Country Practice Area(Lead) Additional Financing Nepal Education P125610,P152377,P152377

Public Disclosure Authorized Independent Evaluation Group (IEG) 1. Project Data Report Number : ICRR0020684 Public Disclosure Authorized Public Disclosure Authorized Project ID P113441 Project Name NP: School Sector Reform Program Country Practice Area(Lead) Additional Financing Nepal Education P125610,P152377,P152377 L/C/TF Number(s) Closing Date (Original) Total Project Cost (USD) IDA-46520,IDA-52500,IDA- H5180,IDA-H8460,TF-98235,TF- A0779,TF-A0843,TF-A0915 15-Dec-2014 9,682,000,000.00 Bank Approval Date 22-Sep-2009 Closing Date (Actual) 15-Jul-2016 IBRD/IDA (USD) Grants (USD) Original Commitment 130,000,000.00 183,300,000.00 Revised Commitment 223,948,711.07 181,060,472.85 Actual 216,222,567.23 154,447,478.45 Prepared by Reviewed by ICR Review Coordinator Group Katharina Ferl Susan Ann Caceres Joy Behrens IEGHC (Unit 2) Public Disclosure Authorized PHPROJECTDATATBL Project ID P152377 Project Name SSRP Additional Financing ( P152377 ) L/C/TF Number(s) Closing Date (Original) Total Project Cost (USD) Page 1 of 16

PHEVALUNDERTAKENLBL Independent Evaluation Group (IEG) 7,047,000,000.00 Bank Approval Date 13-Jan-2016 Closing Date (Actual) IBRD/IDA (USD) Grants (USD) Original Commitment 0.00 0.00 Revised Commitment 0.00 0.00 Actual 0.00 0.00 2. Project Objectives and Components a. Objectives According to the Project Appraisal Document (PAD) (p. ii) and the Financing Agreement of November 27, 2009 (p. 5) the development objective was to increase access to and improve quality of school education, particularly basic education (Grades 1-8), especially for children from marginalized groups. This review does not apply a split rating: Although there were some revisions in outcome targets at additional financing, targets were not lowered; in fact, many were made more ambitious. b. Were the project objectives/key associated outcome targets revised during implementation? Yes Did the Board approve the revised objectives/key associated outcome targets? No c. Will a split evaluation be undertaken? No d. Components The project included three components: Component 1: Basic Education (appraisal estimate for IDA financing US$102 million, Additional Financing US$223.81 million, actual US$292.9 million): This component was to finance activities to increase access and inclusion in basic education such as outreach programs, scholarships and mid-day meals in targeted districts. Also, this component was to finance improving the quality of basic education by implementing different activities such as providing free text books and multilingual teachers, improving the pupil-teacher ratio, updating the curriculum, and introducing an assessment system. In addition, this component was to finance the operational costs of approximately 25,000 Early Childhood and Education Page 2 of 16

centers and community learning centers. Finally, this component was to finance teacher salaries. Component 2: Secondary Education (appraisal estimate for IDA financing US$25 million, Additional Financing US$33.9 million, actual US$70.4 million): This component was to finance activities to increase access and equity in secondary education through the expansion of physical facilities, providing scholarships to targeted groups and alternative open learning centers, non-formal education for adult women, and grant support to traditional schools. This component was also to finance improvements in the quality of secondary education through the development of norms and standards for curricula, education materials, teachers, school environment, and assessment. This component piloted technical education and vocational training for grades 7-10. It financed teacher salaries through the Per Capita Financing approach (i.e. block grants based on the number of students in the school). Component 3: Institutional Capacity Strengthening (appraisal estimate US$3 million, Additional Financing US$5.66 million, actual US$7.4 million): This component was to finance teacher recruitment, licensing, and professional development. The component financed capacity building for schools and the Implementing Agency, as well as Monitoring and Evaluation (M&E) activities e. Comments on Project Cost, Financing, Borrower Contribution, and Dates Project Cost: The project was estimated to cost US$130 million. Actual costs were US$370.7 million, as Additional Financing of US$283 was provided. An amount of US$8.1 million was cancelled and US$29.75 million was undisbursed, but much of this amount is expected to disburse (US$21.8 million) during the post-project period (ICR p. 33). The project lost approximately US$4.8 million due to exchange rate fluctuations. The Borrower reimbursed the Bank of NPR 31.5 million (approximately US$305,463) for ineligible expenditures. Financing: The project was to be financed by a US$130 million loan by the International Development Agency (IDA). Additional Financing of US$283.3 million was provided through five grants and a credit. Development Partners that also provided financing included: the US Agency for International Development (US$2.33 million), the Asian Development Bank (US$163.8 million), the Australian Agency for International Development (US$14.54 million), the Danish International Development Assistance (US$58.68 million), the Department for International Development (financed through the European Commission), the European Commission (US$57.02 million), the Finnish Ministry of Foreign Affairs (US$25.39 million), the Education for All Fast Track Initiative (US$117.8 million), the Global Partnership for Education (US$36.7 million), the Norwegian Agency for Development Cooperation (US$38.3 million), the UN Children s Fund (US$2.55 million) and the Japanese International Cooperation Agency (US$5.67 million). Borrower Contribution: There were no contributions by the Borrower, nor were any planned. Dates: On May 15, 2014 the project was extended for 19 months from December 15, 2014 to July 15, 2016 to align it with the Government's seven-year program timeframe (2009/10 2015/16). The project was restructured three times: On December 7, 2012, the project received a US$120 million grant from the Education for All-Fast Track Initiative. Also, the legal covenants were modified to add the requirement for the borrower to prepare a completion report on the project. On July 8, 2013, the project received an IDA credit in the amount of US$55 million and an IDA grant Page 3 of 16

