Document of The World Bank FOR OFFICIAL USE ONLY Report No: ICR00004950 IMPLEMENTATION COMPLETION AND RESULTS REPORT TF0A3458 ON A SMALL GRANT IN THE AMOUNT OF USD 0.67 MILLION TO THE Ministry of Planning and Finance FOR Survey on Household Living Conditions (P158724) 22 January 2020 Poverty And Equity Global Practice East Asia And Pacific Region Regional Vice President: Victoria Kwakwa Acting Country Director: Gevorg Sargsyan Regional Director: Hassan Zaman Practice Manager: Salman Zaidi Task Team Leader(s): Clarence Tsimpo Nkengne ICR Main Contributor: Kimsun Tong, Obert Pimhidzai ACRONYMS CSO Central Statistical Organization CSPro Census and Survey Processing System GA Grant Agreement GAAP Generally Accepted Accounting Principles GoM Government of Myanmar HIES Household Income and Expenditure Survey ICR Implementation Completion and Results Report IHLCS Integrated Household Living Conditions Survey M&E Monitoring and Evaluation MLCS Myanmar Living Conditions Survey MOPF Ministry of Planning and Finance MPLCS Myanmar Poverty and Living Conditions Survey MSA Myanmar Statistical Association MSDP Myanmar Sustainable Development Plan NGO Non-Profit Organization NSDS National Strategy for the Development of Statistics PAD Project Appraisal Document PDO Project Development Objective STEP Systematic Tracking of Exchanges in Procurement UNDP United Nations Development Programme UNFPA United Nations Population Fund TABLE OF CONTENTS DATA SHEET .......................................................................................................................... 1 I. PROJECT CONTEXT AND DEVELOPMENT OBJECTIVES ....................................................... 4 II. OUTCOMES..................................................................................................................... 6 III. KEY FACTORS THAT AFFECTED IMPLEMENTATION AND OUTCOMES ................................ 8 IV. BANK PERFORMANCE, COMPLIANCE ISSUES, AND RISK TO DEVELOPMENT OUTCOME .. 10 V. LESSONS LEARNED AND RECOMMENDATIONS .............................................................. 11 ANNEX 1. RESULTS FRAMEWORK AND KEY OUTPUTS ........................................................... 13 ANNEX 2. PROJECT COST BY COMPONENT ........................................................................... 17 ANNEX 3. LIST OF PEOPLE MET DURING ICR PREPARATION .................................................. 18 The World Bank Survey on Household Living Conditions (P158724) DATA SHEET BASIC INFORMATION Product Information Project ID Project Name P158724 Survey on Household Living Conditions Country Financing Instrument Myanmar Investment Project Financing Original EA Category Revised EA Category Organizations Borrower Implementing Agency Republic of the Union of Myanmar Ministry of Planning and Finance Project Development Objective (PDO) Original PDO To improve the Central Statistical Organization of Myanmar's capacity to produce and analyze quality household survey data. FINANCING FINANCE_T BL Original Amount (US$) Revised Amount (US$) Actual Disbursed (US$) Donor Financing TF-A3458 670,000 624,673 624,673 Total 670,000 624,673 624,673 Total Project Cost 670,000 624,673 624,673 Page 1 of 18 The World Bank Survey on Household Living Conditions (P158724) KEY DATES FIN_TABLE_DAT Approval Effectiveness Original Closing Actual Closing A 13-Sep-2016 16-Sep-2016 18-May-2018 30-Apr-2019 RESTRUCTURING AND/OR ADDITIONAL FINANCING Date(s) Amount Disbursed (US$M) Key Revisions 20-Mar-2018 0.54 Change in Loan Closing Date(s) Change in Implementation Schedule 27-Nov-2018 0.66 Change in Loan Closing Date(s) Change in Implementation Schedule KEY RATINGS Outcome Bank Performance M&E Quality Satisfactory Satisfactory Substantial RATINGS OF PROJECT PERFORMANCE IN ISRs Actual No. Date ISR Archived DO Rating IP Rating Disbursements (US$M) 01 12-Feb-2017 Satisfactory Satisfactory 0.30 02 27-Feb-2018 Satisfactory Satisfactory 0.54 03 26-Nov-2018 Satisfactory Satisfactory 0.66 ADM STAFF Role At Approval At ICR Regional Vice President: Victoria Kwakwa Victoria Kwakwa Country Director: Ulrich Zachau Gevorg Sargsyan Director: Ana L. Revenga Hassan Zaman Practice Manager: Salman Zaidi Salman Zaidi Page 2 of 18 The World Bank Survey on Household Living Conditions (P158724) Task Team Leader(s): Reena C Badiani-Magnusson Clarence Tsimpo Nkengne ICR Contributing Author: Kimsun Tong Page 3 of 18 The World Bank Survey on Household Living Conditions (P158724) I. PROJECT CONTEXT AND DEVELOPMENT OBJECTIVES Context 1. In 2016, Myanmar was transitioning towards democratic governance and political transparency, and realigning to a more market-orientated rather than state-directed economy. The availability of quality and timely statistics was – and remains – a priority during this transition. In 2011, the previous administration announced an economic policy that included statistical development as one of its four pillars. The first concrete results of the new policy was the successful implementation of the Myanmar Population and Housing Census in 2014. On May 3, 2016, the Government of Myanmar (GoM) announced the formation of the Data Accuracy and Quality of Statistics Committee, under the patronage of Vice President II, and embarked on a process to modernize statistical legislation and institutional frameworks for official statistics in Myanmar. This included the development of the country’s first National Strategy for the Development of Statistics (NSDS). 