community professionals in different areas, which averages to 4.9 per village, only 47% of Comments those CPs could meet all of the criteria of the indicator by EOP. Ultimately, the (incl. % amount of services offered depends on demand for those services. Ultimately, the achievement) target was overly ambitious and would have benefitted from simplification or clarification during restructuring. At least 50% of GSs and VCOs of current program (Phase 1) are federated Indicator 5 : at upazila/cluster level by EOP viii Original Target Actual Value Formally Values (from Achieved at Indicator Baseline Value Revised approval Completion or Target Values documents) Target Years Value (quantitative 0 50% No Change 93% or Qualitative) Date achieved 05/28/2010 06/30/2016 12/31/2015 Target 186% achieved. By the close of SIPP-2, SDF had supported 93 percent of Comments Gram Samitis (GS) to join a federated NJCS. Overall, they formed 124 Nuton (incl. % Jibon Cluster Community Societies (NJCCS) and 15 Nuton Jibon District achievement) Community Societies (NJDCS), including forming key committees and providing initial training to staff. At least ten major business partnership established through public/private Indicator 6 : partnerships with communities Value (quantitative 1 10 No Change 10 or Qualitative) Date achieved 05/28/2010 06/30/2016 12/31/2015 Target 100% achieved. Partnerships and market linkages include— Grameen Comments Danon Foods Ltd, BRAC Chilling Center, Bio-Science Agro, Milk Vita, and (incl. % SSKS. These benefitted over 7000 project households with a variety of services achievement) as well as backward and forward linkages to markets. At least 50,000 direct jobs created through project facilitation & at least Indicator 7 : 50% of them are sustained by EOP Value (quantitative 7893 50,000 (75%) 50,000 (50%) 41,849 (99%) or Qualitative) Date achieved 05/28/2010 06/30/2016 02/19/2014 12/31/2015 Comments Target for job creation substantially achieved—84%. SDF commissioned (incl. % thematic study on youth jobs (the focus of this target) found 99% of youth were achievement) still employed by the end of the project. Staff performance appraisal results indicate 70% of staff appraised Indicator 8 : performing at Grades A and B from year 2 Value (quantitative 0 70% 70% 89% or Qualitative) Date achieved 05/28/2010 06/30/2016 02/19/2014 12/31/2015 Target 127% achieved. SDF had a strong HR policy and manual that was Comments developed with support from the Bank team. There was also a training program (incl. % for staff and a thorough community operations manual to guide staff on achievement) implementation issues. At least 70% of the villages (GS, SAC and VCO) are rated as A or B grades Indicator 9 : (Phase 1 and 2) Value (quantitative 0 70% 70% 86% or Qualitative) Date achieved 05/28/2010 06/30/2016 02/19/2014 02/19/2014 ix Original Target Actual Value Formally Values (from Achieved at Indicator Baseline Value Revised approval Completion or Target Values documents) Target Years Target 123% achieved. According to MIS tracking and project assessments, 2987 Comments village institutions are functioning in a transparent, inclusive, and accountable (incl. % manner (against a target of 2753). Ninety-nine percent of GSs are accessing and achievement) managing project funds as per agreed rules and service standards (target 80%). Successful proposals for Innovation Pilots awarded by EOP under the Indicator 10 : Competitive window Value (quantitative 0 9 9 2 or Qualitative) Date achieved 05/28/2010 06/30/2016 12/31/2015 While this was only 22% achieved, the activity generated some good lessons. A good practice from this activity, which was financed under the competitive window of Component C, is the support for a Marketplace for Youth Comments Employment Generation managed by Bangla-German Sampreeti. BGS trained (incl. % 989 youth, of which 810 were successfully employed with income ranges 7000- achievement) 10000 taka per month for wage employment and from 4000 to 6000 taka for self- employment. It also established youth information centers to increase awareness of employment opportunities, and it produced several valuable lessons for SDF’s skills development work. GAAP recommendations implemented and reviewed annually with the Indicator 11 : community representatives in 70% of target villages Value (quantitative 0 70% No Change 71% or Qualitative) Date achieved 05/28/2010 06/30/2016 12/31/2015 Comments Target 101% achieved. (incl. % achievement) At least 70% of process monitoring recommendations addressed by SDF Indicator 12 : management taking actions Value (quantitative 86% 70% No Change 83% or Qualitative) Date achieved 12/21/2011 06/30/2016 12/31/2015 Comments Target 119% achieved. The high baseline reflects the achievement of the SIPP (incl. % project, so the target reflected the volume managed with the scale-up. achievement) SDF has Complaints/Grievance Redress system established and 80% of Indicator 13 : complaints received have been addressed Value (quantitative 0% 80% No Change 89% or Qualitative) Date achieved 05/28/2010 06/30/2016 12/31/2015 x Original Target Actual Value Formally Values (from Achieved at Indicator Baseline Value Revised approval Completion or Target Values documents) Target Years Comments Target 111% achieved. (incl. % achievement) Targeted vulnerable and marginalized beneficiary population holding Indicator 14 : decision making positions. Value (quantitative 25% 80% 100% or Qualitative) Date achieved 02/19/2014 06/30/2016 12/31/2015 Target 125% achieved. This indicator was added as a core indicator at Comments restructuring. By project close, 37,000 female and 2400 males from P/HCP/V (incl. % held decision-making positions. achievement) Indicator 15 : Grievances related to delivery of project benefits that are addressed Value (quantitative 150 197 or Qualitative) Date achieved 06/30/2016 12/31/2015 Comments Target 131% achieved. SDF received a total of 222 grievances, of which 197 (incl. % were addressed. This indicator was added for official recording as a core achievement) indicator in April 2014. THE FOLLOWING INDICATORS WERE DROPPED DURING THE LEVEL 1 RESTRUCTURING Indicator 16 : 80% of target HHs mobilized and organized into Jibikayan Groups Value (quantitative 40% 80% Dropped 40% or Qualitative) Date achieved 06/30/2011 06/30/2016 02/19/2014 12/05/2013 Comments Indicator dropped at restructuring because information was already captured (incl. % through PDO Indicator 2 “Direct project household beneficiaries�. achievement) 80% of Gram Samitis have HCPs occupying decision making positions (and Indicator 17 : at least 50% are women) Value (quantitative 43% 80% Dropped 40% or Qualitative) Date achieved 06/30/2011 06/30/2016 02/19/2014 12/05/2013 Comments Target dropped because information already captured through Intermediate (incl. % Results Indicator 1 “Targeted vulnerable and marginalized beneficiary achievement) population holding decision making positions.� 70% of the HCP and vulnerable have accessed resources under the project Indicator 18 : (e.g., one-time grant, skill development loan, Shabolombi fund, etc.) and started income generating activities or employment xi Original Target Actual Value Formally Values (from Achieved at Indicator Baseline Value Revised approval Completion or Target Values documents) Target Years Value (quantitative 18% 70% Dropped 57% or Qualitative) Date achieved 12/21/2011 06/30/2016 02/19/2014 02/19/2014 Comments Target dropped at restructuring as it was captured by other indicators. (incl. % achievement) At least 70% of village organizations have applied participatory methods for Indicator 19 : risk mapping /vulnerability analysis and implemented risk reduction and preparedness methods Value (quantitative 16% 70% Dropped 84% or Qualitative) Date achieved 12/21/2011 06/30/2016 02/19/2014 04/19/2014 Comments The indicator has been dropped to reflect the revised PDO, but the target had (incl. % already been surpassed when it was dropped. achievement) Indicator 20 : At least 70% of graded CPs are earning at least 15 days in a month Value (quantitative 0 70% Dropped 10% or Qualitative) Date achieved 05/28/2010 06/30/2016 02/19/2014 02/19/2014 Comments This indicator was dropped at restructuring. (incl. % achievement) At least 40% of target HHs have benefited from value addition activities Indicator 21 : through economic federations Value (quantitative 0 40% Dropped 5% or Qualitative) Date achieved 05/28/2010 06/30/2016 02/19/2014 02/19/2014 Comments Target dropped at restructuring and the data was captured by another component. (incl. % achievement) Indicator 22 : At least 50% of the Ph. 1 federations are rated A and B Value (quantitative 0 50% Dropped 0% or Qualitative) Date achieved 05/28/2010 06/30/2016 02/19/2014 02/19/2014 Comments Target dropped at restructuring. (incl. % achievement) A youth database and employment exchange established and benefiting at Indicator 23 : least 75% of project villages xii Original Target Actual Value Formally Values (from Achieved at Indicator Baseline Value Revised approval Completion or Target Values documents) Target Years Value (quantitative 0 75% Dropped 92% or Qualitative) Date achieved 05/28/2010 06/30/2016 02/19/2014 02/19/2014 Comments Indicator dropped to reduce the number of indicators, but information continued (incl. % to be captured in the village matrix. achievement) At least 4 major partnerships established with organizations spanning Indicator 24 : academia, private sector, research institutes, NGOs and public sector Value (quantitative 0 4 Dropped 2 or Qualitative) Date achieved 05/28/2010 06/30/2016 02/19/2014 02/19/2014 Comments Indicator dropped to reduce the number of indicators in the RF. It was still (incl. % tracked by SDF. achievement) COM is revisited annually incorporating recommendations of process Indicator 25 : monitoring and other field experience and community feedback Value (quantitative 0 5 Dropped 2 or Qualitative) Date achieved 05/28/2010 06/30/2016 02/19/2014 02/19/2014 Comments Target dropped during restructuring, but the contents of the Community (incl. % Operations Manual (COM) continued to be tracked by SDF. achievement) At least 80% of SDF staff receiving positive score cards by the communities Indicator 26 : in annual CAP process Value (quantitative 0 80% Dropped 0 or Qualitative) Date achieved 06/24/2010 06/30/2016 02/19/2014 02/19/2014 Comments Dropped at restructuring because it was too early to be implemented. (incl. % achievement) G. Ratings of Project Performance in ISRs Actual Date ISR No. DO IP Disbursements Archived (USD millions) 1 12/17/2010 Satisfactory Satisfactory 0.00 2 06/25/2011 Satisfactory Satisfactory 5.77 3 12/24/2011 Satisfactory Satisfactory 8.19 4 06/08/2012 Satisfactory Moderately Satisfactory 22.65 xiii Actual Date ISR No. DO IP Disbursements Archived (USD millions) 5 12/19/2012 Satisfactory Moderately Satisfactory 38.52 6 05/30/2013 Satisfactory Moderately Satisfactory 56.97 7 12/20/2013 Satisfactory Moderately Satisfactory 75.66 8 04/19/2014 Satisfactory Moderately Satisfactory 91.84 9 08/23/2014 Satisfactory Satisfactory 101.23 10 03/05/2015 Satisfactory Moderately Satisfactory 111.32 11 05/09/2015 Satisfactory Moderately Satisfactory 111.32 12 11/12/2015 Moderately Satisfactory Moderately Satisfactory 115.64 H. Restructuring (if any) ISR Ratings at Amount Board Restructuring Disbursed at Restructuring Reason for Restructuring & Approved Restructuring Date(s) Key Changes Made PDO Change DO IP in USD millions To reconcile inconsistencies 11/16/2011 S S 6.66 between the Financing Agreement and the PAD. Better align PDO to institutional 02/19/2014 Y S MS 76.39 development focus of the project. To advance the closing date of 12/24/2015 MS MS 115.64 the project If PDO and/or Key Outcome Targets were formally revised (approved by the original approving body) enter ratings below: Outcome Ratings Against Original PDO/Targets Moderately Satisfactory Against Formally Revised PDO/Targets Satisfactory Overall (weighted) rating Moderately Satisfactory xiv I. Disbursement Profile xv 1. Project Context, Development Objectives and Design 1.1 Context at Appraisal 1. At the time of project appraisal, Bangladesh’s poverty rate was approximately 40 percent (down from 60 percent in the early 1990s); however, of the 56 million people remaining in poverty, 85 percent lived in rural areas, and their income opportunities and access to basic services were severely constrained. Nearly one-third of the rural poor suffered from chronic income poverty, and were, therefore, termed hard core poor (HCP). National poverty data on which project design and appraisal were based derived from Bangladesh’s 2005 household survey, and the appraisal team noted that the data most likely did not reflect the impact of natural disasters, like the floods and cyclone Sidr, that occurred between 2005 and 2010. Those poverty numbers were, most likely, on the low end. Natural disasters in Bangladesh can be seasonal or unpredictable―like the major floods and tropical cyclones in 2007 and 2009. The 2007 cyclone and floods were estimated to have caused a combined loss to the economy of about BDT 189.4 billion, or 4.7 percent of 2006 GDP. Bangladesh, particularly its agriculture sector and the rural population, was and remains highly susceptible to natural disasters. The design team recognized the role that climate shocks played in knocking people close to the poverty line back into poverty, and keeping the poor in poverty. The Government of Bangladesh’s (GOB) social safety nets intended for the left-out poor faced serious challenges in establishing clear targeting criteria and reached only about 15 percent of the population in rural areas and 5 percent in urban areas. Evidence at the time suggested that inequality and regional disparities in the country were increasing and that addressing the needs of the poorest was critical to enable more equitable, faster-paced, sustainable growth. 2. GoB was committed to addressing inequity and poverty through a broad reform program designed to accelerate economic growth. Its vision was embedded in the Poverty Reduction Strategy Paper (PRSP), “Unlocking the Potential�, released in October 2005 as the country’ s first National Strategy for Accelerated Poverty Reduction (NSAPR), and which centered on a policy triangle of growth, human development and governance. The follow-on NSAPR-2 (FY09-11) continued to build on this. Amongst the critical and cross-cutting strategies that were identified as essential to achieve the ‘strategic blocks’ of pro-poor economic growth, reduced vulnerability and human development, were participation and inclusion, good governance and improved service delivery. 3. The Bangladesh Country Assistance Strategy (CAS) (proposed 2010-13) aligned IDA assistance with this NSAPR II, as reflected in its two pillars which emphasized empowering the poor, as well as its cross-cutting focus on governance. Expanding poor people’s ability to make choices and take actions is at the heart of empowerment. Developing their capacity to organize, access and take advantage of available opportunities would enable both social and economic empowerment, and reduce vulnerability. 4. The SIPP-1 CDD operation—which was originally initiated in the two poorest north- western districts of Jamalpur and Gaibandha to test new ways of delivering critical infrastructure services and social assistance to the rural poor—successfully demonstrated the capacity of rural 1 villagers, especially poor and hard core poor women, to develop participatory and inclusive village institutions, accountable to their members and capable of prioritizing needs, managing resources and implementing sub-projects. GoB further expanded the program to respond to cyclone Sidr in 2007 and devastation caused by flood during 2008, aimed at livelihoods and asset restoration of the poorest and most vulnerable families in the affected districts. 5. The Empowerment and Livelihood Improvement “Nuton Jibon� Project (hereafter called SIPP-2) built on and scaled up the successful CDD approach pioneered by SIPP-1 and built on the approach by developing federated institutions of the poor to facilitate linkages with government and markets. 1.2 Original Project Development Objectives (PDO) and Key Indicators (as approved) 6. The original Project Development Objective was to improve the livelihoods, quality of life and resilience to climate variability, natural hazards and other shocks of the rural poor, especially the left-out poor and vulnerable households. 7. The PDO indicators were:  Village institutions supported by the project are functioning in a transparent, inclusive and accountable manner, as per project guidelines (EOP Target: 75%)  Increase in HHs income by at least 50 percent against base year for 50 percent of targeted households by the end of the project (EOP Target: 50%)  Targeted households benefiting from improved access to community infrastructure or social services sub-projects (EOP Target: 80%) 1.3 Revised PDO (as approved by original approving authority) and Key Indicators, and reasons/justification 8. The PDO was revised in February 2014 as follows: to improve the livelihoods of extreme poor communities and to strengthen the community institutions in selected districts. 9. The original PDO was somewhat ambitious and incomplete. It not include community institutions which was the center-piece of the development initiative, and included quality of life and resilience attributes which were vague and ambiguous. The project activities directly contributing the improving resilience were not clearly identified or articulated in the project. The resilience attribute measurement was acknowledged in the PAD as a difficult endeavor with a suggestion that “Measuring resilience is difficult, unless and until a shock tests the communities’ ability to withstand such hazards. The design of the project is such that the three outcome indicators should be taken together as an ‘index’ for resilience that addresses communities’ social, physical and financial coping capacities.2 10. The revised PDO refined the original PDO by excluding resilience and including community institutions development. None of the PDO indicators were changed either in 2 SIPP-2 PAD, page 37, footnote 19. 2 substance or in target, and one core PDO indicator was added: number of direct beneficiaries households and a core supplement measure the proportion of beneficiaries that are women. 1.4 Main Beneficiaries 11. SIPP-2 was designed to cover 15 districts (8 new districts and expansion in 7 districts treated under SIPP-1), working in 1,500 new villages while continuing to provide support to the 1407 villages mobilized under SIPP-1. The project aimed to directly benefit about 300,000 target households, or 1.8 million people. Total direct and indirect beneficiaries were estimated to be about 3 million people. 3 The project sought to create and strengthen village and community organizations and help them to form federations and producer groups (second generation institutions) to improve their livelihoods. Within project villages, the targeted households comprised the poor, hard-core poor (HCP), and vulnerable households, based on a participatory identification of the poor exercise. The Community Operations Manual outlines criteria for classifying households. SIPP-2 had a special focus on reaching women and youth, and strategies for each group were developed for the project. SIPP-2 was to cover 15 districts—entering 8 new districts and expanding in the 7 existing project districts, bringing in 1,500 new villages, benefitting about 300,000 target households, with about 3 million population, directly and indirectly. 12. SIPP-2 ultimately reached 16 districts and worked in 3262 villages (the original 1407 villages and 1855 new villages), directly targeting 658,661 households (or an estimated 3.4 million people), of which 94% were women.4 The total number of households in the project was 923,540 (or an estimated 5.1 million people). This represents total population, including those that indirectly benefitted from the project, particularly from infrastructure investments or increased economic activity within the village. 1.5 Original Components (as approved) 13. Component A: Community and Livelihoods Development at the Village Level (PAD: US$82.5 million/EOP: US$96.08 million). This component comprised two sub- components: (1) Development and Strengthening of Community Organizations to mobilize, develop and strengthen self-reliant and self-managed people’s organizations from the poorest and most vulnerable households; and (2) Village Development and Risk Reduction Fund (VDRRF) to directly finance village priorities in livelihoods and community infrastructure incorporating risk reduction plans. 14. Component B: Institutional Development and Livelihoods Promotion at the Inter- village Level (PAD: US$19.5 million/EOP: US$1.98 million). This component worked through three sub-components—(i) Development and Strengthening of Inter-village Organizations; (ii) Business Promotion, Livelihoods and Market linkages and Partnerships; and (iii) Creating Conditions for Employment Generation—to promote inter-village development to consolidate 3 Estimates of beneficiaries in the PAD were based on an average of 5.5 people per household and about 350 households per village, which approximately 200 households falling into the target beneficiary group. 4 The project districts were: Bagerhat, Barguna, Barisal, Dinajpur, Gaibandha, Jamalpur, Kurigram, Mymensingh, Naogaon, Nilphamari, Patuakhali, Pirojpur, Rangpur, Sherpur, Sirajganj, and Sylhet 3 and sustain livelihood activities and investments generated at the village level, promote business and establish market and informational linkages for community organizations, and facilitate job creation. 15. Component C: Capacity Development and Partnership Building from Cluster to National Levels (PAD: US$6.1 million/EOP: US$2.4 million). This component worked through two sub-components—(i) Strengthening the institutional capacity and delivery of services by SDF and other relevant partner agencies; and (ii) Supporting Innovation—to strengthen the capacity of the implementing agency and its partner agencies at all levels to deliver services and build linkages and synergy with the local governments and other public and private sector agencies and to pilot innovations in the areas of CDD approach, local level governance, climate risk management and community adaptation, environmental sustainability and livelihood development. 