in the amount of US$45 million to close a financing gap. On January 22, 2016, the project received additional financing grants in the amount of US$4 million from the Education for All Children Multi-Donor Trust Fund and US$59.3 million from the Global Partnership for Education to close a financing gap and incentivize higher performance towards equity, efficiency, and improved learning outcomes. Five new Disbursement Linked Indicators (DLIs) were included in the Results Framework. 3. Relevance of Objectives & Design a. Relevance of Objectives High: The objectives of the project were highly relevant. Even though Nepal had made significant improvements in its education sector, challenges still remained at the time of project appraisal. The sector suffered from low quality at all levels of schooling and low internal efficiency (as 16% of children dropped out after the first grade and almost 30% repeated the first grade, and less than 60% completed primary grades). Approximately 10% of primary age children were not in school. The objectives of the project were in line with the government s School Sector Reform Program, which aimed to address the challenges noted above. The objectives of the project were in line with the Bank s Interim Strategy (2007-2009), which aimed to ensure equitable access to quality education and protect key reforms undertaken such as decision-making and school management to communities. Also, the objectives of the project were in line with the Bank s Country Assistance Strategy (2014-2016), which focused on more equitable access to education and higher quality and relevance of skills. Rating Revised Rating High --- b. Relevance of Design Substantial: The project was built on lessons learned from previous education projects and programs in the country, which allowed for continuity. Activities to improve access to basic school education included the provision of outreach programs, scholarships and mid-day meals in districts targeted through a mapping exercise. Activities to improve access to secondary education included: construction of facilities, provision of scholarships to targeted groups and alternative open learning centers, non-formal education for adult women, and grants to traditional schools. Activities to improve the quality of education included the provision of free textbooks and multilingual teachers, improvement of the pupil-teacher ratio, revision of the curriculum, and the introduction of an assessment system. Additional activities to be financed were teacher recruitment and teacher professional development, and teacher licensing enhancements. The underlying assumptions about how program actions would lead to intended outcomes were logically linked and aligned with the objectives. However, the project would have benefited from going beyond mainly financing inputs to focusing on how School Management Committees and schools could deliver inclusive education of high quality (ICR p.11). Page 4 of 16