2. Despite improvement, Myanmar was identified as one of 29 countries globally with “extreme data deprivation� in 2015, primarily due to the limited availability of household and enterprise surveys.1 In addition, the country lacked up-to-date statistics on poverty and household welfare at the state and regional level – the last comprehensive Integrated Household Survey having been conducted in 2009/10. In 2015, the World Bank provided support for the implementation of the Myanmar Poverty and Living Conditions Survey (MPLCS) to fill immediate gaps in data at the national level. At such a critical time for development policy and planning in the country, a strengthened national statistical system to support national household surveys that also presented statistics for subnational levels – states and regions – was urgently needed by the government and development partners. 3. The Myanmar National Statistical System was (and is still) highly decentralized and fragmented. Each line ministry and agency collected data for their respective needs and the use of common statistical standards and practices across government ministries, departments and agencies was rare; compliance with international standards was also non- uniform. An example of fragmentation is the high number of household surveys conducted by various ministries and agencies with little or no coordination or protocols in place to assure minimum statistical standards. This led to duplication and uncertainty about the quality of surveys. While the Central Statistical Organization (CSO) is the national statistical authority, its role has evolved and its main focus to date has been on the processing, compilation and dissemination of secondary data. Myanmar’s NSDS gives the CSO a stronger role in conducting and coordinating Household Surveys in the country and outlines the institutional framework for guiding the production and dissemination of official statistics. 4. The Survey on Household Living Conditions Project was designed with the participation of multi-stakeholders and fully aligned with government’s priorities i.e. the key recommendation of the NSDS to harmonize the system of national surveys and the Myanmar Country Partnership Framework (CPF) 2015-2019 (Report No. 95998) – the survey data being an important source for monitoring the CPF’s performance. The overall project was conducted in 1Serajuddin, U., H. Uematsu, C. Wieser, N. Yoshida, and A. Dabalen. 2015. Data Deprivation: Another Deprivation to End. Policy Research Working Paper No 7252. Washington, DC: World Bank, April 2015. Page 4 of 18 The World Bank Survey on Household Living Conditions (P158724) collaboration with UNDP, who provided technical assistance for data entry protocols and also provided broader technical support through an in-country statistics advisor as well as financial support for fieldwork. Project Development Objectives (PDOs) 5. The original PDO, as stated in the Project Appraisal Document (PAD) and the Grant Agreement (GA), was “to improve the Central Statistical Organization of Myanmar’s capacity to produce and analyze quality household survey data�. Key Expected Outcomes and Outcome Indicators 6. The PDO level indicator stated that a Household Survey was to be conducted, with at least 8,000 households surveyed, during a twelve months period (to cater for seasonality). 7. Progress towards the PDO was measured by the following five intermediate indicators:  Survey instrument developed, drawing upon good practices from the Household Income and Expenditure Survey (HIES), the Integrated Household Living Conditions Survey (IHLCS), and the Myanmar Poverty and Living Conditions Survey (MPLCS) and similar surveys from other countries.  CSO survey management team trained in various aspects of household survey management and implementation.  At least 15 days of training delivered to enumerators and supervisors.  CSO and World Bank conducted joint analysis of the Household Survey data.  