16. Component D: Project Management, Monitoring and Coordination (PAD: US$11.9 million/EOP: US$14.54 million). This component will support overall project coordination and management at national, regional and district levels; and monitoring, learning, communication and evaluation. 1.6 Revised Components 17. There were no material revisions to the activities in the components, but there were reallocations between components during implementation (see section below). 1.7 Other significant changes 18. There were significant reallocations and expansion of expenditure categories over the life of the project. 19. Project restructuring. SIPP-2 was restructured three times during implementation—two Level 2 restructurings and one Level 1 restructuring:  Level 2 Restructuring November 2011. The first Level 2 restructure was approved on 15 November 2011 and simply updated wording to fix inconsistencies between the Financing Agreement and the PAD and to correct a table defining the expenditure categories in the PAD. The changes corrected the allocation amounts in USD within the PAD, and allowed 100% financing under grants to communities in the FA (in accordance with the PAD).  Level 1 Restructuring February 2014. A Level 1 restructuring was approved on 19 February 2014, which (a) refined the PDO; (b) added a core PDO indicator; (c) simplified the results framework; and (d) reallocated funds across components and existing expenditure categories and created new expenditure categories to cover and better delineate operational project. The changes to the PDO better aligned it to the PDO indicators and the existing project design. There were no changes to the PDO indicators, and no changes to the scope of the activities or their targets. 4  Level 2 Restructuring December 2015. This restructuring was approved on 24 December 2015 to bring forward the project closing date to 31 December 2015, and to reallocate $16.34 million from expenditure categories 1 and 2 to expenditure categories 3 and 4. It also revised expenditure category 3 to include expenditures under subcomponent A1. The changes were retroactive to 1 December 2013 to address an oversight in the previous Level 1 restructuring. 20. Changes to component funding. Given the overlap of SIPP-2 design with SIPP-1 implementation, it was assumed at design stage that about $40 million from the additional financings in SIPP-1 would be extended and implemented under SIPP-2. SDF costs were calculated based on that assumption and 4 districts and 74 clusters were omitted from the original Costab. However, the additional financing was cancelled. As a result, SDF identified a funding gap was estimated to be as high as US$43.76. SDF, GoB, and the Bank team worked closely to overcome this gap through  Reducing SDF staffing by 288, from 2131 to 1843 positions;  Reducing the allocation for sub-component A2(Financing of Village Development and Risk Reduction) that represents sub-grant allocation for the Gram Samitis to eliminate under-utilized funds sitting idle in accounts; and  Significantly reducing Component B (Institutional Development and Services at Inter- Village and National Levels) by shifting unspent funds allocated for the first year of the project due to implementation delays for that component, and by shifting more tasks from outside consultants to SDF staff. There had been a significant over-allocation for support to second generation institutions that the existing SDF staff complement were able to do.  Eliminating unintended over-financing in the cost table stemming from allocating project funds based on total number of villagers instead of number of target households for villages having more than 500 households.  Exchange rate gains—shifting from 70 Taka to 80 Taka per US$—also generated savings amounting to nearly 15% of the IDA allocation of $115 million. 21. Changes to eligible expenditure categories. At appraisal, there were 2 expenditure categories: Sub-grants for communities under Part A.2 and B.1 of the Project and Goods, works, consultants’ services, training and Operating Costs. By the end of the project there were four categories: 1. Sub-grant for Communities under Part A.2 and B.1 of the project; 2. Goods, works, consultants’ services and training and operating costs—covering 100% for payments made before December 01, 2013 to cover the financing gap created by the cancelled additional financing; 3. Goods, works, consultants’ services and training parts A, B, C, and D—covering 100% for payments made on or after December 01, 2013. 4. Operating Costs under Parts C and D—covering 100% for payments made on or after December 01, 2013. 5 Table 1 Summary of component funding at appraisal and end of project Appraisal End of Project Components (US$m) (US$m) A. Community and Livelihoods Development at the Village Level US$82.5 US$96.08 B. Institutional Development and Livelihoods Promotion at the Inter- US$19.5 US$1.98 village Level C. Capacity Development and Partnership Building from Cluster to US$6.1 US$2.4 National Levels D. Project Management, Monitoring and Coordination US$11.9 US$14.54 22. The increase in funds to Component A are largely attributed to shifting the costs of SDF’s cluster level field teams that provided direct training and guidance to the communities under sub-Component A1 (Development and Strengthening Community Organizations) to be able to implement sub-Component A2 (Financing of Village Development and Risk Reduction Plans). 23. The significant reduction in Component B (subcomponent B1) was due to delays in implementation of that component at the beginning of the project—the readiness of villages to federate and the time to prepare 2nd generation institutions was under-estimated at appraisal—so the first year funds allocated to the component were shifted during the Level 1 restructuring in 2014 (see below). The component initially included costing for expensive external agencies to facilitate activities, but responsibility was shifted to existing SDF staff. The plan for external facilitation at appraisal could have been planned to overcome perceived weakness of SDF as it reorganized to accommodate the rapid scale-up of SIPP-1 and the requirements of SIPP-2. Ultimately, SDF’s existing staff could do the job. 24. The reduction in Component C reflects some difficulty in implementing the innovation fund—only 2 of 9 anticipated innovations were funded—but also the ability of existing field teams, district and regional staff, and the emerging second generation institutions to facilitate partnerships at the community level. This eliminated the need for external facilitation. 25. Despite the significant shifts among components and expenditure categories, it is important to note that SDF was able to achieve or exceed all of its PDO goals and all but three of its intermediate outcome goals. 2. Key Factors Affecting Implementation and Outcomes 2.1 Project Preparation, Design, and Quality at Entry 26. SIPP-2 reflected lessons from the ongoing operation and other examples . The project preparation team built on the successful experience of SIPP-1, which was still implementing when SIPP-2 was being designed and approved. It also drew from international experience— particularly in South Asia and Latin America―and continued to use the CDD approach, which had proven successful in most of the SIPP-1 activities. 27. The team took a strong risk management/mitigation approach to design. At the time of design, SDF was without a Governing Body and permanent Managing Director. It was also 6 accommodating the requirements of the rapid scale-up of SIPP-1 and preparation of SIPP-2, reorganizing its structure, reforming its human resource systems, and reconfirming and/or hiring approximately 2000 staff. SIPP-2 started with SDF capacity stretched—approximately 60% of staff were new—and it was implementing 4 other investments in addition to the new project. To mitigate the risks to implementation, the team and SDF took several steps, including:  Creating a field team architecture in existing districts to serve as a resource to new districts and newly recruited staff and building a network of CPs and federated societies to supplement SDF’s district resources and to enable scaling up in a sustainable and cost - effective manner;  Implementing a new HR policy and strategy focused on selection and retention of qualified and committed staff, the adoption and implementation of which was a legal covenant of the project and progress was initially tracked in the results framework;  Ensuring GoB creation of an endowment fund to ensure the sustainability of SDF, particularly in instances where bridge financing is required and ensuring the structuring and performance of the SDF Governing Body;  Building on the successful mobilization approach developed under SIPP-1 to identify beneficiaries in a participatory way, make sure village organizations functioned according to transparent rules and were led by the target beneficiaries—all of which was tracked through a village grading system and measured in the first PDO indicator;  Instituting a grievance redressal mechanism (GRM) to flag and manage issues with delivery of benefits to targeted communities, the activities of which were ultimately tracked in the results framework;  Developing manuals to codify good practice in management, human resources, and operations―including, a human resource policy and manual, manuals for financial management, for procurement, for economic activity federations and Nuton Jibon Community Societies (NJCSs), and a 15-volume community operations manual;  Instituting a Governance and Accountability Action Plan (GAAP) and implemented a governance, fraud, and corruption risk mitigation system at all levels of the project that incorporated governance into the mobilization process and reinforced it with social audit committees, community operations manuals and training, appraisal monitoring team, a process monitoring agency, and internal audit within SDF; and  Closely monitoring project activities through a large number of indicators in the results framework, supplemented by monitoring of physical and financial targets on a regular basis.5 28. Project activities aligned to the PDO, but funding allocations were misestimated for various reasons. The bulk of project activities defined in the PAD focused on building quality institutions of the poor and building their capacity to manage grant funds and loan funds to 5 The number of indicators (24 at the beginning of the project) would ultimately be reduced to fewer than 15 during the Level 1 Restructuring in 2014. 7 support infrastructure investments and livelihood support (Component A). Components B and C were designed to contribute to the long-term sustainability of the outcomes of Component A. The initial allocations to Component B were much higher than needed because costs for external facilitation and support had been incorporated and ultimately were not needed. Costs for Component C also reduced significantly because some activities under subcomponent C1 did not progress as expected while others required fewer funds than anticipated. Despite the shift in funds, both components achieved most of their targets (see Sections 1.7 and 3.5 for more details). 29. Miscalculations in project costing led to the need for restructuring and reallocations during implementation. Project costing for SIPP-2 initially excluded 4 districts and 74 clusters because there was an assumption that costs would be covered by $40 million from the SIPP-1 additional financing; however, those funds were cancelled in April 2011, leaving a significant financing gap for SDF. SDF, GoB, and the Bank team were ultimately able to enact changes to allocations to cover that gap (see Section 1.7). 2.2 Implementation 30. The project was not in “at risk� status at any point during implementation. It received either satisfactory or moderately satisfactory ratings for both ID and PO throughout implementation. Effectiveness was postponed from October 2010 to December 2010 to give the government sufficient time to meet all effectiveness covenants imposed by the Bank. Physical and financial progress lagged against planned targets at the very beginning of the project due to the delay implementation resulting from (1) limited human resources at the very beginning of implementation as SDF was contracting a human resources agency to manage the recruitment of over 2000 positions; and (2) SDF was completing the investments under SIPP-1, which closed June 30, 2011. However, after Q4 of 2011, actual disbursements consistently exceeded planned and the project was able to use all of the IDA credit six months ahead of schedule. 31. Human resource development of SDF. The HR policy and strategy implemented by SDF, albeit difficult to start up, resulted in what the closing impact assessment team described, “motivated, highly skilled and dedicated staffs, who have the sincere potentiality and expertise…�. Having the rigorous system to identify and retain good talent on the ground certainly contributed to SIPP-2 exceeding project targets and implementing the project so rapidly. 32. Community Professionals. These community resource people, which SDF began to develop under SIPP-1 and continued under SIPP-2, provide additional on-the-ground knowledge and capacity for mobilization and training of new communities. Given that they have come from SDF communities, they can relate to new communities and help them understand and accept the project principles more readily. 33. Mid-term Review. The project mid-term review (MTR) took place in October 2013, and it was informed by a mid-term assessment was commissioned from Development Technical Consultants Pvt. Ltd. The assessment reaffirmed the design of the project in terms of contributing to the project’s goals: “Adoption of the concept and design by SIPP-II in implementing the project activities has been able to place the rural poor and hardcore poor segments of the community at the center of all decision making process as related to planning and implementation of self-development sub-projects. As such, the design of the project as followed in the first three 8 years of SIPP-II should be continued for the remaining project period and even beyond, albeit with a few modifications in staffing positions.� The methodology of the assessment included review of secondary project documents, solicitation of perceptions from project stakeholders— Gram Parishad members (demand side) and the SDF program managers from all levels (supply side) through Focus Group Discussions (FGD) and Key Informant Interview (KII), respectively. A structured questionnaire was also administered among the randomly selected beneficiaries. The MTR team also observed the physical structures in the villages. It rated the progress towards achieving the development objective as satisfactory, noting that, according to the mid-line survey, 83 percent of village institutions were functioning in a transparent manner. The other two PDO indicators were close to being achieved. 2.3 Monitoring and Evaluation (M&E) Design, Implementation and Utilization 34. SDF developed under SIPP-1 a robust system of program monitoring from the village to the national level that SIPP-2 continued this system and improved upon. The system incorporated community self-monitoring mechanisms, routine monitoring data collection (e.g., for MIS), and a system of quarterly internal learning forums at all levels of the organization to address implementation issues and share successes. The intact system was supplemented by third-party process monitoring, special studies, and a rigorous impact evaluation. 35. Community Level Monitoring. SDF has developed a monitoring and benefit tracking system for Gram Samiti members to use that captures key implementation milestones, selection of beneficiaries, functioning of community level institutions, and use of funds. This system is centered on the Village Matrix, which captures beneficiary targeting, funds allocation, progress on subprojects, and status of revolving funds, income generating activities, and youth skill training and employment, among other data. In addition, the project had social accountability mechanisms—Social Audit Committees, procurement committee, public information displays, reporting in Gram Parishad (village assembly) meetings, and a participatory conflict resolution system, and annual community assessment process (score cards) and grading of villages. 36. Management Information System (MIS). SDF developed a comprehensive MIS— including hardware, software, communication links, wide area network, and information technology training for staff as well as installation, maintenance, and support—with the objective of connecting the headquarters with its field offices. The MIS comprises five modules: (a) Financial Information System; (b) Project Monitoring System; (c) Procurement System; (d) HR & Payroll System; and (e) Inventory System. Data is compiled by villagers with support from field implementation teams through a Village Matrix. SDF field teams check the accuracy of the village data and pass on to district teams for compilation and passing on to the regional level. Spot checks of data are supposed to be taken at the district and regional level as well. Regions submit the monthly and quarterly progress reports. SDF’s ML&E Unit in Dhaka consolidates all data and generates village, cluster, district, and project level reports. It also coordinates creation of the Annual Planning Report and the internal learning workshops. 37. Third Party Monitoring. SIPP-2 used third party monitoring by Center for Natural Resources Study (CNRS) & Maxwell Stamp, Ltd. To examine processes and activities on a monthly basis that would feed into quarterly monitoring meetings held in each region. CNRS uses 9 five types of tools for data collection and covers 200 villages every month. The process monitoring agency submitted monthly reports on key project activities and quarterly reports to summarize key lessons learned, case studies, and follow-up status. These are used in the quarterly internal learning workshops at the field level. 38. Impact Evaluation. An impact evaluation (IE) also measured progress against project goals by measuring impact in treatment and control villages through baseline, mid-line, and end- line surveys. The IE covered 480 villages (269 treatment, 211 control) in four districts. The surveys covered about 4,701 households living in these villages. Sampling was developed in two stages: (a) using secondary information to identify a sample of project and control villages and compiling a census to classify households into poor, hardcore poor, and vulnerable categories; and (b) drawing up a probability sample of households from these classified census lists. CNRS and the Flood Hazard Research Centre produced the baseline report. Bangladesh Institute of Development Studies (BIDS) conducted the mid-line survey. Survey data analysis informed the MTR assessment and the implementation completion findings. 39. Additional thematic studies. SDF also commissioned special studies―a mid-term assessment as well as thematic studies on infrastructure, youth development, livelihoods, and community finance just before project completion to identify successes, failures, and capture lessons from implementation. In addition, SIPP was assessed under the Citizen Action for Results, Transparency, and Accountability (CARTA) program, which sought to (a) gather information from communities, committee members and leaders to assess SDFs village loan process; and (b) support the level of effectiveness of different village-based institutions. CARTA focused on Jamalpur and Gaibandha districts, surveying a random sample of 400 respondents, performing a social audit (input tracking) of village loan systems, as well as FGDs and KIIs. 40. Communications and Information. SDF undertook a communication needs assessment of villages to identify ways to improve the flow of information to and within villages, and the communications unit is working with the technical and field teams to increase the flow of information in a form that will be most useful for communities —e.g., posters, billboards, video, etc. 41. System Implementation. As the passage above shows, SDF had an ambitious monitoring and evaluation program to implement. This presented a challenge for an agency that was managing such a rapid scale up and presented challenges to SDF’s capacity. The system progressed on all fronts, but progress was slow—e.g., difficulties in timely delivery of quarterly progress report, delays in MIS software upgrading to reflect the project design, delays in contracting the third-party process monitoring agency due to difficulties with the procurement process,6 and delays in producing the baseline survey which established many of the baseline values in the results framework. Delays in producing the mid-line survey limited its use for the mid-term assessment, but data did inform the restructuring and implementation thereafter. Once commenced, process monitoring progressed smoothly. It was used in quarterly internal learning workshop, and the use of that data was tracked in the results framework. However, the mid-term 6 Empowerment and Livelihood Improvement "Nuton Jibon" Project (P073886) Implementation Status and Results Report. Sequence 4. May 2012. 10 report noted that the PMA agency did not feel independent it was contracted by the system it was monitoring. In addition, supervision reports noted that the PMA reports were not shared regularly (or in a timely manner) with the Bank team. 42. The mid-term report also noted the need to improve data verification from the field, and SDF has worked on doing this. By project end, there was significant progress on having a complete and robust monitoring system, and the team was able to get significant amounts of data that could be cross checked with special studies and impact evaluation data to make a confident analysis of the SIPP-2’s progress. 