Rating Revised Rating Substantial --- 4. Achievement of Objectives (Efficacy) PHEFFICACYTBL Objective 1 Objective To increase access to school education, particularly basic education (Grades 1-8), especially for children from marginalized groups: Rationale Outputs: Between 2009 and 2016 the number of schools increased from 32,130 to 35,222. During the project lifetime, 32,000 classrooms were constructed to accommodate 95,000 students. The number of schools offering the full cycle of basic education nearly doubled from 1,597 schools in 2009/10 to 3,175 schools in 2016/17. The percentage of basic education and secondary students in community schools remained approximately at 85% and 66%, respectively, out of all schools. 12,472 schools were managed by communities since 2014. Between 2009 and 2016, the number of school grade levels and early childhood development centers increased in all eco-belts except for a minor decrease at basic education in Kathmandu Valley. The size of student population increased all grade levels, except primary education between 2009 and 2016. An equity index was developed to identify districts with the highest rate of out-of-school children. A Welcome to School campaign was conducted in all 75 districts to identify and enroll out-of-school children. Free basic education strategies were implemented to target 12 regions which were lagging behind and support them with financial and other resources. Tracking of out-of-school children was implemented In 53 districts. A geographic information system mapping capability was established to inform school planning based on population need. A mid-day meal program was provided to approximately 800,000 students. 32 million textbooks were supplied for free to schools within the first two weeks of the school year. Curriculum was digitalized and made available on websites and through mobile applications for greater accessibility nationwide. Primary education curriculum and textbooks developed in 21 languages. 42 supplementary reading materials were produced in 15 languages. Page 5 of 16

By 2015/16 69 local languages were used as a medium of instruction benefiting a total of 850,831 primary students in over 7,676 schools. 82% of schools had toilet facilities (among which 85% have separate girls toilet facilities). 15,000 teachers were supported through per-capita financing on an annual basis. By 2015/16 scholarships were provided through per-capita financing to over 900,000 Dalits (members of the lowest social class in the Hindu caste system), almost 40,000 children with disabilities, over 150,000 children from other deprived communities, and 1.8 million girls. Furthermore, in 2015/2016 almost 20% of all secondary students received scholarships, including over 80,000 Daltis and 15,401 students from other marginalized groups. However, the completion report (p. 58) noted a continued need to improve targeting and coherence, prioritization and efficiency in planning and implementing strategies for equal access. Outcomes: The net enrollment rate for primary education (grades 1 to 5) increased from 92% in 2009 to 97% in 2016, nearly achieving the target of 98% for all phases. The net enrollment rate for basic education (grades 1 to 8) increased from 73% in 2009 to 91% in 2016, surpassing the target of 85% for phases 1 and 2, nearly achieving the target of 92% for phases 3 and 4. The gross enrollment rate for primary education (grades 1 to 5) decreased from 147% in 2009 to 134% in 2016, achieving the targets of all phases. The decrease in the rate shows the trend of participation of appropriate age group in the respective education level. The gross enrollment rate for basic education (grades 1 to 8) decreased from 123% in 2009 to 122% in 2016, almost achieving the target of 132% for Phases 1 and 2 and achieving the target of 120% for Phases 3 and 4. The gross enrollment rate is expected to decrease during the next program, as a result of increased efficiency at primary education during the project period. The net enrollment rate for secondary education (grades 9 to 12) increased from 21% in 2009 to 39% in 2016, achieving the targets for all phases. This result reflects the increase in completion rates at the primary and basic education levels. The average net enrollment rate in basic education in the 15 districts categorized as the most lagging areas increased from 73.6% in 2009 to 85.6% in 2016. Between 2009 and 2016 the number of students per teacher decreased from 32 to 22 students per teacher for primary education, and from 34 to 24 students per teacher for basic education (grades 1-8) and 40 to 34 students per teacher (grades 6-8). Also, during the same time period the number of students per teacher decreased in secondary education from 35 to 24 students per teacher (grades 9-12), and from 27 to 24 students per teacher (grades 9-10). For grades 11-12 the ratio increased from 20 to 25 students per teacher. The gender parity index, which measures the relative access of males and females, for primary education (grades 1 to 5) increased from 98% in 2009 to 99% in 2016, almost achieving the target of 100% for all phases. The gender parity index in enrollment for basic education (grades 1 to 8) increased from 95% in 2009 to 100% in 2016, achieving the target of 100% for all phases. The gender parity index in enrollment for secondary education (grades 9 to 12) increased from 91% in Page 6 of 16