Double data entry conducted for at least 8,000 households. Components 8. The project had two components: (i) training and fieldwork for the Household Survey; and (ii) data processing and analysis of data from the Household Survey. Component 1: Supporting survey training and fieldwork (US$ 536,000) 9. This component included financial provisions to cover operating costs to conduct training and to implement the large-scale Household Survey. The operating costs included field costs incurred during training and data collection, including the hiring of enumerators on short-term contracts, and transportation costs. Complementary technical assistance, which was financed through a Bank-Executed Trust Fund, focused on enhancing capacity and quality during design, training and fieldwork. Component 2: Supporting data processing and analysis (US$ 134,000) 10. This component supported data processing and analysis of the Household Survey by the CSO. This included the acquisition of goods – software, computers and office equipment – needed to conduct data entry, manage the survey Page 5 of 18 The World Bank Survey on Household Living Conditions (P158724) and to conduct data analysis. Complementary technical assistance financed through a Bank-Executed Trust Fund focused on enhancing analytical capacity. II. OUTCOMES Assessment of Achievement of Each Objective/Outcome 11. The achievement of the Project Development Objective (PDO) is rated Satisfactory. The PDO had two dimensions: (i) to improve the CSO’s capacity to produce quality Household Survey; and (ii) to improve the CSO’s capacity to analyze the data from the Household Survey. Achievement is rated satisfactory for both dimensions. For the first dimension, the capacity and skills of CSO staff – in terms of survey design/implementation, procurement, financial management and Household Survey data analysis – have significantly improved since the beginning of the project. Capacity was enhanced through a series of face-to-face and on-the-job trainings on Household Survey management and implementation, which targeted both managerial and technical level staff. Although the Survey Section of the CSO had never conducted a complex Household Living Conditions Survey before this project, they assumed leadership roles in survey design and implementation, with technical support from the World Bank and UNDP teams. The CSO and government staff in charge of procurement and financial management for the project have learned – to a large extent – how to use various World Bank systems, such as Client Connection and the Systematic Tracking of Exchanges in Procurement (STEP), which enabled them to produce the financial statement reports required by the World Bank financial management team. 12. Regarding the second PDO dimension, CSO staff were trained in the use of statistical software (mainly CSPro and Stata), and on core methods of Household Survey analysis, particularly basic statistics and advanced poverty measurement techniques. CSO staff who received training are now able to independently analyze Household Survey data on non-monetary indicators across a wide spectrum of topics. The first Key Indicators Report, focusing on non- monetary indicators, was finalized in June 2018; the Poverty Report was finalized and widely disseminated in June 2019; and the Socio-Economic report is scheduled to be completed by the end of 2019. Thanks to their acquired skills, CSO staff have been able to substantially contribute to the content and dissemination of these reports. 13. While most of the data collected through this project has been deemed to be of high quality, the quality of food consumption data in two out of fifteen states and regions (Kayin and Rakhine) was compromised due to: (1) the mishandling of non-standard units for food by some enumerators; and (2) possible double counting of food consumed away from home. This highlights the importance of improving quality standards in a harmonized manner across all states and regions. There is also an identified need for further capacity building in analytical skills, including poverty measurement – and more broadly how to use microdata to answer policy questions, and ultimately inform policy decisions. Page 6 of 18 The World Bank Survey on Household Living Conditions (P158724) Overall Outcome Rating 14. In line with the achievement of the PDO rated as Satisfactory, the overall outcome rating is Satisfactory. Significant achievements that can be attributed to the project, include:  A questionnaire for the Living Standard Measurement Survey (which successful integrated questionnaires from three main surveys i.e. HIES, IHLCA and MPLCS) was developed through a wide consultation process that included relevant ministries and other stakeholders (NGOs, Academia, think tanks, development partners, etc.).  A nationally representative sampling frame was developed using the 2014 Population and Housing Census.  