2.4 Safeguard and Fiduciary Compliance 43. Environmental and Social Safeguards. SIPP-2 was environmental category B project, and the only safeguard policy triggered was OP/BP 4.01 Environmental Assessment. The project disallowed any involuntary resettlement—requiring that any project activities be contained to publicly owned land that was free from squatters, encroachments and other encumbrances. Acquisition of private land was done through purchase at market rates or voluntary donation in accordance with guidelines and procedures set forth in the PIP and COM. 44. Environment. The project prepared an ESMF to (a) protect environmental health and safety of poor and vulnerable communities; (b) prevent or mitigate negative environmental impacts as a result of either individual schemes or their cumulative effects; and (c) enhance positive environmental outcomes. The ICR mission found that none of the 3,682 infrastructure subprojects were approved or funds transferred without an environmental screening and Environmental Management Plans (EMP). Only 13 cases of environmental problems were identified after construction, and they were mitigated immediately by the Gram Samiti. SDF tested for arsenic contamination on all 7138 tube-wells implemented under the SIPP and SIPP-2 and sealed 141 as a result. 45. SDF has institutionalized an environmental monitoring system in SIPP-2 and collected data on 24 indicators from the field on a quarterly basis to monitor the changes in infrastructure and livelihood interventions. A Technical Audit in October 2015 confirms that SDF schemes, in general, have not created any major harm to the environment, while two positive environmental impacts from subprojects― water logging by culvert and reduction of surface water pollution due to installation latrines―are evident. SIPP-2 villagers planted a total of 1,015,372 trees at Gram Samiti Office premises and rehabilitated earthen road sides and homestead areas for environmental enhancement and soil erosion protection. 46. From “do no harm� to “do good�. During implementation, the Bank recommended SDF to adopt two innovations: (1) vermi-compost to enhance soil fertility and supplement family income; and (2) improved cook stove (ICS) to reduce fire wood usage and air pollution. By end of project, nearly 1,950 community members have installed ICS and received training on its benefits. SDF also facilitated a partnership with Bio-Science Agro to provide training on the production of vermi-compost in Giabandha. Five beneficiaries adopted it as a supplementary livelihood and nearly 1,300 households benefited from the technology. These approaches would be good to scale up under the follow-on NJLIP. 11 47. Social. SIPP-II did not trigger OP 4.12 Involuntary Resettlement or OP4.10 Indigenous People (IP), although the project intervention areas in Naogaon, Dinajpur, Mymensingh, Barguna, Patuakhali, Jamalpur, and Rangpur did cover pockets where small ethnic and vulnerable community people (SEVCP) live. Of the SEVCP people who did fall in the project area, 100 percent of the households joined NJG and 77 percent had received SF loans. There were 765 NJG members holding decision-making positions in 128 villages. No negative discrimination was reported, and instead this group has gained positive benefits in terms of economic improvement and enhanced social status in their communities by taking on leadership roles in their villages. The project interventions such as cow-fattening, cattle rearing, poultry, fisheries, vegetables, and small shops were socially and culturally suitable for women from SEVCPs. Land was purchased directly for the Gram Samiti office buildings, so no involuntary resettlement occurred as a result of project activities. The other subproject activities did not require land acquisition. 48. Financial Management. The overall financial management performance rating of the SIPP-2 is satisfactory considering the SDF’s initiatives to reinforce internal controls and accountability. SDF’s FM staffing is adequate and satisfactory under the project, which comprises of a number of professional and experienced personnel and headed by Finance Director. On planning and budgeting, adequate procedures were followed in the budgeting process in line with government requirements, and project financial plans are prepared based on the procurement plan. Upgradation process of automated accounting information system with budgetary control feature is under progress. The upgraded system would enable project accurate tracking of actual expenditure on a similar chart of accounts. This is expected to be operationalized by September 30, 2016 (under the follow-on project). SDF’s Financial Manual was updated and circulated in July 2015. Modification and updating of the HR manual with adequate interface with FM functions has been completed with adequate clarity on compensation and benefit was provided. The audited financial statements for FY 2015-16 is required to be submitted to the World Bank by December 31, 2016. Management reply on the observations on Integrated Fiduciary Review (IFR-16) is yet to be submitted to the Bank.� 49. Procurement. SIPP-2 had a combination of centralized and decentralized procurement. The centralized part was done by the SDF head office while the decentralized procurement activities were carried out at beneficiary community level. The centralized procurement, in principle, followed the Government of Bangladesh’s national procurement laws. Procurement was reasonably timely until early 2014. In June 2014, SDF revised its HR policy and manual in accordance with instruction from the Ministry of Finance to set an appropriate retreatment age in line with the prevailing procedure of the GoB. The changes were made in consultation and with the no objection of the Bank. As a result, the retirement age for SDF staff became 60 years of age, and this meant that SDF’s Procurement Manager had retire immediately. The rapid departure of a seasoned staff from such a critical position did lead to delays in procurements. Ideally, a 3-6 month overlap between the outgoing manager and their replacement would have eased the transition. There was one case of misprocurement involving shopping contracts, to which SDF responded following notification and took sanctions against the firms who performed the contracts. Community procurement followed a simplified procurement manual developed by SDF. In general, this was a successful system with no significant deviation from agreed procedures. 12 50. The lessons learned from SIPP-2 was used in designing the NJLIP follow-on project. 2.5 Post-completion Operation/Next Phase 51. The follow-on operation to SIPP-2, Nuton Jibon Livelihood Improvement Project (P149605), became effective in August 2015. SDF has taken several steps to ensure the sustainability of the community institutions, the livelihoods, and the sub-project investments made during SIPP-2. 52. Community institutional sustainability. The CDD approach used by SDF has built a network of strong institutions at the village level with a powerful underlying mobilization that has raised the commitment of the target population to sustain what they have built. The system of governance and accountability established in the villages and throughout SDF should safeguard the village institutions from political interference and corruption. Registering the GSs as joint- stock companies—already underway during SIPP-2—will give them the legal identity they need to interact with government and the private sector as well. 53. The critical next step, which SDF is taking to reach beyond the village and network these institutions at the cluster and district level through Nuton Jibon Community Societies (NJCS). By the close of SIPP-2, 93 percent of Gram Samitis (GS) had joined a federated NJCS. The financing model seems strong for a start—a corpus fund put into an interest-bearing account to cover sustaining costs, membership dues from member GSs, and fee-based services to members. A priority for NJLIP will be to strengthen the capacity of the NJCS to be able to develop and execute business plans and to have the technical skills that the market will pay for. Similarly, SIPP-2 organized 90,526 producers into 5,561producer groups, which helps the producers achieve scale to better negotiate with input and output markets. These linkages with private sector providers and buyers can boost the potential sustainability of the livelihood investments. 54. SDF Sustainability. The Government of Bangladesh has established an Endowment Fund with the equivalent of US$21.5 million. Earning 10% interest, it is expected to sustain SDF core staff, which had provided stability for human resources in the agency. The NJLIP team will continue to dialogue with GoB regarding future needs for the fund. In addition, SDF needs to market itself and its approach to other donors so that it reduces is reliance on a single donor. 55. Infrastructure Sustainability. Of the 3262 village served under SIPP-2, 3006 have operations and maintenance committees and 3005 have started O&M funds to pay for future maintenance. The NJLIP will have to ensure that continued mobilization around the need to plan for the unexpected happens with villagers to ensure the sustainability of these activities. 3. Assessment of Outcomes 56. Since SIPP-2 had a Level 1 Restructuring, the assessment of outcomes is split according to pre-restructuring and post restructuring. Annex 1 outlines the calculation of the ratings. 3.1 Relevance of Objectives, Design and Implementation Rating: Pre: Modest/Post: High 13 57. Relevance of Objectives. SIPP-2’s original PDO was to improve the livelihoods, quality of life and resilience to climate variability, natural hazards and other shocks of the rural poor, especially the left-out poor and vulnerable households. In February 2014, the PDO changed to improve the livelihoods of extreme poor communities and to strengthen the community institutions in selected districts. In terms of PDO elements:  Both PDOs included livelihoods development (income generation) and inclusion (reaching the poor);  The first PDO included the concepts of resilience, particularly the climate variability and natural hazards, and quality of life, but it did not include elements of institutional development, even though there were PDO indicators measuring that.  The second PDO dropped the element of resilience and realigned the PDO to reflect the high investment in community institutional development. 58. Both PDOs were relevant to the goals of the Government of Bangladesh and the World Bank in terms of promotion of livelihoods and inclusion of the left-out poor. In fact, SIPP-2 interventions will be consolidated and scaled up through the Nuton Jibon Livelihood Improvement Project (P149605), which was approved in 2015.The project prioritized poverty reduction and inclusion―emphasizing access of poor to assets and means of production and strengthened delivery of human development services to the poor―which are top priorities of the Government as expressed in its national development strategy, the Seventh Five Year Plan (FY16-20) and its Vision 2021. Similarly, in 2013, the World Bank committed itself to the twin goals of eliminating extreme poverty by 2030 and boosting shared prosperity, measured as the income of the bottom 40 percent in any given country. The new Country Partnership Strategy (FY16-20) supports the twin goals, and specifically for Bangladesh, emphasizes the need for increased human capital development and job creation to boost growth and advance poverty reduction. The activities in the CPF are structured around three focus areas of growth, social inclusion, and climate and environmental management. 59. While the original PDO was changed to better reflect the indicators and project design, it remains relevant to the goals of the Bank or the Government of Bangladesh. The 7th Plan’s Core Targets in the Context of the Vision 2021 include “environmental, climate change and disaster risk reduction considerations are integrated into project design, budgetary allocations and implementation process�. 7 The CPF supports the government’s adaptive delta management program, which “seeks to ensure that sectoral investments take into account the long term uncertainties related to climate change and growth�.8 60. Relevance of Design. The project components remained the same throughout the project though allocations across components did change. The theory of change underlying the design of 7 Government of Bangladesh. 2015. Seventh Five Year Plan FY2016 – FY2020: Accelerating Growth, Empowering Citizens. Prepared by the General Economics Division (GED), Planning Commission, Government of the People’s Republic of Bangladesh 8 International Development Association, International Finance Corporation, Multilateral Investment Guarantee Agency. 2016. Country Partnership Framework For Bangladesh For The Period FY16-FY20. March 8, 2016. Prepared by the Bangladesh Country Management Unit, South Asia Region. Report No. 103723-BD 14 SIPP-2 is to identify the poor, organize them and build their capacity, get them access to assets, training, and finance to overcome obstacles to productive livelihoods, and they will begin to climb out of poverty.  The goal of inclusion, organization, and capacitation was addressed through component A1 and tracked through PDO indicators 1 and 2 along with relevant intermediate outcome indicators.  The goal of access to assets, infrastructure and social services, and livelihood finance was addressed through Components A1 (establishment and management of the VDRRF), Component B (creating systems for linking to government and markets), and Component C1 (establishing strategic partnerships to crowd in services for the poor). 61. The original PDO—to improve the livelihoods, quality of life and resilience to climate variability, natural hazards and other shocks of the rural poor, especially the left-out poor and vulnerable households—addressed the inclusion and livelihoods development goals of the project design; however, it included other elements that were not explicitly measured by any PDO indicators. The goal of reduced vulnerability to climate and natural disaster risk was mainstreamed into the project design, however. Risk for natural disasters was part of the upazila selection criteria and SDF introduced participatory vulnerability analysis as part of the village planning process to underpin investment priorities, particularly for infrastructure. The appraisal team did note that “the three outcome indicators should be taken together as an ‘index’ for resilience that addresses communities’ social, physical and financial coping capacities.� 9 The PDO element of “quality of life� is not explicitly measured by a particular PDO or intermediate outcome indicator, and the significant project activity of institutional development was not included in the PDO. 62. The restructured PDO—to improve the livelihoods of extreme poor communities and to strengthen the community institutions in selected districts—aligned much more clearly to the design of the project in terms of inclusion, livelihoods development, and institutional development. 63. Relevance of Implementation. Use of the CDD approach was relevant to the project given the objectives of inclusion, institutional development, and livelihoods development, and it supported the theory of change underlying SIPP-2 (and all livelihoods projects). The approach moves SDF beyond being a provider of finance to being a facilitator of the poor to organization for their own development. The finance helps develop livelihoods, and the technical assistance/training help makes the livelihoods sustainable. Under SIPP-1, the project demonstrated that the approach could target the left-out poor, and that the left-out poor were more than capable of borrowing and repaying loans to support their livelihoods. The MTR assessment reaffirmed the appropriateness of the approach to achieving project goals. 3.2 Achievement of Project Development Objectives Rating: Pre: Substantial/Post: High 9 SIPP-2 PAD, page 37, footnote 19. 15 64. Although the PDO was changed to exclude resilience and include institutional development post-restructuring, the PDO indicators and their targets did not. By project closing, all PDO indicators were exceeded. Basically, the revised PDO excluded resilience and quality of life, but the project components continued to have activities potentially directed to this attribute resilience. Indicators for quality of life were not defined, but the borrower ICRR includes findings from FGDs to address this element. 65. Assessing the achievement of the PDO will track the following indicators: a. Pre-restructuring: the three PDO indicators, supported the intermediate outcome indicator on the use of the VDRRF, general achievements on resilience planning and training, and FGD feedback on quality of life; and b. Post-restructuring: the four PDO indicators, supported by achievement of inclusion, financial access, livelihoods activities, and FGD feedback from technical reports. 66. Creation of transparent, inclusive, and accountable village institutions. SDF mobilized 1855 new villages under SIPP-2 and continued supporting the 1407 villages created under SIPP-1. Villages self-identify to join the project and are admitted to the project upon completion of readiness milestones, at which point, the project mobilizes the villages to form institutions—the Gram Smiti (GS), Gram Parishad (GP), Sanchay Sangrakhan Committee (SSC), Village Credit Organization (VCO), and the Nuton Jibon Groups (NJG)—which provide a platform for the poor and HCP to come together, identify problems and solutions, and access the funds necessary to work toward solutions. 67. GPs and GSs were the primary village organizations overseeing all aspects of the project implementation. Under the GS, the VCO managed the Shabolombi Fund (SF), which provided loans for income generation investments. NJGs comprising 10-15 individuals who contribute savings for inter-lending within the group form the base of the institutional pyramid. They federate into SSCs, which manage savings for the NJG member groups and facilitate applications to the SF. This indicator refers to the village level institutions —GPs and GSs—and their effectiveness is based on how well the village meets its milestones for funding and how well the funds provided to the village are managed. The BIDS mid-line survey for the impact evaluation found that 83 percent of village institutions were functioning in a transparent, inclusive and accountable manner by project mid-term, so the target of 75 percent had been met by mid-term. Assessment at end of project shows that this had increased to 92 percent of these institutions meet the criteria of transparency, inclusiveness, and accountability. In addition, by the end of the project 37,000 women and 2400 men from the poor, HCP and vulnerable target group were holding decision-making positions in community institutions. 68. Once the institutions are established and functioning, the GS receives the Village Development and Risk Reduction Fund (VDRRF) which includes five sub-funds: (a) one-time grant facility for the vulnerable poor; (b) a youth employment fund; (c) a youth skills development fund; (d) the Shabolombi Fund (SF) for livelihoods finance; and (e) a fund for village investment in priority infrastructure and capacity building. Options b, c, and d operate as revolving funds, but only the SF is managed by the Village Credit Organization, the others are managed by the Gram Samiti. 16 69. Reaching target beneficiaries, especially women. This indicator was added during the Level 1 Restructuring in February 2014 as a core sector indicator with a sub-indicator that measured the percentage of women among the beneficiary population. While it was not previously in the results framework, SDF had been collecting the data (even during the SIPP-1 project), so this could be measured. By project end, SIPP-2 had served 658,661 households (exceeding PAD target of 652,000) and provided to villages covering 658,661. Based on the PAD assumption of 5.5 people per household, this represents 3.4 million direct beneficiaries, of which 94 percent were women (exceeding PAD target of 90 percent).Early during implementation of SIPP-2, SDF focused its mobilization efforts on women. This is consistent with good practice from Bangladesh and other countries, where program experience shows that women are more likely to act collectively and repay loans, and that the income increase controlled by women is more likely to accrue to the entire household. By the end of the project, 94 percent of direct beneficiaries were women, and as shown above, a large proportion of leadership positions are held by women. This was also evident during field visits of the ICRR missions. 70. The information of impact on women is a bit mixed. The impact evaluation, noted a significant impact on women’s access to finance and women’s savings activity (see section 3.5). However, it did not find a significant impact on women’s agency on economic activities—most of which are family enterprises 10 —such as spending and investment decisions. The borrower completion study, which HH interview, participant’s observation, FGD, KII, and case studies— but on a significantly smaller sample size of 30 households—found that women reported increased mobility, increased respect from family members and the community, reduced domestic violence, and increased role in household decision-making.11 It could be that the project, thus far, has impacted women’s household and civic agency, but economic agency (particular give that most livestock and agricultural enterprises are family enterprises) is still to be achieved. This is an area for further investigation, but clearly the project has had as significant impact on women’s access to finance, on their emerging leadership roles in communities, and on their own perceptions of their level of empowerment. 