2009 to 98% in 2016, achieving the target of 95% for phases 1 and 2 and almost achieving the target of 100% for phases 3 and 4. The Public Expenditure Tracking Survey found that 35% of the education budget at basic level benefited 20% of the poorest students. This was a result of the pro-poor budgeting and increased expenditure on targeted scholarships to promote access and retention in school. Under the project the scope and coverage of targeted scholarships increased from 31% to 57% of all children enrolled in basic education. 2016/2017 data of the Education Management Information System showed an increase in enrollment shares of Dalits at primary, basic and secondary levels. For example, the enrollment rate in grades 6 to 8 increased from 13% in 2009 to 16% in 2016/17. Rating High PHREVDELTBL PHEFFICACYTBL Objective 2 Objective Improve quality of school education, particularly basic education (Grades 1-8), especially for children from marginalized groups: Rationale Outputs: 5,500 new Early Childhood Education Centers were established. 2,200 Community Learning Centers were constructed. 42,000 teachers were recruited through per-capita financing or through schools own resources. 4,500 libraries were constructed. Information Communication Technology facilities for teachers were constructed in 4,000 schools. Alternative secondary education for adult learners was provided through grants to 400 schools for open education, 150 schools running non-formal secondary education, and 100 religious schools. Standards and norms for curricula, education materials, teachers, school environment and examination systems were developed. The number of schools managed by communities increased from 8,500 schools in 2009 to 12,471 schools in 2016. The target of 26,500 schools for phase 1 and 2 and the target of 16,000 schools for phase 3 and 4 were not achieved. The original target was reviewed during the midterm review and assessed as too ambitious. A student learning assessment for a sample of students in grades 3, 5 and 8 was implemented, achieving the target for all phases. An equity strategy for basic education was developed and implemented, meeting the target. A total of 32 million full sets of textbooks were distributed to students under the project. By 2016/17, 90% of students in grades 1 to 10 received their full set of textbooks at the start of the Page 7 of 16

school year, compared with about half of students (grades 1 to 10) in 2009. 10 day Teacher-Professional Development modules benefitted 141,000 teachers using training packages developed by the National Council for Educational Development. Outcomes: The percentage of fully trained teachers (completed National Teacher Professional Development Program) at basic and secondary education level increased from 91% in 2009 to 93% of basic education teachers and 82% of lower secondary education teachers. The target of phases 1 and 2 was fully achieved while the target for phase 3 and 4 was nearly achieved for basic and secondary education teachers. By the end of the project, nearly half of the students who initially enrolled in Grade 1 remained in school to Grade 10. The completion rate for primary education (grade 5) increased from 58% in 2009 to 81% in 2016, achieving the targets of all phases. The completion rate for basic education (grade 8) increased from 41% in 2009 to 70% in 2016, achieving the targets for all phases. Drop-out rates decreased for grade 1 from 10% in 2009 to 5% in 2016/17, for grade 5 from 8% in 2009 to 4% in 2016/17, for grade 8 from 7% in 2009 to 6% in 2016/17, for grades 9-10 from 9% in 2009 to 5% in 2016/17. Promotion rates increased for grade 1 from 64% in 2009 to 81% in 2016/17, for grade 5 from 86% in 2009 to 92% in 2016/17, for grade 8 from 85% in 2009 to 90% in 2016/17, for grades 9-10 from 85% in 2009 to 91% in 2016/17. Repetition rates decreased for grade 1 from 27% in 2009 to 14% in 2016/17, for grade 5 from 7% in 2009 to 4% in 2016/17, for grade 8 from 12% in 2009 to 4% in 2016/17, for grades 9 to 10 from 7% in 2009 to 3% in 2016/17. In the 15 most lagging districts the average cohort survival rates to Grade 5 and Grade 8 were 87% and 76%, respectively, in 2015/16 and therefore on the same level as the national averages. In these 15 districts the number of student years invested per basic education graduate was 11.4 years, also close to the national average of 11.3 years. As of 2015/16, 69 local languages were being used as the medium of instruction, supporting the education of marginalized groups. Furthermore, as of 2016, curriculum and textbooks were being developed in 21 languages for the primary level. The ICR reports considerable fluctuations in the School Leaving Certificate and Higher Secondary Education pass rates for several decades. Through the Additional Financing, the project introduced a single-subject certification of the student s performance in individual subjects instead of an aggregate pass/fail evaluation allowing students, who would have dropped out after Grade 10 to continue their studies resulting in an increase of the system s efficiency, average educational attainment, and knowledge accumulation. While the completion report provided evidence of improved progression and flow, it did not provide robust Page 8 of 16