The CSO Survey Management Team was trained in various aspects of Household Survey management and implementation.2  Enumerators and supervisors received training for 20 days, which was considered an appropriate length for the training period.  Double data entry using CSPro – a software package used for entering, editing and tabulating census and survey data – was implemented.  The CSO, UNDP and the World Bank jointly analyzed various aspects of the Household Survey and published – or will publish – the following joint-reports: (1) Key Indicators Report; (2) Technical Report; (3) Poverty Report; and (4) Socio-Economic Report (forthcoming). Other Outcomes and Impacts 15. Consultations with current and former CSO Director Generals, CSO staff, development partners (UNFPA, UNDP), and the Myanmar Statistical Association (MSA) revealed that the Key Indicators Report (June 2018) and the Poverty Report (June 2019) are being widely used by policymakers. For example, the share of people living in poverty was highlighted as one of the key indicators for budget allocation for the fiscal year beginning October 2019. The State Counsellor has also referred to the survey findings on several occasions to advocate for policy reforms. 16. The World Bank and UNDP provided several trainings to the CSO. Eight out of 40 participants who attended the training on Listing were CSO staff, and most of them are still working for the CSO in various capacities. All trained supervisors were CSO staff who continue to contribute to various surveys that are implemented by the Survey Section. The Survey Section successfully managed to enter survey data in the field and completed the double data entry within two months after the fieldwork, an important achievement given it was the first time the section had managed a 12- month nationally-representative Household Survey. This provides concrete evidence that the CSO’s capacity to conduct surveys has significantly improved. The various trainings, however, were organized separately by the World Bank and UNDP as part of the collaboration arrangement, and some training content was not fully in line with the training program as originally envisaged by the CSO. In summary, some of the topics that the CSO had originally planned to cover in training were not included due to a lack of coordination between the World Bank, UNDP and the CSO. 2In addition to the DG and the Deputy DG, core members of the CSO survey management team include staff from the survey section and IT section (under which the procurement and financial management team is embedded). Page 7 of 18 The World Bank Survey on Household Living Conditions (P158724) III. KEY FACTORS THAT AFFECTED IMPLEMENTATION AND OUTCOMES 17. The project was approved by the Bank on 13 September 2016, for a Grant of the amount of US$ 0.67 million. The Grant Agreement was signed on 16 September 2016 and the project became effective on the same day (16 September 2016). UNDP contributed significantly to the project by funding the cost of the pilot survey, and per-diems for field work through their existing country program and by providing the necessary technical assistance to ensure that the survey was prepared, implemented and disseminated to the highest quality standards. The division of funding responsibilities between UNDP and the World Bank had been clearly defined during the project preparation stage to minimize financial risks. 18. The original closing date for the project was May 18, 2018. Project activities started on time – the training of listers began in October 2016; household listing took place in November 2016; and the survey was implemented in December 2016, being completed in December 2017 as planned. Data entry began on time and was completed in January 2018. Data analysis, however, was still on-going so the closing date for the project was initially extended to 30 November 2018 to allow for the completion of analytical and dissemination activities. The completion deadline was subsequently further extended to 30 April 2019 for the following reasons: (1) the final tranche of goods procurement was delayed due to the revised government procurement processes; and (2) data analysis took longer than anticipated because of data inconsistencies in the food consumption module for Kayin State and Rakhine State – resulting in a delayed report launch. 