71. Increasing household income.SIPP-2 sought to increase household income by 50 percent for at least 50% of the targeted households. The impact evaluation data showed that 62 percent of target households had met this goal.12The end-line survey showed that SIPP-2 had a significant impact on income (+BDT 12,000) and on net income from enterprises (+5788). The findings are also supported by a review of livelihoods and finance activities toward the end of SIPP-2. 72. For the vulnerable poor, households with essentially no source of income, SIPP-2 provided one-time grants of BDT 5000 to start some form of income generating activity. By 10 Mghenyi and Kitsuki. 2016. “Impact Evaluation for SIPP -II.� Data report from the impact evaluation of the Empowerment and Livelihood Improvement "Nuton Jibon" Project. Washington, DC: World Bank. Mimeo. 11 Hussein, Akhter. 2016. Borrower’s Completion Results Rep ort. Mimeo. 12 SDF internal reporting based on Mghenyi and Kitsuki (2016): Impact Evaluation for SIPP-II, Mimeo 17 project end, SIPP-2 had identified 18,717 vulnerable households and provided grants to 17,712 of them. It also continued to provide livelihood technical support to the grantees under SIPP-1.13 73. According to this review, 432,919 NJG members engaged in income generating activities by taking a loan from the Shabolombi Fund, and 60 percent of those took loans for livestock sector activities—primarily beef fattening and dairy. Figure 1 Sector breakdown of SF fund uses Fisheries Others 4% 1% Small trade 18% Agriculture sector 12% Livestock sector 60% Transport sector 5% 74. Based on data collected by the livelihoods and community finance units, members who engaged in 2 cycles of SF loans saw a 39 percent increase in net income over the first cycle. Those who engaged in 3 cycles of SF loans (steadily growing their enterprise) saw an 89 percent increase in net income over their first cycle. Data show that of the 432,919 NJG members have taken one loan, 235,619 NJG members have completed a 2ndloan cycle, and 122,889 members have completed 3rdloan cycle. Figure 2 Growth in net income from multiple loan cycles for livelihoods 1st TK 2nd TK 3rd TK Cycle 21,919 Cycle 30,347 Cycle 41,436 75. The impact evaluation data also shows that the project had a significant impact on livestock related livelihoods (beef fattening, eggs, dairy), including livestock sales for large and small ruminants and poultry. In agriculture, the only crop showing significant impact is paddy production, which is the major crop in the project area.14 76. Income growth is further supported by the formation of production groups. SIPP-2 mobilized 90,527 producers into 5561 Producer Groups (PGs) to achieve economies of scale in accessing input and output markets. PGs also facilitated access to training, which had led to about 60 percent of PG members are using scientific production methods which is leading to significant productivity increases. Business training and increased bargaining power have helped about 60 percent of PG members are able to sell their products competitively higher prices. 13 The vulnerable poor are a relatively small proportion of the target population. These are the poorest of the poor. In the baseline survey, the team was identifying 1-5 families in a village, and some villages had none. 14 Mghenyi and Kitsuki (2016) 18 77. SDF also had a special strategy to promote youth employment—both wage and self- employment—through skills enhancement, linking to job opportunities, and financial assistance for enterprise development. Under SIPP-2, SDF created 41,849 of the target 50,000 direct jobs According to a special study commissioned by SDF, which reviewed to review youth development activities, 80 percent of males and 95 percent of females were unemployed before receiving skills training, after which 99 percent and 98 percent, respectively, reporting being gainfully self-employed. Of the youth who pursued wage employment, there were similar 15 movements from unemployed to employed. However, the impact evaluation shows no impact on incomes from wage employment. According to the current CPF, having formal secondary education is a key determinant of accessing remunerative wage employment. According to the SIPP-2 baseline survey, the average years of education for males aged 17-35, which is the target group for SIPP-2 youth employment initiatives is 4.3, and for females it is 4.4. This indicates that a more nuanced strategy for wage employment will be needed. 78. Benefitting from access to infrastructure. SIPP -2 sought to benefit 80 percent of target households with infrastructure or social investments. Every village that met the requirements for participating in the project, received a Gram Samiti office building, so every village had that investment as a baseline. 16 The design of the buildings included a sanitary latrine, and most served as safe spaces for villagers during floods. 79. The mobilization process helped villages identify additional priority investments in infrastructure and social services based on the participatory planning process and vulnerability assessment. As shown in table 6, the large number of subproject invested in infrastructure that would benefit a broad section (if not all) of the village, even non-target households. In addition, latrines installed on school grounds or serving multiple families benefitted a larger number of people than the number of subproject would indicator. 80. By project end, SIPP-2 infrastructure and social investments exceeded the target by benefitting 92 percent of target households. According to SDF MIS, 3006 of the 3262 villages had an O&M plan and 3005 had started an O&M fund, thus contributing to sustainability. However, the technical audit of 51 villages found only 47 percent of sample villages had O&M plans. The study surmises that this could be due to the recent completion of subprojects. Still, O&M is historically a problem area for CDD projects, and this will have to be closely monitored and supported by technical assistance to ensure the sustainability of investments. The same study found that 83 percent of investments were in working order during the audit visit. Table 2 Outputs for Community Infrastructure and Social Fund Infrastructure No. of Subprojects Achieved Gram Samiti Office17 1,937 193418 15 Iqbal, S. M. Hasan. 2015. “Survey / Impact Assessment on Youths Employed in Different Industries/Places through SIPP Project.� Mimeo. 16 There were a few villages that had not constructed their GS office building by project end due to legal issues over the land to be used for the building. According to the project rules, in accordance with World Bank safeguards, these buildings will not be constructed without clear title. 17 The Gram Samiti offices include a sanitary latrine, so the number reported for latrines is lower than actual achievements. 19 Table 2 Outputs for Community Infrastructure and Social Fund Infrastructure No. of Subprojects Achieved Gram Samiti Office17 1,937 193418 Earthen Road 964 615 km Culvert (Ring/Box) 854 2541 Tube-well 805 4198 Latrine / Urinal 63 67 Market development 2 2 School Development 71 71 Drain 61 5993 m Other 195 330 81. The impact assessment of SIPP-2 reports no significant impacts on infrastructure in treatment villages; however, the number of subprojects and the size of the subprojects is small relative to total need, so it is possible that there were not enough subprojects to register a significant impact.19 Focus group discussions undertaken for the Technical Audit report that the investment in the Gram Samiti office buildings has raised the social status of the GS members. The office serves as a status symbol for GS members, and FGD participants noted an increase in respect from other villagers. Some FGD participants plan to generate income for the GS in the future by renting out the GS office to others for various purposes to others —which could raise additional funds for O&M. Some FGD members from Barisal and Jamalpur reported already using the local offices as shelters during natural disasters, and offices in Jamalpur and Sherpur had been used to host immunization programs. The earthen roads in these 51 villages improved access to markets and to critical health care. The culverts significantly reduced water pollution, water logging and erosion of roads. Linking back to the vulnerability reduction elements of project design, four villages raised their school grounds, which resulted in another safe place for villagers during a natural disaster in addition to removing stagnant water after rains, which reduced the number of diseases contracted by children. In addition, the infrastructure project generated a significant number of labor days for beneficiaries during construction. The Technical Audit reports that, on average, the highest labor days generated were for GS Office (11 days/sqm), followed by culverts (6 days/cubm), and bridged (5 days/m).20 82. Progress towards resilience to climate variability, natural hazards and other shocks. Based on the assumptions by the project appraisal team, the significant achievements on mobilization, institutions, and infrastructure combine to reduce the vulnerability of the population. Certainly, the members have more income to meet their consumption needs, and the access to affordable finance and increased savings can provide a cushion in the event of a natural disaster or other hardship. Many infrastructure investments were developed with disaster risk management principles in mind—e.g., GS buildings that can serve as shelters during disasters, improved drainage and culverts to reduce waterlogging and flooding, and raised school yards to provide a safe space during disasters as well. 18 Three were nearing completion but halted due to legal issues as land for GS offices needed to be donated or purchased at market prices with appropriate documentation to comply with OP 4.12 19 Mghenyi and Kitsuki (2016): Impact Evaluation for SIPP-II, Mimeo 20 Zaman, Dr. Asif M. Technical Audit & Impact Assessment of SIPP-II. Paper submitted to SDF, October 2015. This examined the impact of infrastructure investments in 51 village Samitis across Jamalpur, Barisal and Rangpur regions. 20 83. As mentioned before, SIPP-2 also integrated risk for natural disasters within the project design. SIPP-2 incorporated environment, climate change, and risk reduction (ECCRR) into its village planning activities so that risks were integrated into village The PAD included one intermediate outcome indicator directly related to disaster vulnerability: village organizations have applied participatory methods for risk mapping /vulnerability analysis and implemented risk reduction and preparedness methods. While this indicator was dropped at restructuring, ultimately 2359 villages (72% of total villages) carried out participatory vulnerability analysis. Disaster volunteer teams were formed and trained in 2353 villages, and 2116 villages had purchased disaster management tools. 84. Quality of Life. The PDO element of “quality of life� was not defined in the PAD and had no correlating indicators; however, In FGDs from the analysis of Technical Audit and Impact Assessment Report of SIPP-II participants reported that their social status within the village increased dramatically, and they attributed this to their improved economic condition, the improvement of their living standard, and the construction of GS Office building and other infrastructure. Examples cited, include: rich and the middle class villagers showing them honor; being invited for functions, such as social and religious rituals; having the rich and middle class address them honorably and enquire about their families. They also reported that their children can play with the children of rich and middle class members. Feedback from FGD’s indicates less impact on social cohesion within the village, and less than one-fourth of the FGD participants in all surveyed districts admitted that social harmony is in the increase —e.g., less conflict, social value, rich-poor relationship. Data from the formal impact evaluation survey only shows a small, but insignificant decrease in conflict [estimated impact -0.026, P-value (0.612)].21 Achievements in income appear to have impacted food security and quality of diet. While the formal impact evaluation does not record a significant increase in diet, it does capture a change in the quality of the diet, with a significant increase in consumption of meat, dairy, pulses, and even sweets among treated households.22 FGDs for the borrower completion report that they are able to eat three meals a day and no longer experience hunger. According to FGDs for Technical Audit and Impact Assessment Report of SIPP-II, almost 100% members of the families are using sanitary latrines because their economic condition has improved and their consciousness and awareness has increased. The FGD participants strongly stated that the GS of SDF helped them increase their earning sources and ensure their self- employment, as a result they can afford sufficient food and clothes and can send their children to school.23 3.3 Efficiency Rating: Pre: Substantial/Post: Substantial 85. The EFA conducted as part of the PAD projected benefits to some 1,500 villages and 300,000 poor and vulnerable households, deriving from: (i) investment in livelihood activities; 21 Mghenyi and Kitsuki. 2016. 22 Mghenyi and Kitsuki. 2016. 23 Hussein, Akhter. 2016. 21 (ii) community-led livelihood sustainability; (iii) community infrastructure; (iv) vulnerability reduction; and (v) community-led management. 86. Economic and financial rates of return were provided for livelihood activities, as well as for the other categories defined above (table 1). However, no indications were provided in the PAD EFA or accompanying documents as to how these rates of return were estimated, although the conversion process from financial to economic returns was briefly described. 87. SIPP-2 operated in a total of 3,262 villages in three regions (Rangpur, Jamalpur, and Barisal) participated in SIPP-2 with the 658,661 Nuton Jibon group members– in both cases more than double the number initially envisaged. Of the almost 800,000 IGAs reported, 60 percent were for livestock (22 percent dairy, 22 percent beef and the remaining 16 percent small stock and poultry). Data from the Bank internal analysis of household expenditure data is also used.24 88. The Technical Audit of SIPP-2 infrastructure investments and their impacts did a cost comparison of SDF community infrastructure to similar in the Local Government Engineering Department (LGED) and the Department of Public Health Engineering (DPHE), and the results show significant efficiency in some but not all investments (see table 3). One exception is the cost of latrines, for which SDF’s maximum cost is TK 87,037 compared to Tk 10,250 for LGED and Tk 30,000 for DPHE; however, the study explained that SDF builds permanent sanitary latrines (for example, they include a tube-well, concrete foundation, etc.) compared to the temporary structures built by the other departments, so the price differential is understandable. Table 3 Comparison of infrastructure unit costs Measurement SDF Max Difference from Max Rate Investment Units Rate LGED DPHE Earthworks Tk/cubm 108 -113 15 RCC work Tk/cubm 5803 -6315 -2745 Brick work Tk/cubm 4666 -702 -430 GI pipe Tk/m 394 -586 -400 PVC pipe Tk/m 96 -602 -1284 Palasiding Tk/m 3080 -1558 Tubewell Tk/no. 62541 n/a 7541 89. Financial Returns (both IRR and NPV) calculated are presented in Appendix 6 and summarized in Table 8 below. At the financial level, these returns are very strong, in large part due to the assumptions of returns generated after the completion of the implementation period. In order to show this clearly, FIRRs and FNPVs are given both for the implementation period (to 2015) and are estimated at 37.7 percent and BDT 2,275 million, as well as until 2030, when the community level funds are assumed to be virtually exhausted. This latter duration yields a FIRR of 52.6 percent and an FNPV of BDT 9,058 million. 90. These rates of return compare strongly with those in the PAD, which were projected to reach just under 24 percent. The increased ex-post rate of returns to the project over that projected in the PAD can largely be attributed to the more than doubling of beneficiary populations over 24 Mghenyi and Kitsuki. Impact Evaluation for SIPP-II, Mimeo, World Bank. 2016 22 those foreseen during preparation. Thus benefits accrued to a much greater number of participants without any increase in initially budgeted costs. It is also likely that any inclusion of labor costs for enterprise operations - that is, where labor costs are not charged to the enterprise, but are seen as part of the overall net income figure – will increase the net returns calculated. In fact, the World Bank Impact Analysis already cited found an increase of 13.7 percent in day labor for project beneficiaries, suggesting that participation does result in increased labor use. These two considerations suggest that although the true FIRR may well be lower, the investment would still have generated a worthwhile return. 91. Economic Returns are adjusted from financial returns largely on the basis of the divergence of the Nominal Effective Exchange Rate (NEER) and the Real Effective Exchange Rate (REER) with respect to the Bangladesh Taka. The recent (late 2015) Article IV consultations between the IMF and the Government of Bangladesh25 indicate an estimated NEER of BDT 95.4 per US dollar, while the REER is calculated at 120.5, on the basis of significantly higher inflation rates in Bangladesh than is the case in its main trading partners. The REER thus exceed the NEER by 26.3 percent, and the cost of project funds (although not those contributed by the beneficiaries) has been increased to reflect this divergence. As few traded resources are used in the small scale investments undertaken by project beneficiaries, no specific adjustments for internationally traded goods are required. 92. A further factor usually considered in undertaking adjustments for economic, rather than financial, costs is the real cost of labor. No recent shadow labor rates could be identified for Bangladesh, but the country enjoys a relatively low unemployment rate (4.3 percent in 2014 according to World Bank data). However, national underemployment rates have been estimated as high as 40 percent, 26 suggesting that financial prices for labor are probably overstated. Nevertheless, no data was collected during the implementation with respect to the labor component of investments, and the detailed enterprise figures shown in Appendix 1 do not include labor, suggesting that these costs were subsumed within the net returns. As a result, no adjustments have been made to labor costs for the economic analysis. 93. With the higher value of project funds (provided in SDRs), the EIRR and ENPV for the project are lower than those estimated on a financial basis—see annex 3. Appraisal End Of Project Indicator Calculations Calculations ENPV (BDT billion) 2.9 1.9 ERR 25.3% 33.0% FNPV (BDT billion) 3.4 2.3 FRR 23.8% 37.7% 3.4 Justification of Overall Outcome Rating 94. According to World Bank ICRR guidelines, the overall outcome rating based on the standard 6 point scale. 25 Bangladesh. Staff Report for the 2015 Article IV Consultations. IMF. January 5, 2016. 26 http://www.indexmundi.com/bangladesh/unemployment_rate.html 23 Against Original PDO Against Revised PDO Relevance Modest High PDO Achievement Substantial High Efficiency Substantial Substantial Overall rating MS = 4 S=5 Disbursement Weight 66% 34% Final Rating 2.64 1.7 Final Score 4.34 Overall Rating Moderately Satisfactory 95. SIPP-2 met or exceeded all of its PDO indicators by end of project. Overall, progress was moderately satisfactory before restructuring and satisfactory after restructuring, giving a weighted score of substantial. 96. Similarly, the initial PDO was relevant to government and Bank objectives/strategies; however, it was overly complex and not directly reflected in the PDO indicators. The revised PDO more accurately reflected the true project objectives and remains relevant to the goals of the Government of Bangladesh and the World Bank. The weighted score is substantial. 97. While there was no mid-term economic analysis undertaken, so the rating is based on the end of project EFA, which show the calculated and projected ERR and FRR exceeding PAD expectation. That rating is substantial, particularly because it shows that the project design should yield positive returns for several years. 