evidence of improved quality, and acknowledged that "Although the SSRP made major progress in improving access to education, there is less evidence on improvement in quality of education" (ICR p. 40). The detailed results of a National Assessment of Student Achievement were not shared in the completion report, but it noted that the results raised awareness among parents and educational community regarding the low achievement at all levels. One of the Development Partners (p. 59) wrote that the results have so far been disappointing in terms of learning outcomes. Rating Modest PHREVDELTBL PHREVISEDTBL 5. Efficiency Substantial: Both the PAD and the ICR conducted an economic analysis. The PAD (p. 15) included a costbenefit analysis identifying education quantity, education quality and relevance, and education s internal efficiency as benefit streams. The measure of impact of the project on these benefits was estimated as the difference between with the project and without the project. The costs consisted of the cost of the project, household expenditure and opportunity costs in the form of foregone earnings. The analysis estimated an Internal Rate of Return (IRR) of 39% and a Net Present Value of US$931 million, assuming a discount rate of 12%, indicating that this project provides economic value. The ICR conducted similar calculations as the PAD. The ICR s cost-benefit analysis estimated, under different scenarios, an IRR ranging from 13.7% to 17.9% and a Net Present Value between US$500 million and US$637 million, suggesting that the project has been a worthwhile investment. Although the project was extended for 19 months, the extension was done to align with the Government's program timeline and because of the Additional Financing provided. During the Additional Financing, the overall financing framework of the Government s program was re-examined to take different scenarios into account, in order to prioritize spending, identify savings, and make more efficient use of resources. However, the project also experienced some inefficiencies such as the cancellations of funds and the accumulation of ineligible expenditures. Some of the cancellations occurred because disbursements were based on actual expenditures of the Government. Part of the cancelled funds remained within the Bank's country program for Nepal. Considering all of these factors, efficiency is rated Substantial. Efficiency Rating Substantial a. If available, enter the Economic Rate of Return (ERR) and/or Financial Rate of Return (FRR) at appraisal and the re-estimated value at evaluation: Page 9 of 16

Rate Available? Point value (%) *Coverage/Scope (%) Appraisal 0 ICR Estimate 0 0 Not Applicable 0 Not Applicable * Refers to percent of total project cost for which ERR/FRR was calculated. 6. Outcome Relevance of objectives was high, given the challenges Nepal s education sector was facing. Relevance of design was substantial, as the underlying assumptions about how program actions would lead to intended outcomes were logical aligned with the objectives. Achievement of the objective to increase access to education was High. Achievement of the objective to improve quality of education was Modest. Efficiency was rated Substantial. Outcome was rated Satisfactory, implying minor shortcomings in design and implementation. a. Outcome Rating Satisfactory 7. Rationale for Risk to Development Outcome Rating The Government is committed to reform the country s education sector and developed a School Sector Development Plan for the period 2016-2023 and the Development Strategy 2030. This project is followed by a new Bank project ( School Sector Development Program ) which aims to improve the quality, equitable access, and efficiency of basic and secondary education. The National Planning Commission and the Ministry of Finance expressed willingness to increase the share of budget devoted to the education sector. However, financing by Development Partners is still needed. Also, addressing the damages from the devastating 2015 earthquakes put even more constraints on an already tight government budget. Taking everything together, the risk to development outcome rating is Modest. a. Risk to Development Outcome Rating Modest 8. Assessment of Bank Performance Page 10 of 16

a. Quality-at-Entry The Bank team collaborated with the Government and with other Development Partners during project preparation. The project was built on lessons learned from the Education For All Program, Community School Support Program, the Secondary Education Support Program and Teacher Education Project. The Bank team identified relevant risk factors. These risk factors included the stalling of reforms and financial sustainability of populist programs, delay in legislative processes, and a gap between Nepal s legal framework and its capacity to implement its laws. However, the design stage assumed an active local government at the district and village level, but processes were not fully linked with responsibilities and authorities, which negatively impacted implementation (and reduced full benefit to students) (ICR p. 13). Preparation would have benefited from more diagnostic work on the differences between the new program and the previous Education-for-All program in order to identify potential implementation challenges and stronger analysis of the processes critical for achieving the quality, rather than focusing on inputs. The indicators listed in the Results Framework as measures of improved quality were in reality measures of student flow and progress. Quality-at-Entry Rating Moderately Satisfactory b. Quality of supervision The Bank team collaborated well with the Development Partners throughout project implementation ( ICR p. 13). The Bank was proactive in identifying development challenges, provided enhanced supervision and technical assistance. The Bank team supported the Government in staying on track towards achieving the development objective. The Bank team provided intensive fiduciary supervision. The Bank restructured the project several times to provide Additional Financing and modify covenants to include more detail. In the restructurings in 2013 and 2016, Incentive-linked indicators and Disbursement-linked indicators were added to the Results Framework to leverage financing for results. The Bank team strengthened the procurement capacity at the Department of Education (DOE) by providing extensive technical assistance and monitoring. Implementation Status Reports were candid and substantive and Bank management provided close oversight. However, internal ratings of safeguard compliance were consistently satisfactory, while an internal assessment in 2014 noted insufficient budget to ensure compliance at the school and district level to implement the Environmental Management Framework and Social Management Framework. Quality of Supervision Rating Satisfactory Overall Bank Performance Rating Moderately Satisfactory 9. Assessment of Borrower Performance a. Government Performance This program was the Government's number one priority (ICR p. 12). The Government worked closely with Page 11 of 16