19. While the Household Survey was successfully implemented in most areas of the country, it is important to note that the project came short in delivering 100 percent coverage of the initially selected national representative sample due to inaccessibility to armed conflict-affected areas – there was an increase in armed conflict in parts of the country during the survey period (December 2016 to December 2017). The CSO made great efforts to access these areas using the local NGO network and conflict advisers to help facilitate dialogue with ethnic armed organizations in Kachin; by holding direct discussions with the Wa administration in Shan; and by exploring options with NGOs to cover the townships in Northern Rakhine. The CSO also engaged conflict experts to provide support to assess the situation in these areas. Despite these extensive efforts, the CSO was unable to reach seven enumeration areas in Northern Shan controlled by the Wa authorities and two townships in Northern Rakhine State. Inaccessible enumeration areas were replaced with other areas in Rakhine in order to meet the representative sample size for the State. Although the supplemental sample maintained the effective sample size, it did not reduce the bias. The seven enumerations areas in Wa Self-Administered Division were not replaced as Shan State already had a larger proportion of enumeration areas due to its large size (it is the largest State/Region in area in Myanmar). Overall, seven out 1,152 enumeration areas were not covered by the Survey. 20. The low quality of food consumption data in Kayin and Rakhine States was largely due to the mishandling of non-standard units for food by some interviewers, and the possible double counting of food consumed away from home. As a result, the analysis team had to impute total food consumption by using the within-survey imputation method to construct welfare aggregates for poverty measurement and analysis for this sub- sample. Page 8 of 18 The World Bank Survey on Household Living Conditions (P158724) 21. At project design, it was decided not to set up a standalone steering committee to oversee project implementation; instead this role was deferred to the NSDS Cluster on Survey Coordination and Statistical Standards. Following recommendations by the NSDS, a total of 10 clusters were set up to improve coordination across the national statistical system. Most of these clusters, however, were not fully functional and the absence of a proper steering committee limited interaction with the project and the monitoring of project implementation by policymakers. 22. Monitoring and Evaluation (M&E): Progress toward the PDO was monitored through five intermediate indicators: (1) the Household Survey instrument developed; (2) training for the CSO survey management team on Household Survey management and implementation conducted; (3) training for enumerators and supervisors conducted for at least 15 days; (4) the CSO and World Bank conducted a joint analysis of the Household Survey; and (5) double data entry conducted for at least 8,000 households.  M&E design: The PDO indicator and intermediate indicators were well-defined, with clear milestones. The PDO milestones had a logical and sequential set-up, whereby the achievement of each intermediate indicator reflected an important step towards the attainment of the PDO. Overall, the PDO indicators were adequate.  M&E implementation: The monitoring and evaluation of the project was undertaken by the CSO survey management team, with the support of the World Bank team. The PDO indicator and the intermediate indicators were monitored annually through project reports prepared by the CSO.  M&E utilization: The monitoring and evaluation arrangements established for this grant were straightforward and in line with the project PDO and components. Given the logical and sequential set-up of the PDO milestones, the information gathered for M&E was used to inform decisions on readiness to move from one core activity to the next.3 23. Financial Management: Prior to the project, the CSO had no experience in managing donor-funded projects. The existing government financial rules and regulations were also outdated – particularly for travel expenses which constituted a substantial amount of the project costs. These factors had the potential to contribute to increased risks of the misuse of funds and constrain the CSO in conducting the Household Survey effectively. For example, the transportation allowance was not sufficient to motivate enumerators to reach the selected, and sometimes remote, enumeration areas. Furthermore, enumerators were required to submit invoices for accommodation or transportation even if the rates had been agreed in advance. 