3.5 Overarching Themes, Other Outcomes, and Impacts (a) Poverty Impacts, Gender Aspects, and Social Development 98. SIPP-2’s impact on gender is mixed based on the results of the impact evaluation, a nd it raises the question of what additional interventions might be needed to achieve empowerment for women. The impact evaluation shows a clearly positive impact on the loan amounts taken by women and controlled by women. Table 4 Impact on Women's Access to Finance Baseline mean value Estimated impact Indicator (standard deviation) (P-value) 10,593 2,759*** Loan amounts (27466) (0.008) 5,803 3,961*** Loan amounts taken by woman (12527) (0.000) 493.9 1,9501*** Loan amounts controlled by woman (4983) (0.000) 0.639 0.101*** Proportion of loan amounts taken by woman (0.453) (0.000) 0.624 -0.079*** Proportion of outstanding loan (0.367) (0.000) 0.0588 0.127*** Proportion of loan amounts controlled by woman (0.234) (0.000) 0.784 0.036* Proportion of savings by woman in total savings (0.393) (0.057) 24 Source: Mghenyi and Kitsuki (2016): Impact Evaluation for SIPP-II 99. However, the IE survey shows no impact on women’s control over business decisions, control over profits from business, or control over cropping decisions. Feedback from focus group discussions done as part of some of the thematic studies indicate that women have greater mobility within the village and outside of the village. Some report less domestic violence, and an increased role in household decisions such as education and marriage of their children, purchasing household items. Since the IE will continue in NJLIP, it would be good to add survey questions that start to describe the level of impact on women’s agency—mobility, household decision making, reduced domestic violence, working outside the home, and hours per day spent on unpaid work relative to the males in the household. (b) Institutional Change/Strengthening 100. One of the critical issues for the success of the model introduced by SIPP-1 and refined under SIPP-2 is the strength, sustainability, and effective performance of SDF. 101. SDF has pioneered a new form of development—community driven—in Bangladesh over the last 13 years. GoB has supported SDF’s efforts as part of its vision to cover the entire nation with this approach to bring the left-out poor into its growth agenda. Estimates based on the implementation experience of SDF indicate that nation-wide expansion would take no less than 10 to 15 years to complete. Therefore, SDF needs to develop a strategy for managed growth that would allow it to expand while maintaining quality. To achieve nation-wide coverage, 64 districts would be involved and all districts would be fully covered. This involves a several fold expansion in SDF activities, though of course matched by a reduction in activity as groups exit through graduation. This expansion will take no less than 10-15 years to complete. An assessment undertaken by a consultant for the World Bank in the wake of identifying the funding gap during implementation, outlined a series of measures for improved effectiveness:  Clear articulation of SDF's role and responsibility and its accountability. In response to the funding gap, SDF drafted a vision and organizational strategy that would guide its human resources and other management decision-making. This vision was agreed with GoB, and SDF began implementation during SIPP-2. The follow-on project should solidify the reforms.  Improvements to its organogram and job descriptions. In response to the funding gap, SDF rationalized its organogram, streamlining teams at different levels of the organization and defining the staff complement required at each level. This was agreed with GoB as well, and has been implemented. The creation of Community Professionals—numbering 16,000 by the end of SIPP-2—contributes significantly to SDF’s ability to implement effectively on the ground. As CPs’ skills increase, and as the Nuton Jibon Community Societies to which they belong are strengthened, SDF should regularly review the existing organograms and job descriptions to ensure the proper separation of tasks for effective service delivery.  Recruitment of quality staff and rigorous performance management. SDF has instituted, with the help of the Bank, and new HR policy and strategy that includes clear job descriptions and performance indicators that are reviewed annually. SIPP-2 tracked 25 progress on this reform—monitoring if staff performance appraisal results indicate 70 percent of staff performing at Grades A and B—in its results framework. By the end of SIPP-2, 89 percent of staff were meeting that criteria, and the borrower completion report exercise noted the professionalism, dedication, and technical ability of field staff.  Upgraded internal processes quick adoption of IT systems to raise operational efficiency. SDF rolled out five modules as part of its computerized MIS system: (a) Financial Information System; (b) Project Monitoring System; (c) Procurement System; (d) HR & Payroll System; and (e) Inventory System. It is also piloting the use of ICT for data collection in the project monitoring system at the village level. These technologies, once fully deployed system-wide should increase operational efficiency even more.  Effective monitoring and evaluation arrangements. SDF has developed a thorough and rigorous monitoring and evaluation system, which is described in section 2.3. The follow-on project will continue to refine this system.  Adequacy and continuity of funding. At the beginning of SIPP-2 SDF relied completely on World Bank funding, even for its staffing costs. By the end of the project, GoB had created an endowment fund of US$21.5 million earning about 10 percent in income annually, which would cover the staffing costs of SDF to allow for continuity of operations should a break in financing flows occur. Moving forward, it is absolutely critical for SDF to market its development model and its achievements to the corporate sector and to other donors to diversify its funding sources. This will require documenting and communicating the CDD approach for the target donor/partner audience and using the results of impact evaluation data and other thematic studies to market SDF as the partner of choice in Bangladesh for poverty reduction. The consultant also recommended that SDF expand the non-government role in its governance and accountability to gain greater credibility and additional financing support, which can be done through the governing board. 102. Overall, SDF has made significant progress in developing its organizational structure and operations. It has also identified the additional actions that are needed to ensure the sustainability of the organization. (c) Other Unintended Outcomes and Impacts (positive or negative) 103. Impact on food security. Feedback from FGDs conducted for the technical audit indicates that food security has increased among SIPP-2 beneficiaries. Participants claim that they are eating more, are no longer going hungry, and are having 3 meals per day.27According to the impact evaluation data, there was no significant change in overall food consumption; however, data does show that diets seem to have diversified—milk and dairy product consumption up 39 percent and meat and egg consumption up 49 percent. 28 This suggests that increased income could be increasing the quality of diets and dietary diversity within the project. This would be a good sign because the diet choices appear to be healthy, often income increases lead to increase 27 Zaman, Dr. Asif M. 2015 28 Mghenyi and Kitsuki (2016): Impact Evaluation for SIPP-II, Mimeo. World Bank. 26 in junk food consumption. However, the consumption fruit and vegetables was not increased, except for spinach and data. As NJLIP will support a specific nutrition function, it would be good to examine consumption patterns more methodically to see if increased income is translating into better diets, and if additional inputs (e.g., behavior change communication) could increase consumption of fruit and vegetables. 3.6 Summary of Findings of Beneficiary Survey and/or Stakeholder Workshops 104. There was no beneficiary survey; however, the special studies commissioned by SDF to assess the impact of SIPP-2 did include focus group discussions and key informant interviews that are reflected in the sections above. 4. Assessment of Risk to Development Outcome Rating: Moderate 105. SIPP-2 outcomes of SIPP-2 investments can be divided into three areas: institutions, financial systems, and livelihoods. The overall risk to these outcomes can be considered moderate because they have been flagged by both SDF and the Bank, and mitigating actions have been started and will be continued in the follow-on project. 106. Risk to community institutions. SIPP-2 supported 3262 villages, each of which has a GP, GS, VCO, and multiple SSCs and NJGs. Together these institutions serve as a platform for the poor, HCP, and vulnerable to participate effectively in their village, identify problems and solutions, and access the finance and technical assistance necessary to engage in profitable and sustainable livelihoods. The institutional structure needs to be continuously refreshed with retraining and new (advanced) training to ensure sound leadership in perpetuity and to ensure that the institutions remain representative, transparent, and accountable. SDF, and increasingly the NJCSs will be critical to providing the necessary support to keep these institutions strong. Toward the end of SIPP-2, SDF and the Bank agreed to legally register Gram Parishads as Joint Stock Companies under the Societies Act of 1860, and this process has begun. This is a very positive development as it will give the GPs legal status, allowing them to enter into contracts and engage more effectively with government, other civil society organizations, and the private sector. 107. Nuton Jibon Community Societies. SIPP-2 formed 124 Nuton Jibon Cluster Community Societies (NJCCS) and 15 Nuton Jibon District Community Societies (NJDCS), including forming key committees and providing initial training to staff. All NJCSs have offices and have received BDT 131 million for capacity building and institutional development. Part of the funds have been put into an accounting earning approximately 10 percent interest to generate a sustaining income of the NJCSs, which will contribute to their sustainability. A total of 2910 villages have joined an NJCS and 95 percent had deposited their membership fee—amounting to BDT 27.64 million—by the close of the project. The NJCS are the linchpin of sustainability for the SDF system, as they will take on the sustaining services to village members as SDF staff are deployed to new areas. If they cannot forge partnerships that crowd in resources for their members, provide relevant and high quality technical and institutional support to their members, their relevance will wane. The NJDSs are being formally registered under the Societies Act, 27 which, as with the GPs, will give them the important legal status to carry out their function. Building the capacity of these organizations is, perhaps, the most critical need moving forward, and the follow-on project is addressing this need. 108. Risk to livelihoods and incomes. Most of the livelihood activities that were supported by SIPP-2 built on and strengthened the existing livelihoods of the target beneficiaries —e.g., cropping (largely paddy) and livestock production (largely beef fattening and dairy). This bodes well for sustainability because the producers are familiar with the production processes and have a solid understanding of local markets. The sustainability of these livelihoods as sources of incomes is solid. The risk to livelihoods lies in continued access to finance by the beneficiaries to maintain (and ultimately grow) their enterprises and access to inputs and lucrative markets to increase profitability. To address this, SDF is increasing the size of loans available from the Shabolombi Fund, and it has started to link producers to MFIs and commercial banks to meet their financing needs. The creation of producer groups (5561 created during the project) is the direction that will help to make the livelihoods more sustainable in the long run. 109. Risk to the implementing agency. SDF’s financial position was strengthened during SIPP-1 when GoB created an endowment fund of US$21.5 million that is expected to yield 10% per year. This provides continuity for current staff under SDF’s organogram. However, the critical priority for SDF is to diversify its sources of finance by marketing its approach to donors and the corporate sector. The plan is to pursue this goal under the follow-on project which is currently underway (see section 3.5(b) for more information). 5. Assessment of Bank and Borrower Performance 5.1 Bank Performance (a) Bank Performance in Ensuring Quality at Entry Rating: Moderately Unsatisfactory 110. The Bank team had the difficult task of designing the project while implementing the SIPP-1 operation, and it did a good job of taking on lessons from implementation of SIPP-1. The team also designed a robust governance and monitoring system covering the national to community level that has kept corruption/mismanagement to a minimum in a very complex setting. However:  the team’s choice of PDO was not linked sufficiently to the PDO indicators and did not sufficiently reflect the project focus on institutional development at the community level.  the team created an extensive list of monitoring indicators with unclear measurement units and overlaps in measuring between components. These caused confusion in measurement of indicators that were not clarified until 2014.  the team should have ensured the full, secure financing of all project costs instead of keeping some operating costs of SDF dependent on the additional financings from the on- going SIPP-1 project. This ultimately led to the need for the Level 1 Restructuring. 28 111. Given the number of problems and the implications of those problems, the team’s performance has to be rated moderately unsatisfactory. (b) Quality of Supervision Rating: Moderately Satisfactory 112. Three Bank TTLs supervised implementation during the life of the project, but the Bank was careful to overlap the tenure of the TTLs whenever possible to ensure a smooth transition. The Bank team fielded implementation support missions twice per year, and quickly responded to issues as they emerged—e.g., project restructuring, issues around fiduciary management. The team also provided sound technical advice and suggestions to the SDF team on implementation issues—e.g. workshops offered on FM issues at the outset of the project, recommendation on environmentally sustainable innovations―and accepted realistic adjustments to project design— e.g., moving from Community Professional Centers to Nuton Jibon Community Societies to better support long-term sustainability of the institutions. Moreover, the team worked closely with SDF to design the NJLIP follow-on project to start in time to overlap with SIPP-2 so that there would be no break in funding for the agency. 113. One shortfall on the part of supervision was the handling of the results framework. The initial ISR is meant to lock on the results and outcome indicators; however, the ISRs did not the indicators until 2013. There was a delay in setting baseline targets due to the baseline survey and the fact the SIPP-1 closed after SIPP-2 became effective—both of these were sources of the baseline. However, not getting the list of indicators from the PAD into the ISR format early and clarifying the measure units (e.g., headcount vs. households, percentage vs. numbers) led to unnecessary confusion—e.g., the restructuring paper dropped indicators that had not been in the results framework, and measurements contradicted each other from one ISM to another. (c) Justification of Rating for Overall Bank Performance Rating: Moderately Satisfactory 114. The Bank lending team was pressed to design and deliver a project while still implementing and closing out SIPP-1 and its three additional financings; however, very important details were missed that led to a funding crisis for the client and a restructuring on the part of the Bank. The original PDO, while noble in aspiration, did not adhere to the principles of being specific and measurable, and the project design did not fully align to it. 115. The supervision team did a good job trying to bring consistency in TTL switchovers by overlapping the people for handover. They managed one Level 1 and two Level 2 restructurings and a fiduciary problem, in a timely and professional manner. The presence of TTLs in the country or the region, and a technical staff member based in the Dhaka office increased the ability of the team to provide close support to the SDF team. The fiduciary and safeguard staff also contributed proactively to the technical quality of the project by providing workshops on various approaches, and by recommending and supporting the implementation of innovative and environmentally sound investments. 5.2 Borrower Performance (a) Government Performance 29 Rating: Satisfactory 116. The Government of Bangladesh continued to show its strong support for SDF and for the CDD approach by meeting all of the lending conditions presented in the project. More specifically, it reinstated the Governing Board of SDF and formed the audit committee as required by the Bank in a timely manner, and both bodies have been performing throughout the project. GoB also set up the endowment fund for SDF with a commitment of US$15.0 million, which it then increased to US$21.5 million when SDF’s higher funding requirements became clear in early 2013. The actions of the Government were consistent with its vision of having SDF grow to fill this poverty reduction mandate on a national scale, and the support was critical to the success of the project. (b) Implementing Agency or Agencies Performance Rating: Moderately Satisfactory 117. SDF staff designed SIPP-2 with the Bank while implementing SIPP-1 without a permanent managing director or a sitting board. At the same time, it was implementing a massive reorganization and human resource reform as it moved away from contracted partner organizations to using its own staff. At the start of the project, approximately 60 percent of SDF staff were new and had to be trained quickly in order to implement. Despite this, once implementation started SIPP-2 was meeting or exceeding physical targets and disbursement faster than the original plan. The field team architecture where staff in existing districts help new staff and new districts learn the project quickly, as well as a focus on staff training and a clear human resources strategy certainly helped in that outcome. SDF is also to be commended for designing the follow-on project so that there be no break in service between SIPP-2 and the NJLIP follow- on project. Challenges did emerge during implementation—e.g., management and use of the robust M&E system was a challenge for much of the implementation period, delays in recruiting key positions (e.g., senior staff in procurement and financial management) did impact the pace of some activities. Some recommendations made at mid-term regarding simplification of the village matrix were implemented, but confusion over the matrix remained according to focus group feedback at the end of the project. This contributes to M&E challenges. Finally, the introduction of new activities, such as the innovation fund, was not fully successful, but could have been with more effort and drawing on good practice examples. SDF faced an incredible number of challenges when SIPP-2 began—managing the implementation of SIPP-1, reorganizing its operations, and scaling up its staff—but it still managed to meet or exceed most of its targets. It is to be commended for that. (c) Justification of Rating for Overall Borrower Performance Rating: Moderately Satisfactory 118. Both the Government of Bangladesh and SDF worked well to meet the requirements laid out by the Bank to get the project approved, and both rose to meet changing needs throughout implementation. 30 6. Lessons Learned 119. The results from SIPP-2 reinforce the SDF approach as an effective way to identify and include marginalized groups—poor, HCP, vulnerable, women, and youth—into the growth agenda. Evidence from field visits show the enterprising nature of the poor, many of whom are developing diverse sources of income for their households. 120. SDF has created a stairway out of abject poverty—one-time grantinternal lendingShabolombi fundcommercial finance (complemented with technical assistance and external linkages) that should be thoroughly documented in terms of approach and replicated in other regions. Bringing more focus on convergence with government programs that provide critical safety nets for the poor can boost the impact of this stairway, particularly for the vulnerable poor who often have trouble accessing these benefits. 121. Community Professionals are a key element of success for SIPP-2 and should be developed in other CDD projects that want to enhance the sustainability of their investments. The CPs are embedded in the target villages and have the additional capacity to act as change agents within their own communities and beyond. 122. SDF has created a thriving village finance model with over 450,000 NJG members taking loans for livelihoods. Introducing a comprehensive program to identify and screen for members with entrepreneurial attributes and to provide them training and mentorship move some members from self-employment to entrepreneurship, increasing the potential for creating jobs. 123. The formation of producer groups can help to introduce modern cultivation technology to ultra-poor producers and can increase market margins for producers through collective actions, which can boost income growth beyond production increases. 124. Mobilization of women has led to increased confidence, increased access to finance, and increased savings among women. Women report that they are more respected by the their families and communities, that they have a larger role in household decision making, and that they even face less domestic violence. At the same time, the impact evaluation is showing limited impact on women’s agency within their enterprises. These elements would be good to explore further. In particular, lessons from other programs shows a risk of increased domestic violence as women become empowered, and SDF seems to have avoided this. Understanding how would be important for other programs. 125. The phasing of the project and steadily scaling up in geographic focus is a strength of the SDF model. Village institutions that have gone through an entire project cycle are federated once they have achieved a certain level of quality, which seems to support the quality of the second generation institutions. 126. Moving away from partner organizations to SDF staff for implementation seems to have contributed to consistency of support and efficiency of the model. Of course, the CPs and Nuton Jibon Community Societies are a critical element of this system. There appears to be a sense of belonging on the part of SDF staff that is bolstered by the creation of an endowment fund to ensure SDF institutional sustainability. 31 127. The field based mobile Capacity Building Cell (CBC) attached to the Program Division of the National office worked as an independent special purpose vehicle and coordinated all the capacity building activities. The CBC designed program activities and prepared implementation plan for capacity building programs according to the emerging issues from Appraisal and Monitoring Team (AMT), PMA report, observation of the monthly progress review and monitoring meeting at the Regional, District and Cluster levels and the issues identified in internal learning and review meetings at various levels from cluster to national. All Capacity Building events were undertaken as per the demands that come from Cluster, District and Region. 