the Bank during project preparation and was committed to achieving the development objectives throughout project implementation despite political instability and frequent changes in government (and an earthquake that killed nearly 9,000 people). While the Government passed supporting reforms, there were delays in the legislative process (i.e. approval of the new constitution or the amendment of the education act, the main pillars for the reform program) that negatively impacted implementation. There were delays in approving the budget, thus impacting the construction of civil works. However, budget releases to the local entities (DEOs, Regional RCs, and schools) were predictable. The Office of Auditor General reviewed financial statements and followed up on audit observations and recommendations for audit reports. The Office of Auditor General and the Office of Financial Controller General improved internal controls to reduce the level of ineligible expenses. Government Performance Rating Moderately Satisfactory b. Implementing Agency Performance The Department of Education (DOE) acted as the Implementing Agency while the Ministry of Education (MOE) was the executing agency. Both entities shared the overall responsibility for M&E. The DOE prepared and implemented the Annual Work Plan Budget and the Annual Strategic Implementation Plan. The DOE coordinated activities that were implemented by other agencies. However, there was inadequate capacity at the school and district level and inadequate supervision of these entities in relation to financial management, procurement, and environmental management. For example, there was evidence of leakages in the use of per-capita financing resources because of the weak capacity at the DOE. The internal control framework at the MOE had shortcomings, leading to additional risks. All of this was compounded by the fact that MOE had to take responsibility for the district level implementation (as there was no local government with elected representatives) until the constitution was approved. On the positive side, the frequent turnover of higher administrative officials to other sections in the MOE and DOE positively impacted on institutional memory and capacity, as these officials implemented lessons learned. Implementing Agency Performance Rating Moderately Satisfactory Overall Borrower Performance Rating Moderately Satisfactory 10. M&E Design, Implementation, & Utilization a. M&E Design The objectives of the project were clearly specified and reflected in the selected indicators, with the exception of measuring improved quality. The original Results Framework included four PDO indicators and 22 Intermediate Outcome Indicators. The selected indicators were measurable. Page 12 of 16

The Education Management Information System (EMIS) incorporated school censuses from beginning and end of the school year. Additional data were collected from household surveys such as the Nepal Living Standard Survey, Demographic and Health Survey, and Population Census to verify data collected through EMIS. In addition, qualitative and quantitative studies were to be conducted every year. b. M&E Implementation During two of the project restructurings the Results Framework was revised. In 2013 four Incentive-Linked Indicators (ILIs) and in 2016 three indicators related to equity, efficiency and learning outcomes, and two Disbursement-Linked Indicators (DLIs) were added. The Bank team stated that the indicators were measured on a regular basis. The MOE conducted a census twice a year, analyzed the results, and compiled them into a report for review at each joint annual review and joint consultation meeting. A software-based school-based Integrated Education Information Management System was implemented in 72 of all 75 districts during project implementation. The Bank team stated that the data in the Flash system were of adequate quality and steadily improved as a result of various interventions over the project s lifetime. From the outset, the responsibilities for data collection and cross check on consistency were well defined for each agency/level, and validation rules were incorporated into the Flash data entry software. Technical support helped improve institutional mechanisms for data quality, verification, and follow-up on results. During the program period, EMIS moved from paper-based data collection to excel-based (electronic off-line) data collection, which improved the timeliness and reliability of data. Several quantitative and qualitative studies were conducted to inform and support the program implementation. c. M&E Utilization The M&E system is the Government s main data base. However, the data could be used more effectively to inform decision making at central and local levels (ICR p. 16). M&E Quality Rating Substantial 11. Other Issues a. Safeguards The project was classified as Category B for OP/BP 4.01 (Environmental Assessment) and triggered OP/BP 4.10 (Indigenous People). The project prepared an Environmental Management Framework, a Social Management Framework, and a Vulnerability Community Development Framework. Building on its political commitment to gender and social inclusion, the MOE included Gender Equity and Social Inclusion in its program and monitoring framework. A DOE staff member was responsible for ensuring compliance with the safeguard policies. While the Bank rated environmental and social safeguard compliance consistently Page 13 of 16