24. Mitigation measures were agreed and put in place, and included: (i) experienced staff from the finance section were assigned to the project (at least at Assistant Director level with reasonable English skills) to oversee sensitive aspects of the project, such as procurement processes with international firms); (ii) an agreement was reached in advance on the expenditure approval procedures and rates to be used for travel/salary to enumerators; (iii) trainings were provided on financial management and disbursement by the World Bank team; and (iv) after the project ended, an audit was conducted by the Office of the Auditor 3 For example, survey instrument must be developed before conducting training of supervisors and enumerators. Page 9 of 18 The World Bank Survey on Household Living Conditions (P158724) General of the Union of Myanmar. The auditor’s conclusion is that the financial statements of the MLCS were in accordance with Generally Accepted Accounting Principles (GAAP). In addition, the CSO hired an international firm to assist with the development of the reporting template and prepare the statement of expenditure and the 1st Implementation Financial Report, which had to be submitted to the World Bank no later than 45 days after every 6 months. However, 45 days was not enough time to validate all transactions given the CSO’s lack of experience with donor-funded projects – resulting in a delay to the submission of the 1st Implementation Financial Report. Mitigation measures overall, however, were successful in reducing identified risks. 25. Procurement: The procurement capacity of the CSO was very low given their lack of experience with donor- funded projects – procurement experience was limited to small budgets and small purchases. In addition, there was no national legislation on public procurement and no comprehensive procedures on procurement existed within the Ministry of Planning and Finance. The overall procurement risk rating was high during the project appraisal. However, the risk was mitigated by training, advice and supervision from the Bank’s procurement team. Procurement procedures were also revised to meet new government internal processes during the project’s implementation. The new procedures required the establishment of a new tender committee with Deputy Minister level representation. Consequently, the final round of goods procurement equivalent to USD 30,000 was delayed. There was also a high turnover among procurement staff at the CSO that negatively affected procurement performance. 26. Strong ownership and commitment of the CSO: The two broad activities supported by the project i.e. to improve household data production and analytical skills were designed in response to urgent needs (NSDS recommendations) and requests from the Government of Myanmar. The management and staff of the CSO involved in the project saw an opportunity to deliver high quality Household Survey data and improve their own analytical skills, therefore they dedicated the necessary time and efforts to make sure the project was delivered successfully. IV. BANK PERFORMANCE, COMPLIANCE ISSUES, AND RISK TO DEVELOPMENT OUTCOME 27. Bank Performance: The World Bank Task Team worked closely with UNDP to provide technical support to the CSO during the project design, preparation and implementation. Although the task team leader (TTL), along with the procurement specialist, changed during the project implementation, the Bank ensured a smooth transition and these changes did not adversely affect the implementation of the project. The decision to base a TTL in-country improved the World Bank’s ability to support and respond to government requests in a timely manner. As a result, the Bank’s performance is rated Satisfactory. 28. Compliance: No environmental and social safeguards were triggered during the preparation as the project was rated Category C. Page 10 of 18 The World Bank Survey on Household Living Conditions (P158724) 29. Risk to Development Outcomes: The project was designed to improve the CSO of Myanmar’s capacity to produce and analyze quality Household Survey data. While significant progress has been made towards achieving these objectives, the ICR team notes that there remains a need for further capacity building in both the quality control/monitoring of Household Surveys and analytical skills, particularly in regard to monetary welfare indicators. In this regard, the CSO has requested that the World Bank continues to provide support for the preparation and implementation of the next Myanmar Living Conditions Survey – tentatively scheduled for 2022, including training in poverty measurement. 30. Staff turnover and the lack of knowledge transfer mechanisms within the CSO were considered as risks to the sustainability of project achievements. Importantly, the Household Living Conditions Survey is currently only available to development partners that financed the survey. The project will be more sustainable if the survey data are accessible to a broader set of data users, such as the research community, other development partners, stakeholders and students. 