128. Piloting environmentally friendly enterprises, such as vermi-compost, shows additional potential income sources that could be developed around the primary agricultural livelihoods of the target population, and should be scaled up under the Nuton Jibon Livelihood Improvement Project (NJLIP). 129. SDF showed that focused effort to match skills training to labor market opportunities that suited to the ultra-poor population can generate good-paying sustainable employment or enterprises. The project was able to help 51,054 un/under employed youths received skill development training through different initiatives under SIPP II project which added substantial value to the national economy through development of skilled manpower. 130. Development of Nuton Jibon Community Society (NJCS) as second tier institutions created a sustainable platform for villages to continue their development activities in a sustainable manner even after the formal exit of SDF. 7. Comments on Issues Raised by Borrower/Implementing Agencies/Partners (a) Borrower/implementing agencies 131. SDF, the implementing agency, provided thorough comments on the text of the document—providing clarifications on data and on issues related to implementation, and identifying additional lessons, which have been incorporated into the text. The comments and suggestions provided by SDF we well-reasoned and have been integrated into the report wherever appropriate. The primary issues raised by SDF focused on ratings. The rating of the project outcome and the implementing agency—both Moderately Satisfactory—was viewed as too low, considering the overall implementation progress (closing 6 months ahead of time) and the achievement of revised PDO. These ratings were impacted by the low relevance of the original PDO, which was in place for 66 percent of the project implementation period. This is explained in Annex 1. SDF also requested a reconsideration of the Bank’s quality at entry rating (moderately unsatisfactory); however, the combination of problems with the original PDO and issues with project costing represented significant short-comings in the design that led to multiple restructurings. The ICRR reviewers and sector management were in agreement. Additional data and lessons suggested by SDF have been reflected in the ICRR text, and all thematic studies will be added to the project files for future reference by project team. 132. The Ministry of Finance Economic Relations Division received the report in July and they have seen the comments provided by SDF. They have confirmed with SDF that they had no additional comments to add. 32 (b) Cofinanciers 133. There were no institutional cofinanciers. Cofinancing only came from communities through cash or in-kind contributions to subprojects. (c) Other partners and stakeholders 134. Not applicable. 33 Annex 1. Overview of Split Rating for Restructuring 1. Since SIPP-2 underwent a formal Level 1 Restructuring in February 2014, the assessment of the relevance of the project development objective and the achievement of project objectives before restructuring and after restructuring. The exercise requires the author to look at the relevance of the PDO before and after restructuring. Each rating uses a four-point scale: 1=Low/Negligible; 2=Modest; 3=Substantial; 4=High. A score is applied to each element for pre- and post-restructuring weighted according to the proportion of the credit amount that had been disbursed when restructuring occurred. For SIPP-2, the amount disbursed at the time the restructuring was cleared by the World Bank on 19 February 2014 was US$76.39 (66 percent of the credit amount). Therefore, ratings before restructuring were weighted 66 percent, and ratings after restructuring were weighted 34 percent. See section 3 for a justification of the ratings. Relevance Ratings Against Original PDO Against Revised PDO Rating Modest High Rating Value 2 4 Disbursement Weight 0.66 0.34 Weighted Value 1.32 1.36 Final Rating 2.68 Overall Rating Substantial Achievement of Objectives Against Original PDO Against Revised PDO Rating Substantial High Rating Value 3 4 Disbursement Weight 0.66 0.34 Weighted Value 1.98 1.36 Final Rating 3.34 Overall Rating Substantial Efficiency Against Original PDO Against Revised PDO Rating Substantial Substantial Rating Value 3 3 Disbursement Weight 0.66 0.34 Weighted Value 1.98 1.02 Final Rating 3 Overall Rating Substantial 34 Against Original PDO Against Revised PDO Relevance Modest High PDO Achievement Substantial High Efficiency Substantial Substantial Overall rating MS = 4 S=5 Disbursement Weight 66% 34% Final Rating 2.64 1.7 Final Score 4.34 Overall Rating Moderately Satisfactory 35 Annex 2. Project Costs and Financing (a) Project Cost by Component (in USD million equivalent) PAD Project Close Allocated Allocated Amount Amount Components/Activities (USD m) % (USD m) % % PAD A. Community and Livelihood Development 77.5 67.4% 96.1 83.5% 124.0% at Village Level A.1 Development and Strengthening 7.5 6.5% 23.3 20.3% 310.9% Community Organizations A.2 Financing of Village Development and 70.0 60.9% 72.8 63.3% 103.9% Risk Reduction Plans B. Institutional Development and Services at 19.5 17.0% 2.0 1.7% 10.2% Inter-Village and National Levels B.1 Development and Strengthening of Inter 15.5 13.5% 1.7 1.5% 10.9% village Organizations B.2 Business Promotion and Livelihoods and 1.8 1.6% 0.1 0.1% 5.6% Market Linkages B.3 Creating Conditions for Employment 2.2 1.9% 0.2 0.2% 8.6% Generation C. Capacity Development and Partnership 6.1 5.3% 2.4 2.1% 39.3% Building from Cluster to National Levels C.1 Capacity Building of SDF Staff and partner 3.2 2.8% 2.2 1.9% 69.7% org. C.2 Supporting Innovations 2.9 2.5% 0.2 0.1% 5.9% D. Project Management and Coordination 11.9 10.3% 14.5 12.6% 122.2% D.1 Overall coordination, mgt. and 9.4 8.2% 12.8 11.1% 136.2% implementation D.1.1 Project Management – National 2.9 2.5% 3.2 2.8% 112.3% D.1.2 Project Management – Regional 2.3 2.0% 4.0 3.5% 177.0% D.1.3 Project Management – Districts 4.3 3.7% 5.6 4.9% 130.5% D.2 Monitoring, Learning, communication and 2.5 2.2% 1.7 1.5% 69.6% evaluation Total Project Cost (USD m) 115.0 100.0% 115.0 100.0% (b) Project Financing (USD million) Appraisal Actual/Latest Estimate Estimate Percentage of Source of Funds (USD millions) (USD millions) Appraisal People’s Republic of Bangladesh29 0.00 0.00 International Development Association (IDA) 115.00 115.7230 101% 29 At appraisal, GoB had committed to contribute US$15 million towards an Endowment Fund for the sustainability of SDF. By the end of the project, they had contributed US$21.5 million. 30 Variation due to exchange rate fluctuation. Final amount includes US$30,929.97 in canceled funds 36 Appraisal Actual/Latest Estimate Estimate Percentage of Source of Funds (USD millions) (USD millions) Appraisal Subtotal 115.00 115.72 101% Local Communities 5.00 N/A31 N/A Total: 115.00 120.00 115.7232 31 Not measured 32 Reflects full Bank financing as planned. 37 Annex 3. Outputs by Component % Indicator Target Achievement Achieved Component A: Community and Livelihoods Development at the Village Level A.1: Development and Strengthening of Community Organizations Targeted vulnerable and marginalized beneficiary 80 100 125% population holding decision making positions 80 percent of GSs accessing and managing project funds as per agreed rules and service standards as set 80 98 123% out in the COM All of the 39,255 decision-making positions in the village organizations are held by members of the poor, HCP, and vulnerable (against a target of 80%) and 94% of these office holders are women. Of the 658,661 households benefitting by the project, 651,480 were mobilized into Nuton Jibon Groups (NJGs), which are clustered into 6957 Sanchay Sangrakham Committees (SSCs). According to MIS and internal assessments, 2987 village institutions are functioning in a transparent, inclusive, and accountable manner (against a target of 2753). Ninety-nine percent of GSs are accessing and managing project funds as per agreed rules and service standards (target 80%). A.2: Village Development and Risk Reduction Fund (VDRRF) 80 percent of JG members have mobilized savings and accessed additional sources of finance (e.g., 80 91 114% Shabalombi Fund; banks; MFI, etc.) At least 80 percent of completed community infrastructure subprojects are being operated and 80 89 111% maintained by communities Many infrastructure investments were developed with disaster risk management principles in mind—e.g., GS buildings that can serve as shelters during disasters, improved drainage and culverts to reduce waterlogging and flooding, and raised school yards to provide a safe space during disasters as well. Ultimately 2359 villages (72% of total villages) carried out participatory vulnerability analysis. Disaster volunteer teams were formed and trained in 2353 villages, and 2116 villages had purchased disaster management tools. 38 % Indicator Target Achievement Achieved Component B: Institutional Development and Livelihoods Promotion at the Inter-village Level B.1: Development and Strengthening of Inter-village Organizations 50 percent of villages have at least 5 Community Professionals (CPs) identified, trained and offering 70 47 67% services and 70% by EOP At least 50 percent of GSs and VCOs of current program (Phase 1) are federated at Upazila/cluster 50 93 186% level SDF identified and trained SDF is taking is to reach beyond the village and network these institutions at the cluster and district level through Nuton Jibon Societies (NJS). By the close of SIPP-2, 93 percent of Gram Samitis (GS) in phase one villages had joined a federated NJS. The financing model seems strong for a start—a corpus fund put into an interest-bearing account to cover sustaining costs, membership dues from member GSs, and fee-based services to members. B.2: Business Promotion, Livelihoods and Market linkages At least ten major business partnership established 10 10 100% through public/private partnerships with communities SDF forged or facilitated 6 national-level partnerships that benefitted over 15,000 youth with training and employment opportunities through companies/agencies―such as, Apex Footwear, Apex Pharma, Rangpur Foundry, G4 Security, and the Department of Youth Development. Another 10 partnerships/linkages with organizations/companies such as— Grameen Danon foods Ltd, BRAC Chilling Center, Bio-Science Agro, Milk Vita, and SSKS—SIPP-2 community organizations benefitted over 7000 project households with a variety of services as well as backward and forward linkages to markets. An additional 12 additional linkages between communities and government departments, companies and civil society helped over 89,000 project households access services such as--training, inputs, and access to commercial finance. B.3: Creating Conditions for Employment Generation At least 50,000 direct jobs created through project facilitation (and at least 75 percent of them are 50000 41849 84% sustained at EOP) 48,781 youth were employed through the formal labor market as well as self-employment (23,186 wage employment, 25,595 are in self-employment). An independent thematic study based on a small sample conducted 3 months before project close found that 99% were still employed by the end of the project. For wage employment, the top sectors for male youth were textile/garments (27%) and computer operating (27%), and security services (11%). Female wage earners found work in 39 % Indicator Target Achievement Achieved garments/textiles (45%) and tailoring shops (38%). The diversity of wage jobs for women were impacted by physical and cultural reasons. The top sectors young men who became self-employed were: computer composing (17%), tailoring (16%), electrician (12%), and mobile servicing (11%). For young women, self- employment was confined to tailoring (88%), and handicrafts (7%). Component C: Capacity Development and Partnership Building from Cluster to National Levels C.1: Strengthening the institutional capacity and delivery of services by SDF and other relevant partner agencies Staff performance appraisal results (following HR Policy) indicate 70 percent of staff performing at 70 89 127% Grades A and B, annually from year 2 At least 70 percent of the villages (Gram Samiti, SAC 70 86 123% and VCO) are rated as A or B grades According to MIS tracking and project assessments, 2987 village institutions are functioning in a transparent, inclusive, and accountable manner (against a target of 2753). Ninety-nine percent of GSs are accessing and managing project funds as per agreed rules and service standards (target 80%). C.2: Supporting Innovation Competitive window: At least 9 successful proposals 9 2 22% for Innovation Pilots awarded by EOP SDF financed 2 innovation pilots by the end of the project. Bangla-German Sampreeti (BGS) partnered with SDF under the development market place window in November 2013 to support project youth with skills development and linkage to the formal labor market. The activity provided skills training to 1,344 youth, of which 141 youth dropped out. The remaining 1,153 youth were facilitated to access both wage and self-employment, and 630 youth were formally employed in various sectors —such as, garment, security, food processing, mobile and hardware. For the 523 youth who became self-employed, the majority (430) were women who took up tailoring at their respective villages while the rest (93) were young men who set up mechanical, electrical, and mobile service shops. Component D: Project Management, Monitoring and Coordination GAAP recommendations implemented and reviewed annually with the community representatives in at 70 71 101% least 70 percent of project villages At least 70 percent of process monitoring recommendations reviewed and implemented/actions 70 83 119% taken by SDF management 40 % Indicator Target Achievement Achieved SDF has Complaints and Grievance Redressal system established and 80 percent of complaints received 80 89 111% have been addressed Grievances registered related to delivery of project N/A 100 N/A benefits addressed (%) Grievances related to delivery of project benefits that N/A 197 N/A are addressed-(number) 41 Annex 4. Economic and Financial Analysis Introduction 1. The SIPP-II Project became effective on 5 December 2010 with a total budgeted cost of SDR 76.2 million (USD 120 million, of which USD 115m was IDA sourced). The original closing date was 30 June 2016, but this was subsequently revised to 31 December 2015. 2. In addition to management, monitoring and coordination (initially USD 11.9 million, later increased to USD 17.8m), the other components were:  Component A: Community and Livelihood Development at Village Level (initially USD 82.5 million later increased to USD 91.5m) to empower and build institutional capacity of pro-poor village level and federation organizations as well as fund village development and livelihood-related investments.  Component B: Institutional Development and Livelihoods Promotion at the Inter- Village Level (initially USD19.5 million, later decreased to USD 2.15m) to promote the development and strengthening of inter-village level organizations and enterprise and business development.  Component C: Capacity Development and Partnership Building from Cluster to National Levels (initially USD 6.1 million, later decreased to USD 3.6m)to strengthen the capacity of the Social Development Foundation (SDF) and other relevant agencies in order to build the capacity to scale up SIPP’s community development program across much of Bangladesh. 3. Three restructurings occurred over the implementation period: (i) a Level 2 restructuring (approved November 2011) to address inconsistencies between the Financing Agreement and the PAD; (ii) a Level 1 restructuring (January 2014) which modified the PDO and several indicators, as well as funds allocation between components, and; (iii) a second Level 2 restructuring (December 2015) which brought forward the project closing date. None of the restructurings resulted in changes in the original components nor in the overall project amount. 4. The EFA conducted as part of the PAD projected benefits to some 1,500 villages and 300,000 poor and vulnerable households, deriving from: (i) investment in livelihood activities; (ii) community-led livelihood sustainability; (iii) community infrastructure; (iv) vulnerability reduction, and; (v) community-led management. 5. Economic and financial rates of return were provided for livelihood activities, as well as for the other categories defined above (table A4.1). However, no indications were provided in the PAD EFA or accompanying documents as to how these rates of return were estimated, although the conversion process from financial to economic returns was briefly described. 42 Table A4.1: Projected Rates of Return Calculated at Appraisal Financial Returns Economic Returns NPV (BDT NPV (BDT Activity IRR (%) billion) IRR (%) billion) Livelihoods 10.5 (0.4) 12.6 0.1 + CDD-led sustainability 16.1 1.3 17.0 1.2 + Vulnerability reduction 18.1 2.0 19.5 1.8 + Community Infrastructure 19.7 2.4 21.2 2.2 + CDD-led Management 23.8 3.4 25.3 2.9 Source: Annex 10, PAD, May 28 2010 2. Project Costs 6. The Project had fully disbursed all funds against final allocations in the 54-month implementation period. Table A4.2: Final Allocations by Component Allocated Amount Component (USD m) % of total % PAD A. Community and Livelihood Development at Village 96.1 83.5% 124.0% Level B. Institutional Development and Services at Inter-Village 2.0 1.7% 10.2% and National Levels C. Capacity Development and Partnership Building from 2.4 2.1% 39.3% Cluster to National Levels D. Project Management and Coordination 14.5 12.6% 122.2% Total Project Cost (USD m) 115.0 100.0% 3. Investment Completion Analysis 7. A total of 3,262 villages in three regions (Rangpur, Jamalpur, and Barisal) participated in SIPP-2 with the 651,480 Nuton Jibon group members– in both cases more than double the number initially envisaged. 8. This completion analysis has been prepared following the same format as in the PAD. Quantified returns are calculated primarily on the basis of Income Generating Activities (IGAs), primarily supported by Shabalombi loans advanced to poor and hard core poor (HCP) households within each community. Analysis has been undertaken on the basis of findings presented in the SDF Final Report on Assessment of Financial Returns of IGAs 33 arising from an analysis of project participants with investments grouped into six categories (livestock, agriculture, transport, small trade, fisheries and other). Of the almost 800,000 IGAs reported, 60 percent were for 33 September 2015, Social Development Foundation. 43 livestock (22 percent dairy, 22 percent beef and the remaining 16 percent small stock and poultry). Data from the Bank internal analysis of household expenditure data is also used.34 Table A4.3: Distribution of IGAs Income Generation Activities (IGAs) Activity Number Percentage Livestock 474,246 60.0 Agriculture 84,579 10.7 Transport 39,417 5.0 Small Business 150,229 19.0 Fisheries 27,633 3.5 Other 15,323 1.9 TOTAL 791,427 100 9. Agricultural investments focused primarily on vegetables, hybrid crops and betel leaf cultivation, while transportation included motorized rickshaws and vans. Small business investments were primarily in tailoring and retailing. Shabalombi-Funded Income Generating Activities 10. The majority of income generating activities (IGAs) were made possible by project financing in the form of community managed revolving Shabalombi loans. Project financing for this purpose35 accounted for BDT 5,805 million (USD 74 million), or approximately 64 percent of total project expenditures. Net incomes resulting from these investments are shown in table A4.4. It can be seen that returns to second cycle investments (those occurring after the first loans had been repaid) are on average higher than for the first cycle. 36 Transportation and small business investments generated the highest average net incomes. A flat interest rate of 10 percent per annum was charged for all loans under the Shabalombi scheme. Table A4.4: Average Net Incomes from Supported IGAs Average Net Income 1st Cycle 2nd Cycle IGA Average Lowest Highest Average Lowest Highest Cow rearing 22,557 1,000 62,500 30,489 2,800 118,600 Beef fattening 11,643 1,200 64,000 17,670 1,600 77,000 Fish culture 17,914 5,000 38,100 31,555 9,000 82,000 Poultry rearing 15,761 3,600 35,700 28,100 4,600 80,600 Koal rearing 15,000 15,000 15,000 20,000 20,000 20,000 Goat and sheep rearing 8,979 200 32,000 18,613 1,200 48,500 Vegetable Cultivation 30,280 3,000 38,800 49,720 16,100 84,000 34 Mghenyi and Kitsuki. Impact Evaluation for SIPP-II, Mimeo, World Bank. 2016 35 This category also included funds for the one-time grants discussed below. 36 It should be noted, however, that second cycle loans are typically greater than those under the first cycle. 44 Average Net Income 1st Cycle 2nd Cycle IGA Average Lowest Highest Average Lowest Highest Small shop 17,800 1,000 32,850 21,000 4,000 49,000 Tailoring 36,333 8,500 74,000 61,000 18,000 148,400 Duck rearing 11,988 1,900 32,000 15,714 5,000 33,000 Hybrid crop cultivation 10,850 2,700 18,050 18,850 3,500 20,500 Vermi compost 5,000 5,000 5,000 69,000 69,000 69,000 Nursery - - - 35,000 35,000 35,000 Transportation 40,231 8,800 77,800 47,173 4,900 78,500 Small business 39,852 800 124,800 61,792 2,200 268,000 Others 34,062 800 78,600 39,532 1,300 152,000 Source: SDF Final Report on Assessment of Financial Returns of Income Generating Activities . September, 2015. Annexure Table 18. 