Satisfactory in its internal reports, an assessment of the status of the school safety program documented in 2014 evidence that there was limited capacity, insufficient budget to ensure compliance at the school and district level to implement the Environmental Management Framework and Social Management Framework. Instances of inadequate construction were found. b. Fiduciary Compliance Financial Management The project s financial management was complex and involved approximately 200 cost centers with the majority of expenditures taking place at the school level. The DOE lacked capacity in terms of planning, budgeting, record keeping and monitoring across all levels. The Bank provided extensive technical assistance to address these issues and conducted a series of independent fiduciary and forensic reviews. The Bank team stated that financial reports were submitted on a regular basis. However, due to the DOE s capacity constraints, the financial reports initially were delayed and of weak quality. DOE s follow-up on the World Bank s detailed feedback on the first versions of reports resulted in final reports submitted that were satisfactory to the Bank. Delays in approving the government s budget led to delays in the construction of civil works, showing a large share of the construction budget as ineligible expenditure in the Office of Auditor General s audit observations. The external audit opinions were qualified for audits from 2009/10 to 2013/14. The Government reimbursed ineligible expenditures. Procurement The Bank s procurement guidelines were followed and implementation delays were not due to procurement issues. However, the ICR (p. 16) states that the procurement capacity at the DEO and at schools was weak for monitoring and supervising the construction of facilities. Additional procurement staff at the MOE were hired. Procurement trainings for district offices were organized by the Bank. c. Unintended impacts (Positive or Negative) None noted. d. Other --- 12. Ratings Ratings ICR IEG Reason for Disagreements/Comment Page 14 of 16

Outcome Satisfactory Satisfactory --- Risk to Development Outcome Negligible Modest Bank Performance Satisfactory Moderately Satisfactory Borrower Performance Moderately Satisfactory Moderately Satisfactory --- Quality of ICR Substantial --- Contiuned need for funding of financing gap by Development Partners. The 2015 earthquakes put additional pressure on an already tight government budget. The ICR also rates Quality at Entry Moderately Satisfactory and Quality of Supervision as Satisfactory. If the ICR had aggregated the two ratings as per guidelines, the overall Bank Performance rating in the ICR would also have been Moderately Satisfactory. Note When insufficient information is provided by the Bank for IEG to arrive at a clear rating, IEG will downgrade the relevant ratings as warranted beginning July 1, 2006. The "Reason for Disagreement/Comments" column could cross-reference other sections of the ICR Review, as appropriate. 13. Lessons The ICR (p. 30-31) included several lessons that IEG synthesizes as follows: Decentralization in the education sector can be beneficial for dividing responsibilities actors on all different levels, but accountability of roles and responsibilities needs to be clearly defined. In this project, the Ministry of Education was responsible for decision making while the district authorities implemented those decisions, and the local communities delivered front line services, allowing for including all different actors critical for achieving the development objectives. However, a compliance system was lacking, and there were accountability gaps. Collaboration between all development partners can lead to more efficient and effective preparation and implementation. In this project, all development partners agreed on a single appraisal document, including a joint economic and financial analysis. Also, when a development partner provided some technical assistance, it benefited the collective preparation process. This allowed for a more coherent approach and reduced the burden on the Ministry of Education. Page 15 of 16

14. Assessment Recommended? No 15. Comments on Quality of ICR The ICR provides a good overview of project preparation and implementation with excellent insights. Economic analysis is part of the report. The report has extensive data from which to draw upon to demonstrate efficacy, particularly in relation to access. The report is concise and well written. The Borrower s report was good quality. The comments by Development Partners were helpful and utilized in this review. While the report provides rich learning for operational staff, the lessons at the end could have been better crafted. Overall quality of the ICR is rated substantial. a. Quality of ICR Rating Substantial Page 16 of 16