31. As the project was being implemented several important milestones were reached that have positive implications for sustainability: (1) a new statistics law was adopted. The new law positions the CSO as the leading agency and coordinator of the national statistics system; (2) the NSDS Action Plan was finalized. The plan includes a survey calendar which states that the MLCS is to be conducted every five years, and that the next MLCS is scheduled for 2022; (3) The Myanmar Sustainable Development Plan (MSDP) was adopted, and the national indicator framework is being finalized. Several of the selected indicators are informed by the MLCS. V. LESSONS LEARNED AND RECOMMENDATIONS 32. Government commitment and ownership. Strong commitment and ownership by the Government of Myanmar was key to the project’s success. With financial and technical support from the World Bank and UNDP, the Government of Myanmar showed a strong commitment to lead the new large-scale Household Survey. The implementation of the MLCS was one of the priority actions identified in the NSDS and the alignment with Government priorities helped to secure ownership. Moreover, during project implementation, regular consultations with various stakeholders, in particular data users, proved to be critical in ensuring that the design and implementation of the project produced relevant data that could be used immediately to inform policy, programs and resource allocation as well as monitoring frameworks. 33. Technical assistance in support of project implementation. It was critical to provide technical assistance from the beginning of the project as this was the first time that the CSO had implemented such a large scale and complex data collection operation. At the beginning of the project, the CSO had very limited experience in conducting such a survey, and very limited experience working with development partners such as the World Bank and UNDP. Technical assistance was provided at various stages of the project to cover all aspects of project implementation including questionnaire design; sampling; hiring and training of enumerators; organization and supervision of field work; and procurement and financial management, which included hiring an international firm to support the financial management team to develop the report template and prepare the 1st Implementation Financial Report. The technical assistant provided contributed significantly to the successful implementation of the project. Page 11 of 18 The World Bank Survey on Household Living Conditions (P158724) 34. Flexibility of rules and regulations given the specificity of the project. Financial management rules and regulations needed to be simplified to facilitate project implementation. For example, enumerators should not have been required to submit accommodation or transportation invoices when the rates had been agreed and approved in advance as this increased the workload of the financial management team. Transportation allowances in remote areas should also have been increased to cover the actual costs incurred to reach these areas. 35. Setting realistic timelines for report delivery. The World Bank team’s schedule to release the Poverty Report four months after the completion of the fieldwork was ambitious – it took six months to finalize the Key Indicators Report and 18 months to finalize the Poverty Report. This was primarily due to the CSO’s lack of experience in managing a large-scale and complex Household Survey, while data cleaning and the construction of consumption aggregates took longer to complete than anticipated. It is important therefore to draw up timelines that are realistic given the capacity of the implementing agency, and to have additional slippage time for unforeseen constraints that may impact on the ability to meet deadlines. 36. Strong coordination. There was a lack of coordination between the project’s partners – the World Bank, UNDP and the CSO – that particularly impacted on the content and timing of various trainings, which resulted in duplication and lengthier training periods for government staff. To ensure relevance and sustainability, it is important for future projects that technical assistance provided by development partners for training activities are anchored to the CSO’s newly-established training center. This would make training activities more effective, sustainable and target a larger number of staffs. Training for enumerators should also focus on strengthening capacity to accurately collect food consumption data. 37. Quality assurance mechanisms. The nationally-representative Household Survey would have benefitted from a steering committee to oversee project implementation and act as a link with high-level management in government to ensure effective implementation of the Survey, and more importantly, to carry over key issues to high-level management within government. In addition, such a Committee would ensure the sustainability of project impact and could help attract funds to finance the next Household Living Conditions, scheduled to be implemented in five years. . Page 12 of 18 The World Bank Survey on Household Living Conditions (P158724) ANNEX 1. RESULTS FRAMEWORK AND KEY OUTPUTS A. RESULTS INDICATORS A.1 PDO Indicators Unit of Formally Revised Actual Achieved at Indicator Name Baseline Original Target Measure Target Completion Household survey is conducted Number 0.00 8000.00 8000.00 13730.00 01-Apr-2016 01-Dec-2016 01-Feb-2017 16-Feb-2018 Comments (achievements against targets): A.2 Intermediate Results Indicators Unit of Formally Revised Actual Achieved at Indicator Name Baseline Original Target Measure Target Completion Household survey instrument is Yes/No N Y Y Y Page 13 of 18 The World Bank Survey on Household Living Conditions (P158724) developed. 