11. As many activities operate on a four-month or six-month cycle, all costs and income have been converted to an annual basis. The livestock IGAs yield annual net returns ranging from 19 percent for poultry operations to 33 percent for beef fattening and 48 percent for dairy from local cattle to 84 percent for dairy from crossbreed cattle. These returns to livestock production are supported by results from the World Bank impact analysis of baseline and completion survey data for SIPP-II, which show a 90 percent increase in the value of livestock production for the treatment group over the implementation period.37 12. It should be noted, however, that the numbers are considered reliable only for the cattle enterprises (local dairy, cross-breed dairy and beef fattening) due to the small sample size available for other activities. In preparing overall returns estimates, therefore, the average net income figures shown in table A4.4 are used for all activities except the cattle activities. On the basis of loan amounts advanced under the Shabalombi scheme, it is assumed in undertaking the analysis that 65 percent of loans are first cycle and the remaining 35 percent are second or third cycle. 13. Overall, beneficiary contributions to total Shabalombi-funded IGA investments averaged 36 percent, with Shabalombi loan funds providing the remaining 64 percent. Grants and Internal Lending 14. In order to reduce vulnerability among the poorest households in participating communities, two sets of measures were introduced under the project to increase the welfare and income of these groups. Firstly, a one-time grant to a maximum amount of BDT 5,000 (USD 64) per household was extended to support households headed by widows and divorced women, the elderly and the handicapped. These grants reached 35,321 households or 96 percent of those households classified as vulnerable. A survey conducted by SDF consultants38 indicated that 95 37 Mghenyi and Kitsuki (2016): Impact Evaluation for SIPP-II, Mimeo. World Bank. 38 Impact of ‘Nuton Jibon’ on the lives of vulnerable communities. Social Development Foundation. Undated. 45 percent of these grants were used for IGAs, with the remainder being used for medicine and similar needs. Given the limited amount of the grant, the purchase of large stock was generally infeasible and over 50 percent of recipients purchased goats, with another 26 percent establishing poultry operations. 15. Secondly, all households—with an emphasis on those classified as vulnerable—were encouraged to join internal savings groups in their communities. Of the 78 percent of vulnerable households which subsequently joined the Nuton Jibon groups, an estimated 85 percent commenced regular savings typically contributing BDT 10 (USD 0.13) per week, while 95 percent of all households that were members of the Nuton Jibons joined savings groups. Savings from these groups were then available to group participants for loans. A total of BDT 488.6 million (USD 6.2 million) had been saved through these groups by the end of the implementation period (see Appendix 4). However, it was found that the small size of internal savings loans - averaging only BDT 2,126 (USD 27.2) - resulted in a low rate of uptake. Instead the Shabalombi loans, averaging over BDT 12,000 by 2015, were strongly preferred as they permitted the type of investments typically sought by participants (particularly large livestock). Shortly before the completion of implementation, steps were taken to consolidate savings across a number of communities, thus allowing larger individual loans, but the results of these measures are not yet apparent. Community-Led Livelihood Sustainability Benefits 16. The creation and/or strengthening of community-level capacity to support the economic well-being of particularly the poorer households within each community was a central element of SIPP-2, as the long term sustainability of project benefits would depend heavily upon these institutions. The Gram Samiti (the community level management committee) and the Gram Parishad (responsible for management of credit activities such as setting policy for the functioning the program in the village and approving development activities) are the two most important organizations for the purposes of this analysis. 17. The PAD projected the establishment of 1,500 Gram Parishads by the completion of implementation, achieving an average loan repayment rate of 95 percent within 30 days of due date. At completion, the project had encompassed 3,262 communities that attained average 30- day repayment rates of between 94 to 96 percent, according to the region. 18. In addition, the PAD projected that a minimum of 75 percent of beneficiaries (increased from a SIPP-I target of 63 percent) would be from poor and vulnerable households in each community. This target was clearly exceeded, with survey data showing that not only were 94 percent of direct beneficiaries women from vulnerable households, but also that the same proportion held decision-making roles within the village institutions. Community Infrastructure Benefits 19. A range of small-scale village infrastructure was anticipated to be funded and completed under the SIPP-II project. These included dirt roads, culverts and bridges, as well as tube wells, Gram Samiti offices and work on village schools. 46 20. As of the completion of implementation, 100 percent of the 6,300 planned sub-projects had been completed. Community infrastructure projects included: - 1,937 GS office buildings; - 954 earthen roads comprising 615 km completed and maintained; - 2,541 culverts; - 4,199 tubewells; - 71school (most of which could be used as shelters during natural disasters); - 5,993 m of drains; and - 330 other works. 21. An external cost efficiency study39 found that – on the basis of works completed and costs incurred - community management saved about 30% of the standard costs, while more than 90 percent of the SIPP villages had established operation and maintenance plan and were using community funds for maintenance of these roads. 22. A technical audit conducted for SDF in late 2015 compared infrastructure construction costs against those used by the Local Government Engineering Division (LGED) and by the Department of Public Health Engineering (DPHE). The results are shown in Table A4.5 below. 23. It should be noted that the only category for which SDF costs are higher than other agencies is for latrines. This is reported to be the case because the SDF financed latrines were permanent and equipped with water supply as opposed to temporary structures. 24. The same study reported that, overall, 82% of the infrastructures audited conformed to SDF’s design and standards. Table A4.5: Comparison of SDF Costs against other Agencies SDF Rate LGED Rate DPHE Rate Item Unit Min Max Min Max Min Max Earthworks Tk/cubm 68 108 85 221 87 93 RCC work Tk/cubm 5,694 5,803 6,381 12,118 7,665 8,548 Brick work Tk/cubm 3,912 4,666 4,614 5,368 5,096 GI pipe Tk/m 394 501 586 300 400 PVC pipe Tk/m 66 96 159 698 171 1,380 Palasiding Tk/m 2,824 3,080 2,793 4,638 n/a n/a Tubewell Tk/no. 8,356 62,541 n/a n/a 7,600 55,000 Latrine Tk/no. 81,740 87,037 10,250 30,000 Source: Dr Asif M. Zaman. Technical Audit and Impact Assessment of SIPP-II. SDF. October 2015. 25. No surveys or other measures were undertaken prior to or during implementation to quantitatively assess the financial and economic impact of community infrastructure investments. However, a 2010 EFA for an Asian Development Bank financed Rural Infrastructure Project covering two Divisions and completed late in the previous year estimated an EIRR of 24.3 percent for activities in Barisal Division, one of the three regions encompassed by SIPP-II. Higher rates of return (43.6 percent) were calculated for Khulna Division, but this Division does not form part of the current project. 39 Dr Akther Husein, Cost Effectiveness of SIPP-II. Social Development Foundation. January 2016. 47 Vulnerability Reduction Benefits 26. The vulnerability of poorer households to natural disasters (e.g. cyclones, droughts and floods), and their frequency within the project target area, was estimated in the PAD to lead to 42 percent of intended beneficiaries requiring livelihood assistance. Without such assistance, the PAD analysis estimated that 40 percent of HCP households would not be able to cope with natural shocks, leading to a continuation of their extremely low levels of income and food insecurity. 27. As presented in the Grants and Internal Lending section above, 96 percent of households classified as vulnerable were initially assisted with one-time grants under the project and 95 percent of these grants were used for IGAs, increasing household incomes and assets. Over 50 percent of vulnerable households surveyed had no assets prior to participation in the project. However, after four years it was found that individuals in this category had accumulated assets worth on average 27,085 BDT.40 28. The same study shows that 60% of vulnerable people surveyed had started small ventures on the basis of loans from the Shabalombi fund and their monthly income from this source averaged BDT 3,662. This would represent an increase of more than 100 percent in the average of vulnerable, poor and hard core poor income determined in the baseline survey. Independent survey data collected for SDF shows the broader situation for vulnerable groups (table A4.6), with increases in average incomes of from 46 to 96 percent over the implementation period. Table A4.6: Increase in Annual Incomes among Vulnerable and Poor Project Beneficiaries (BDT) Baseline Survey Incomes Implementation Completion Survey Vulnerable Vulnerable Income Income Income Division HH HCP Poor HH Increase HCP Increase Poor Increase Jamalpur 18,000 32,707 47,228 26,232 46% 58,873 80% 88,584 88% Barisal 21,000 25,812 40,117 32,400 54% 50,716 96% 78,744 96% Rangpur 15,600 22,060 34,000 23,000 47% 40,500 84% 65,250 92% Source: Technical Audit and Impact Assessment of SIPP-II. SDF. October 2015. 29. It is also worth noting that although an analysis of baseline, control and treatment impact analysis, 41 found no significant change in overall food consumption, it did identify major increases in milk and dairy product consumption (up 39 percent) and meat and egg consumption (up 49 percent), suggestive of important improvement in household nutrition as a result of the project. Community-Led Management Benefits 30. The EFA analysis in the PAD identified three types of community-led management benefits: (i) better community level governance; (ii) efficient delivery and use of funds, and; (iii) efficient targeting of beneficiaries. 40 Impact of ‘Nuton Jibon’ on Vulnerable Communities. SDF. Undated. 41 Mghenyi and Kitsuki (2016): Impact Evaluation for SIPP-II, Mimeo. World Bank 48 31. With respect to community level governance, an independent study conducted by the Bangladesh Disaster Preparedness Centre42 found significant shortfalls in both documentation and disclosure with respect to loan operations, although these were much less common for the key Shabalombi loans, for which 88 to 95 percent of documentation was complete, than for the internal and youth loans, where only 15 to 40 percent of required documentation could be located. In addition, many Nuton Jibon members lacked knowledge of many basic aspects such as the amount of their personal savings under the project or the decision process followed in approving loan applications. Key documents relating to financial transactions (e.g. bank passbooks, bank statements and deposit slips) were also absent in many cases (table A4.7). Table A4.7: Availability of Required Management Documents in Surveyed Communities % of villages where specific document was found Document Gaibanda Jamalpur type Internal Shabalombi Youth Internal Shabalombi Youth Disbursement 61 97 89 10 98 48 register Cheque 11 61 45 80 register Cheque book 47 71 55 72 Collection 53 95 53 98 register Deposit slip 68 53 Bank 42 42 53 60 Passbook Cash Book 100 74 97 82 Service charge 3 10 register Loan security 26 25 register Bank 84 61 18 40 statement Monthly 0 13 0 15 financial report Source: See footnote 32. The delivery and use of funds for project participants required the expenditure of approximately 12 percent of total investment (based on the cost of project management and coordination), while the PAD claims that 30 to 40 percent ‘leakage’ is average and that the Bangladesh government spent BDT 6.55 to transfer BDT 1 to recipients under their Rural Rationing Programme. 33. Although the PAD does not provide specific figures for targeting, it states that all members of Jibikayan Groups (JGs) under SIPP-I were poor, hard core poor or youth, while 99 percent of loans disbursed by communities under the project were to poor or hard core poor 42 Ensured Good Governance at the Community Level: A Challenge in SIPP-II. Bangladesh Disaster Preparedness Centre. June 2015. 49 members. Monitoring data for SIPP-2 shows that 94 percent of direct beneficiaries were women from poor or hard core poor households. Financial and Economic Analysis 34. Returns to investment have been calculated based upon the returns to income generating activities alone. These represent activities financed by one time grants, internal lending from savings groups and Shabolombi Fund loans, skill development loans, and employment loans to youths. It should be noted, however, that as no financial analysis was undertaken during implementation of returns to the first two categories, these categories have been assumed to generate the same benefits as the Shabolombi financed investments, which may well result in an overestimate of total returns, as the most profitable investments occur for large livestock (dairy and beef – see Appendix 1) which are generally not feasible for borrowers under the other funding sources, due to the much smaller size of the loans. Nevertheless, the amount involved in the one-time grants and internal savings loans is considerably smaller than for the Shabolombi Fund, so any difference will be minor. 35. Using the data presented in table A4.4 above, and assuming a ratio of 65 percent first cycle loans and 35 percent second or third cycle loans, the total net benefit for all IGAs can be calculated. It is then possible to calculate the contribution of the beneficiaries, as is shown in Appendix 3. 36. Loan disbursement data for the various funds permit the estimation of the year in which such investments are made, while benefits are assumed to be accrued in the year following the investment.43 It should be noted that, in accordance with guidance contained in the recent World Bank technical note Discounting Costs and Benefits in the Economic Analysis of World Bank Projects, a discount rate of 6 percent has been applied in determining Net Present Values for the investments. 37. As loans to beneficiaries based upon both internal savings and the Shabalombi fund are revolving, benefits will continue to accrue after the completion of the implementation period, although it is assumed that the amount available for lending will decline by 10 percent per annum due to losses on the loan portfolio (see Appendices 4 and 5). 38. Financial Returns (both IRR and NPV) calculated are presented in Appendix 6 and summarized in Table 8 below. At the financial level, these returns are very strong, in large part due to the assumptions of returns generated after the completion of the implementation period. In order to show this clearly, FIRRs and FNPVs are given both for the implementation period (to 2015) and are estimated at 37.7 percent and BDT 2,275 million, as well as until 2030, when the community level funds are assumed to be virtually exhausted. This latter duration yields a FIRR of 52.6 percent and an FNPV of BDT 9,058 million. Table A4.8: Financial Rates of Return to Investments Indicator To end 2015 To end 2030 43 Net returns to individual investments are estimated on the basis of a 12-month cycle. 50 FIRR (%) 37.7% 52.6% FNPV (BDT million) 2,275.0 13,850. 39. These rates of return compare strongly with those in the PAD, which were projected to reach just under 24 percent. The increased ex-post rate of returns to the project over that projected in the PAD can largely be attributed to the more than doubling of beneficiary populations over those foreseen during preparation and the re-lending of Shabolombi funds. Thus benefits accrued to a much greater number of participants without any increase in initially budgeted costs. It is also likely that any inclusion of labor costs for enterprise operations - that is, where labor costs are not charged to the enterprise, but are seen as part of the overall net income figure – will increase the net returns calculated. In fact, the World Bank Impact Analysis already cited found an increase of 13.7 percent in day labor for project beneficiaries, suggesting that participation does result in increased labor use. 40. These two considerations suggest that although the true FIRR may well be lower, the investment would still have generated a worthwhile return. 41. Economic Returns are adjusted from financial returns largely on the basis of the divergence of the Nominal Effective Exchange Rate (NEER) and the Real Effective Exchange Rate (REER) with respect to the Bangladesh Taka. The recent (late 2015) Article IV consultations between the IMF and the Government of Bangladesh44 indicate an estimated NEER of BDT 95.4 per US dollar, while the REER is calculated at 120.5, on the basis of significantly higher inflation rates in Bangladesh than is the case in its main trading partners. The REER thus exceed the NEER by 26.3 percent, and the cost of project funds (although not those contributed by the beneficiaries) has been increased to reflect this divergence. As few traded resources are used in the small scale investments undertaken by project beneficiaries, no specific adjustments for internationally traded goods are required. 42. A further factor usually considered in undertaking adjustments for economic, rather than financial, costs is the real cost of labor. No recent shadow labor rates could be identified for Bangladesh, but the country enjoys a relatively low unemployment rate (4.3 percent in 2014 according to World Bank data). However, national underemployment rates have been estimated as high as 40 percent, 45 suggesting that financial prices for labor are probably overstated. Nevertheless, no data was collected during the implementation with respect to the labor component of investments, suggesting that these costs were subsumed within the net returns. As a result, no adjustments have been made to labor costs for the economic analysis. 43. With the higher value of project funds (provided in SDRs), the EIRR and ENPV for the project are lower than those estimated on a financial basis, as summarized below. Table A4.9: Economic Rates of Return to Investments 44 Bangladesh. Staff Report for the 2015 Article IV Consultations . IMF. January 5, 2016. 45 http://www.indexmundi.com/bangladesh/unemployment_rate.html 51 Indicator To end 2015 To end 2030 EIRR (%) 33.0% 49.4% ENPV (BDT million) 1,907.2 8,690.9 44. Economic rates of return projected in the PAD were 25.3 percent (see Table A4.1). Poverty Alleviation Benefits 45. According to the 2010 Household Income and Expenditure Survey 46 , the most recent source available, the poverty line in rural Bangladesh was set between BDT 13,004 (lower line) and BDT 14,532 (upper line) per annum. With an average rural family size (HIES 2010 data) of 4.53, this would equate to BDT 62,369 per family. By contrast, the baseline survey (see Table A4.6) indicated an average annual income among project participants of BDT 18,200 for vulnerable families, BDT 26,860 for hard core poor families and BDT 40,448 for poor families; considerably below this level. 46. However, by completion of implementation, these levels had increased to BDT 27,210, BDT 50,029 and BDT 77,526 respectively, representing income increases in the range of 50 percent for the most vulnerable households and 90 percent for hard core poor and poor households. Although poverty income levels may have increased since 2010, the data would suggest that the average poor beneficiary family is no longer below the poverty line, while the average hard core poor family is now close to the poverty line. Only the most vulnerable households (typically headed by widows, the elderly or the handicapped) still remain deep below the poverty line, despite the significant increase in household income over the implementation period. Employment Generation Benefits 47. In addition to employment created among the more than 600,000 recipients of one time grants and both internal and Shabalombi loans, many of whom would have been unemployed or underemployed prior to SIPP-2, the project also addressed youth unemployment through skills development training and support for placement of graduates with employers, or access to loans for small enterprise establishment. 48. Over the course of implementation, 25,595 youth established enterprises and a further 23,186 youths obtained employment, with garment factories (for wages) or computer operation (self-employed) being the most common roles for women, and mechanic, driving and security services (for wages) or carpenter or furniture making (self-employed) for men.47 46 Report of the Household Income and Expenditure Survey 2010 . Bangladesh Bureau of Statistics. Ministry of Planning. 47 Survey/Impact Assessment on Youths Employed in Different Industries/Places through the SIPP Project . SDF, October 31, 2015. 52 Table A4.10: Impact of Youth Employment Training on Participants Self-Employed Wage Employed Immediately Immediately Before After Before After Employment status Training Training Present Training Training Present Males Unemployed 80 28 1 98 17 0 Self-employed 8 48 99 1 1 0 Wage employed 12 24 0 1 82 100 Females Unemployed 95 17 2 99 23 0 Self-employed 0 75 98 0 3 1 Wage employed 5 8 0 1 74 99 Source: Survey/Impact Assessment on Youths Employed in Different Industries/Places through SIPP Project . SDF. October 31, 2015. 49. As shown in Table A4.10, survey data indicates that the impact of skills training was significant, with more than 90 percent of participants being unemployed prior to skills training and placement. Even after a period of time that extended over several years in the case of the earliest participants, unemployment was reduced to 0-2 percent. These findings are confirmed by the World Bank SIPP-2 Impact Evaluation which identified a 44 percent increase in factory employment among project participants, although interestingly, it also found a decline in shop keeping and no significant change in salaried employment. 50. The SDF survey also identified substantial increases in incomes for both wage earners (35 percent for men, 20 percent for women) and for self-employed participants (45 percent and 43 percent, respectively). Average incomes were reported of over BDT 9,000 for those in wage employment and BDT 11,560 for self-employed men. Self-employed women earned considerably less – BDT 5,750 – but this may be a result of only working part-time. Significant increases were also found in asset accumulation among the respondents, including household possessions, livestock and land. 53 Annex 5. Bank Lending and Implementation Support/Supervision Processes (a) Task Team members Responsibility/ Names Title Unit Specialty Lending Meena Munshi Senior Rural Development Specialist GFADR TTL Ousmane Seck Senior Rural Development Specialist GFADR Natasha Hayward Senior Rural Development Specialist GFADR Co-TTL C.S. Renjit Consultant Inst. Dev. Pushina Kunda Ng’Andwe Senior Rural Development Specialist GFADR Toufiq Ahmed Procurement Specialist SARPS Henry Bagazonzya Sr. Financial Sector Specialist SASFP Suraiya Zannath Sr. Financial Management Specialist SARFM Luis O. Coirolo CDD Institutions Consultant SASDA Vinaya Bhargava Governance Consultant EASER Shakil Ahmed Ferdausi Environmental Specialist SASDI Farzana Morshed Operation Analyst SACBD Nadia Sharmin Environment Consultant SASDS Winston Dawes Microfinance Specialist SASDA Md. Aminul Islam Khandaker Rural Development Consultant SASDA C.S. Renjit Community Institutions Consultant SASDA Abdur Rahman Khan Technical/Engineering Consultant SASDA M. Jahangir Alam Chowdhury Impact Evaluation Consultant SASDA Chau-Ching Shen Senior Finance Officer CTRFC Mei Wang Sr. Legal Counsel LEGES Brenda Scott Information Assistant SASDO Venkatakrishnan Ramachandran Program Assistant SASDO Amani Haque Program Assistant SACBD Supervision/ICR Ousmane Seck Senior Rural Development Specialist GFADR TTL Pushina Kunda Ng’Andwe Senior Rural Development Specialist GFADR Co-TTL Frauke Jungbluth Senior Rural Development Specialist GFADR TTL Seenithamby Manoharan Senior Rural Development Specialist GFADR TTL Sugata Talukdar Operations Analyst GFADR Overall support Josef Ernstberger Consultant (FAO) Program impl. C.S. Renjit Consultant Inst. Dev. Winston Dawes Senior Rural Development Specialist GFADR Finance Farhat Chowdhury Environmental Safeguards Specialist Environment Sabah Moyeen Senior Social Development Specialist GSU06 Social safeguards Mohammed Atikuzzaman Financial Management Specialist GGO24 Fiduciary Marghoob Bin Hussein Senior Procurement Specialist GGO06 Fiduciary Ishtiak Siddique Senior Procurement Specialist GGO06 Fiduciary Jaynab Akhtab Consultant - Social Social safeguards Moutushi Islam Consultant SASDA Tahsina Akbar Team Assistant Amani Haque Program Assistant 54 (b) Staff Time and Cost Source Of Funding Labor Travel Other Total FY10 Bank Administrative Funds 312,041.67 140,631.78 40,537.16 493,210.61 BBFAO 119,660.00 119,660.00 Grand Total 312,041.67 140,631.78 160,197.16 612,870.61 FY11 Bank Administrative Funds 198,844.96 53,653.78 24,671.65 277,170.39 BBFAO 67,310.00 67,310.00 Grand Total 198,844.96 53,653.78 91,981.65 344,480.39 FY12 Bank Administrative Funds 195,271.07 49,419.81 12,266.56 256,957.44 BBFAO 74,930.00 74,930.00 Grand Total 195,271.07 49,419.81 87,196.56 331,887.44 FY13 Bank Administrative Funds 142,256.11 58,840.13 12,488.36 213,584.60 BBFAO 78,650.00 78,650.00 Grand Total 142,256.11 58,840.13 91,138.36 292,234.60 FY14 Bank Administrative Funds 213,507.03 43,565.69 7,694.79 264,767.51 BBFAO 30,360.00 30,360.00 Grand Total 213,507.03 43,565.69 38,054.79 295,127.51 FY15 Bank Administrative Funds 87,640.78 11,849.43 2.93 99,493.14 Grand Total 87,640.78 11,849.43 2.93 99,493.14 FY16 Bank Administrative Funds 96,615.37 29,172.79 14,015.41 139,803.57 BBFAO 34,485.00 34,485.00 Grand Total 96,615.37 29,172.79 48,500.41 174,288.57 55 Annex 6. Summary of Borrower's ICR and/or Comments on Draft ICR 1. Social Development Foundation (SDF) has undertaken a Five-year Empowerment and Livelihood Improvement Nuton Jibon Project, popularly known as SIPP II (2010-2016) which is a sequel of Social Investment Project (SIPP-I) using community-driven development (CDD) approach. Due to acceleration of disbursement of funds to the beneficiaries and achievement of the targets earlier than the scheduled time the Government of Bangladesh (GoB) and World Bank has closed SIPP-II project ahead of official closing date i.e. December 31, 2015 instead of June 30, 2016. The BICCR report is prepared covering all aspects of the project including assessment of project outcomes, outputs as well as to the extent to which the project has been able to deliver expectations and the improvement of livelihood status of beneficiaries from the project given the investments made. The objectives for the study is documenting and providing evidence of the extent to which Nuton Jibon Project (SIPP-II) achieved its objective under the original Project Development Objective (PDO) and achievements under the revised Project development Objective and intermediate outcomes by end of the project and operational lessons of the SIPP-II that can be applied to other Livelihood programs and improve the implementation of the recently approved Nuton Jibon Livelihood Improvement Project (NJLIP). The study has been coordinated by secondary data from several reports of SIPP-II and field data to triangulate the assessment from secondary data. The project serves a total of 658,661 direct beneficiaries whilst the targeted number of direct and indirect beneficiaries stands at 6.0 million. 2. The project was implemented to achieve PDO through employing a CDD strategy to (i) empower the poor and strengthen local governance by developing sustainable, participatory and accountable rural community institutions; (ii) reduce vulnerability of the poor to risks, in particular those associated with natural hazards and climate variability; (iii) increase employment opportunities by enhancing skills, supporting the start up or expansion of income generating activities and strengthening access to markets and financial institutions; and (iv) support small - scale demand - driven community investment sub-projects that are prioritized, implemented and managed directly by the rural poor. 3. Under the four components of the project, the interventions of Nuton Jibon SIPP-II include initial information and communication dissemination, building up rural socio-economic institutions for poor and hardcore poor like NJG, SSC, VCO, GS etc. and One Time Grant Recipient Vulnerable falling under HCP group. The components also focus on preparing village development plan and providing fund directly to the community for embarking on different income generation activities in view of their skills, suitability etc. SDF facilitates the poor community to establish linkages with various service providers, preparation of community action plans based on need assessment. Different community infrastructure sub-projects, income generating activities, skill development training, and employment generation of the unemployed youths being implemented under the component and sub-components. All these interventions are mainly targeted to the improvement in the status of livelihood of the poor and hard core poor people 56 4. Under the Component A: Community and Livelihood development at village level, all the target intermediate results indicators were fully achieved. Targeted Female beneficiaries, targeted beneficiary (vulnerable and marginalized population)/people participating in community based decision making and management strategy, total targeted vulnerable and marginalized beneficiary population holding decision making positions has been achieved beyond the project target. Out of 39,225 total key positions, 94% (36,900) women occupy in decision making positions in different village institutions e.g. GS, VCO, SAC etc. and 91% target HHs has been mobilized into NJG as of December 2015 under SIPP-II. The village development and risk reduction fund consists of a budget envelop for directly financing to the village institutions and to build the capacities of the poor and empower the communities to manage their funds by themselves as per Community Operational Manual (COM). The total resource allocation for an individual village is calculated on the basis of Taka 15,000 per household. Total of 6,328 community infrastructure sub-projects (including GS office building) have been completed out of which 1,937 GS office buildings being maintained regularly. The remaining 4,391sub-projects are operated and maintained by the community annually. The OTR of Shabolombi Fund (Livelihood fund) is 94% which is close to the internationally accepted standard of 95%. Gram Samities are properly addressing risk areas during conducting Participatory Vulnerability Analysis (PVA) and preparing social maps indicating the risks. Total 2,388 villages completed PVA and 17,628 disaster volunteers received training on disaster preparedness, rescue, rehabilitation and early warning system those villages. Total of 2,278 villages implemented risk reduction plan by the active participation of community people through analysis the risks and vulnerability of individual villages by the leadership of Disaster Volunteer Team and Gram Samiti since inception. 5. Component B: Institutional Development and Livelihood Promotion at the Inter- Village Level was concerned with the promotion of the development and strengthening of inter- village level organizations and enterprise and business development for higher level/value livelihood activities. In order to achieve these objectives, a substantial progress has been made in livelihood sectors during the project period. About 85% NJG members have received Shabolombi loan for undertaking different income generating activities. About 45% NJG members have completed 2nd cycle IGA and able to come out from poverty level. Supporting the most vulnerable to gain membership in VCOs, a total of 35,422 vulnerable received one-time grant from project out of which 17,611 from SIPP-II. As of now, 33,966 (96%) most vulnerable graduated and are accessing to Shabolombi loans from VCOs and started IGAs. A total of 16,082 (99%) trained CPs have been identified from 3,012 villages out of which there are 1,403 villages having 5 CPs are offering services. Total 8 partnerships and 10 effective linkages have been established between service providers/ market operators and producers organization for ensuring qualitative inputs, technical services and products marketing at higher price. Capacity of producers has increased in terms of production, marketing, value addition in the products through producer group activities. As a result, the communities started to operate larger IGAs from small scale IGAs investing larger amount of loan. It has also been observed that the investment in dairy sector has significantly increased and investment in homestead goat and poultry rearing has decreased. Communities are embarking on comparatively large scale IGA like beef fattening, dairy, commercial vegetable cultivation, high value crop and grocery shop with higher amount of loan rather than small scale IGA like rickshaw/van, small business, homestead goat rearing and poultry. Net income of the poor and hardcore poor has increased after completing successfully 57 2nd and 3rd cycle of IGA. 50% project HH people have been able to increase their income by 40%-50%. The rate of wage employment in Barisal region is higher (58%) than the self- employment. On the other hand rate of self-employment in Jamalpur (60%) and Rangpur (54%) regions are higher than wage employment. 6. The second generation institution NJCS was formed to support and strengthen Gram Parishads for maintaining their organizational quality, ensuring accountability and compliance to the rules of business and guidelines for sustainable performance, scaling up information sharing and communication among all its members, aggregating various services and mobilizing producer groups and livelihood development activities including linkages with local government, private sector and financial institutions. 7. The target intermediate results indicators under the Component C: Capacity Development and Partnership Building from Cluster to National Level were also substantially achieved. Under the pilot project titled “Marketplace Development for Youth Employment Generation (MDYEG), a total of 1,344 youths passed out by successful completion of their course duration. A total of 1,344 youths received skill development training. Under the project so far 630 youths have been provided wages employment in companies and industries through the project initiative. 523 youths could be self-employed in their own areas or nearby towns. BGS has helped them to secure loans from banks, NGOs and development partners. During the process of implementation, some challenges and constrains were faced such as home sickness, shorter project life-span, non-provision of wage loss etc. In order to automate, ensure smooth and secured tracking of community finance activities (loan management) of project villages, SDF has already initiated piloting computerization of community finance activities for four months (Nov’15-Feb’16). As a part this piloting, SDF has already developed a user -friendly and interactive web-based Loan Management System (LMS) with the help of our software development consultant. The user interface i.e. labeling of data entry form and some sorts of reports are in Bangla. A total of 3,171 villages have been graded as per Phasing out plan and 2,703 (85%) are graded A & B. 8. Component D: Project Management Monitoring and Co-ordination aimed to oversee, coordinate, plan and implement the project at national, regional, district and field level and track project implementation and service delivery with a view to respond to operational aspects across regions, districts and villages. Since January 2013, staff performance appraisal system has been introduced and being conducted on a yearly basis. This is an ongoing process, and staff appraisal is used as the basis for awarding contracts, annual increments and termination. Major achievement is completion of substantial staff recruitment, institutionalization of GAAP at the regional and district and AMT Office levels. So far, a total of 1,217 staffs have been reallocated/deployed under NJLIP. SDF follows a “Zero� tolerance policy for financial misappropriation. The rules of HR Policy and Manual are strictly enforced among the staff at all levels. Process monitoring recommendations have been reviewed by SDF on a monthly/quarterly basis. The outputs of field observations are being shared with district and regional officials for comments, decisions and necessary actions. Each of the actions based on management decision related to PMA findings are followed up by the M&E staff of SDF as well as PMA staff. A total of 448 issues identified by PMAs in three regions (Northern & Southern zones) and 372 (83%) issues have been resolved as of June’15. Under the SDF -MIS, Datasoft System Ltd. developed 5 58 (five) modules which are Financial Information System (FIS), Project Monitoring System (PMS), Procurement System, HR and Payroll System; and Inventory System. All modules are being used under SIPP-II satisfactorily. 9. The revised PDO of the project has been sustainably achieved. The first PDO indicator relates to village institutions supported by the project. At the end of the project 92% (2,987) villages those have access to VDRRF are functioning in a transparent, inclusive and accountable manner, as per project guidelines where the target was to bring at least 84% (2,753) villages under this PDO. The second indicator of the PDO is the direct project HHs beneficiaries and 658,661 targeted beneficiaries have been mobilized into NJG which is101% of project target 652,000. The next PDO indicator counts direct beneficiaries women as a share of direct project beneficiaries. Total 617,165 women (94%) have been provided as the project beneficiary against the project target 586,800 (90%) women. 62% targeted HHs have increased in income by at least 50% against base year for 50% of targeted HHs which shows the successful achievement of fourth PDO indicator. Among targeted 80% HH, 92% HH has been benefited from improved community infrastructure or social services sub-projects completing the final PDO indicator. Infrastructure works have been completed in 2,987 villages where the direct benefited HHs from completed infrastructure sub-projects are 597,400 which is 92% of total direct beneficiaries 10. The cumulative Fund disbursement as of December 31, 2015 is BDT 9,107.52 million excluding community contribution which is 100% against the total project allocation of BDT 9,107.52 million (US$ 115.0 million). 11. A total of 124 cluster community society under Jamalpur, Gaibandha, Nilphamari, Barguna, Pirojpur, Bagerhat, Sirajgonj, Sherpur, Mymensingh, Barisal, Patuakhali, Rangpur, Sylhet, Naogaon, Kurigram and Dinajpur districts (16 districts) have been formed and started functioning as per COM 15. All the societies both at cluster (124 offices) and district (15 offices) level have established their own offices in a rental premises. The inclusion of villages as member of NJCS is going on in districts under the regions as planned. A total of 2,877 (Jamalpur- 938, Rangpur- 1,231 & Barisal- 708) villages under 16 districts of three regions have mobilized as member of concern NJCS. More than 88% of villages under SIPP II in all districts enrolled as members of the NJCS. An amount of BDT 28.94 million have been deposited as membership fees by member villages of 16 districts under Rangpur, Jamalpur and Barisal regions meaning 90% member villages have already been deposited their membership fees. To implement the phasing- out activities and make the NJCS sustainable a 2-days orientation on NJCS activities have been conducted at SDF-HQ with the participation of DOs of phased out districts, RDs of all 3 regions. Also The HQ team facilitated the regional team to carry out the capacity building tasks for CFs of phasing out districts. Accordingly, all region teams with the assistance of District teams provided training on role and responsibilities of CFs in phased out districts. 12. The overall conclusions that the Empowerment and Livelihood Improvement Nuton Jibon Project (SIPP-II) was very successful and has been rated as “satisfactory�. The project serves a total of 658,661 direct beneficiaries whilst the targeted number of direct and indirect beneficiaries stands at 6.0 million. The project was implemented through a strong CDD approach involving much exchange of ideas and learning. Key lessons learned have been compiled and are summarized in this report. These concern the following- 59 1. Implementation of project needs strong management with close access to key decision makers. 2. Strong, competent and balanced project teams should be established at an early stage. 3. Develop public-private partnership arrangements whereby specialized private sector service providers and NGOs can be used to complement the work of government agencies in the implementation of project. 4. Youth employment opportunities can be considered working with the villagers for independent work and other employment opportunities around their locality. 5. Participatory Identification of Poor (PIP) can be redesigned every year, so that the group individuals who stay out of the scope of PIP, are later incorporated into the following rundown of beneficiaries. 6. Capacity building preparing should be a consistent procedure for the further improvement of diverse IGAs in the objective ranges. 7. Tracking market pattern at general provisional and renovating the capability preparing educational modules is required. 60 Annex 7. List of Supporting Documents [DTCPL] Development Technical Consultants Pvt. Limited. 2013. “Mid-Term Review of the Empowerment & Livelihood Improvement “Nuton Jibon� Project�. Mimeo Hussein, Akhter. 2016. Borrower’s Completion Results Report. Mimeo. Mghenyi and Kitsuki. 2016. “Impact Evaluation for SIPP -II.� Data report from the impact evaluation of the Empowerment and Livelihood Improvement "Nuton Jibon" Project . Washington, DC: World Bank. Mimeo. Iqbal, S. M. Hasan. 2015. Survey/Impact Assessment on Youths Employed in Different Industries/Places through SIPP Project. SDF. October 31, 2015. Mghenyi and Kitsuki. 2016. “Impact Evaluation for SIPP -II.� Data report from the impact evaluation of the Empowerment and Livelihood Improvement "Nuton Jibon" Project. Washington, DC: World Bank. Mimeo. World Bank. 2010. Project Appraisal Document on a Proposed Credit in the Amount of SDR 76.2 Million (US$115 Million Equivalent) to the People’s Republic of Bangladesh for the Empowerment and Livelihood Improvement “Nuton Jibon� Project (Also Called Social Investment Program Project - SIPP II). May 28, 2010 World Bank. 2011a. “Aide Memoire: Technical Support Mission for the Social Investment Program Project (SIPPI) and the Empowerment and Livelihood Impro vement “Nuton Jibon� Project (SIPP II), July 20- August 4, 2011. World Bank. 2011b. Restructuring Paper on a Proposed Project Restructuring of Empowerment and Livelihood Improvement "Nuton Jibon" Project (P073886, Cr.4757-BD) In The Amount Of SDR 76.2 Million (US$115 Million Equivalent) May 28, 2010 to the Government of The Peoples’ Republic Of Bangladesh. November 9, 2011. Report No. 65595 v.1. World Bank. 2013. Restructuring Paper on a Proposed Credit Restructuring of the Empowerment and Livelihood Improvement "Nuton Jibon" Project. (IDA Credit 4757 BD) Approved June 23, 2010, to the People’s Republic Of Bangladesh. December 15, 2013. Report No. 83308-BD. World Bank. 2015. Restructuring Paper on a Proposed Credit Restructuring of the Empowerment and Livelihood Improvement "Nuton Jibon" Project. (IDA Credit 4757 BD) Approved June 23, 2010, to the People’s Republic Of Bangladesh. Zaman, Dr. Asif M. 2015. “Technical Audit & Impact Assessment of SIPP -II.� mimeo 61 MAP 62