01-Apr-2016 01-Dec-2016 01-Feb-2017 16-Feb-2018 Comments (achievements against targets): Unit of Formally Revised Actual Achieved at Indicator Name Baseline Original Target Measure Target Completion Training of CSO survey Yes/No N Y Y Y management team on household survey management 01-Apr-2016 01-Dec-2016 01-Feb-2017 16-Feb-2018 and implementation. Comments (achievements against targets): Unit of Formally Revised Actual Achieved at Indicator Name Baseline Original Target Measure Target Completion Training of enumerators and Days 0.00 15.00 20.00 20.00 supervisors is given. 01-Apr-2016 01-Dec-2016 01-Feb-2017 16-Feb-2018 Comments (achievements against targets): Page 14 of 18 The World Bank Survey on Household Living Conditions (P158724) Unit of Formally Revised Actual Achieved at Indicator Name Baseline Original Target Measure Target Completion CSO and World Bank conduct Yes/No N Y Y Y joint analysis of the household survey. 01-Apr-2016 01-Dec-2016 01-Feb-2017 16-Feb-2018 Comments (achievements against targets): Unit of Formally Revised Actual Achieved at Indicator Name Baseline Original Target Measure Target Completion Double data entry will be Number 0.00 8000.00 8000.00 13730.00 conducted 01-Apr-2016 01-Dec-2016 01-Feb-2017 16-Feb-2018 Comments (achievements against targets): Page 15 of 18 The World Bank Survey on Household Living Conditions (P158724) B. ORGANIZATION OF THE ASSESSMENT OF THE PDO Objective/Outcome 1. Household Survey was conducted with 13,730 observations Outcome Indicators 1. Survey instrument was developed 2. CSO central staff were trained in various aspects of household survey management and implementation Intermediate Results Indicators 3. Enumerators and supervisors were trained for 20 days 4. CSO and World Bank conducted joint analysis of household survey data 5. Double data entry was conducted 1. MLCS Technical Report 2. MLCS Key Indicators Report Key Outputs by Component 3. MLCS Poverty Report (linked to the achievement of the Objective/Outcome) 4. MLCS Socio-economic Report (to be published by end 2019) 5. MLCS dataset, with a total of 13,730 surveyed households Page 16 of 18 The World Bank Survey on Household Living Conditions (P158724) . ANNEX 2. PROJECT COST BY COMPONENT Amount at Approval Actual at Project Percentage of Approval Components (US$M) Closing (US$M) (US$M) Supporting survey training 0 .54 0 and fieldwork Supporting data processing 0 .13 0 and analysis Total 0.00 0.67 0.00 Page 17 of 18 The World Bank Survey on Household Living Conditions (P158724) ANNEX 3. LIST OF PEOPLE MET DURING ICR PREPARATION Development Partners  Mr. Daniel Msonda, Program Specialist, UNFPA, msonda@unfpa.org  Sa Si Thu Htike San, Project Analyst, UNDP, sa.si.thu.htike.san@undp.org Myanmar Statistical Association (MSA)  U Nyan Lin, Senior Survey Statistician, Former Team Leader of IHLCA Project Technical Unit  Prof. Lay Kyi (Ph.D), Former Pro-Rector/Head of Department of Statistics, Yangon University of Economics  Htun Htun Oo, Managing Director, Decision Support Service Co., Ltd., Former Statistician/Economist of IHLCA Project Technical Unit CSO, Management Team  U San Myint, Director General  Daw Khin Seet Yi, Deputy Director, Survey Section  Dr. Wah Wah Maung, Director General, Pension Department, Ministry of Planning and Finance – Former Director General of CSO CSO, Procurement Team  Daw Yu Yu Naing, Deputy Director, System and Application Section (IT)  Daw Hnin Pyae Pyae Nyo, Assistant Director, System and Application Section (IT)  Daw Hsu Hnin Wai, Assistant Director, System and Application Section (IT)  U Win Zaw Oo, Staff Officer, Survey Section  U Aug Moe Oo, Deputy Staff Officer, Survey Section CSO, Financial Management Team  Daw Su Su Aung, Assistant Director, Account Section  Daw Than Than Soe, Staff Officer, Account Section CSO, Technical Team  Daw Khin Seet Yi, Deputy Director, Survey Section  Daw Hnin Wai Khine, Deputy Staff Officer, Survey Section World Bank, Procurement and Financial Management Specialist  Keymon Soe, Financial Management Specialist  Yin Win Khine, Procurement Specialist Page 18 of 18