The World Bank Report No: ICR00004216 IMPLEMENTATION COMPLETION AND RESULTS REPORT (IDA 57520 MG) ON A CREDIT IN THE AMOUNT OF SDR 39.4 MILLION (US $ 55.0 MILLION EQUIVALENT) TO THE REPUBLIC OF MADAGASCAR FOR A RESILIENCE DEVELOPMENT POLICY OPERATION December 18, 2017 Macroeconomics and Fiscal Management Global Practice AFCS2 Africa Region CURRENCY EQUIVALENTS (Exchange Rate Effective as of November 30, 2017) Currency Unit = Ariary US$ 1 = Ariary US$ 1 = SDR FISCAL YEAR [January 1 – December 31] ABBREVIATIONS AND ACRONYMS AfDB African Development Bank AGOA African Growth and Opportunities Act AUGURE Application Unique pour la Gestion Uniforme des Ressources Humaines de l’Etat (Unique Identifier for Human Resources Management) BCM Central Bank of Madagascar BIANCO Independent Anti-Corruption Bureau BOP Balance Of Payments CPIA Country Policies and Institutional Assessment CPF Country Partnership Framework DeMPA Debt Management Performance Assessment DGE Large Companies Department DTIS Diagnostic Trade Integration Study DPO Development Policy Operation EDBM Economic Development Board of Madagascar EDP Economic Development Paper ENSOMD Millennium Development Goal Surveys EPM Household Survey EPZ Export Processing Zone FDI Foreign Direct Investment FSAP Financial Sector Assessment Program FY Fiscal Year GDP Gross Domestic Product GRS Grievance Redress Service IDA International Development Association IEG Independent Evaluation Group IFC International Finance Corporation IMF International Monetary Fund INSTAT National Institute of Statistics JIRAMA State-Owned Electricity and Water Company (Jiro sy Rano Malagasy) LDP Letter of Development Policy LMIL Large Mining Investments Law MFB Ministry of Finance and Budget MCS Ministry of Civil Service, Work and Social Laws NDP National Development Plan OGT Opération Globale du Trésor (Monthly Treasury Operations report) PEFA Public Expenditure and Financial Accountability PFM Public Finance Management PGDI Projet de Gouvernance et de Développement Institutionnel PPP Purchasing Power Parity RCF Rapid Credit Facility SADC Southern African Development Community SCD Systematic Country Diagnostic SDR Special Drawing Rights SOE State-Owned Enterprise SORT Systematic Operations Risk-rating Tool SSA Sub-Sahara Africa TSA Treasury Single Account US$ United States Dollar VAT Value Added Tax WDI World Development Indicator Vice President: Makhtar Diop Country Director: Mark Lundell Country Manager: Coralie Gevers Global Practice Director: Paloma Anos Casero Practice Manager: Mathew Verghis Task Team Leader: Keiko Kubota ICR Primary Authors Natasha Sharma ICR Team Leader Natasha Sharma MADAGASCAR Resilience Development Policy Operation CONTENTS Data Sheet A. Basic Information i B. Key Dates i C. Ratings Summary i D. Sector and Theme Codes ii E. Bank Staff ii F. Results Framework Analysis iii G. Ratings of Program Performance in ISRs (all fields are entered by the system) vii H. Restructuring (if any) (some fields are entered by the system) vii 1. Program Context, Development Objectives and Design ................................................................... 1 2. Key Factors Affecting Implementation and Outcomes ..................................................................... 6 3. Assessment of Outcomes ................................................................................................................ 14 4. Assessment of Risk to Development Outcome ............................................................................... 24 5. Assessment of Bank Borrower Performance................................................................................... 25 6. Lessons Learned .............................................................................................................................. 27 7. Comments on Issues Raised by Borrower/Implementing Agencies/Partners ................................ 29 ANNEX 1. Bank Lending and Implementation Support/Supervision Processes ......................................... 30 ANNEX 2. Summary of Borrower’s ICR and/or Comments on Draft ICR .................................................... 32 ANNEX 3. List of Supporting Documents ................................................................................................... 33 ANNEX 4: Follow-up of the reforms in the PFSI programmatic DPO 2017-2018........................................ 35 MAP 37 A. Basic Information Resilience Development Country Madagascar Program Name Policy Operation Program ID P153084 Loan Number IDA – 57520 MG ICR Date 12/18/2017 ICR Type Core ICR Lending Instrument DPF Borrower Government of Madagascar Original Total USD 39,400,000 Disbursed Amount USD 39,400,000 Commitment Implementing Agencies: Ministry of Finance and Budget B. Key Dates Revised / Actual Process Date Process Original Date Date(s) Concept Review: 09/28/15 Effectiveness: 12/28/15 12/28/2015 Appraisal: 11/02/15 Restructuring(s): Approval: 12/10/15 Mid-term Review: Closing: 12/30/16 12/30/2016 C. Ratings Summary C.1 Performance Rating by ICR Outcomes Moderately satisfactory Risk to Development Outcome Substantial Bank Performance Moderately Satisfactory Borrower Performance Moderately Satisfactory C.2 Detailed Ratings of Bank and Borrower Performance (by ICR) Bank Ratings Borrower Ratings Quality at Entry Moderately Satisfactory Government: Moderately Satisfactory Quality of Supervision: Satisfactory Implementing Agency: Moderately Satisfactory Overall Bank Overall Borrower Moderately Satisfactory Moderately Satisfactory Performance Performance i C.3 Quality at Entry and Implementation Performance Indicators Implementation QAG Assessments Indicators Rating: Performance (if any) Potential Problem Program No Quality at Entry (QEA) None at any time (Yes/No): Problem Program at any Quality of Supervision No None time (Yes/No): (QSA) DO rating before Closing/Inactive status D. Sector and Theme Codes Original Actual Sector Code (as % of total Bank financing) Central Government (Central Agencies) 50 50 Other Public Administration 50 50 Theme Code (as % of total Bank financing) Public Finance Management 25 25 Macro-financial policies 24 24 Fiscal Policy 49 49 E. Bank Staff Positions At ICR At Approval Vice President: Mahktar Diop Mahktar Diop Country Director: Mark Lundell Mark Lundell Country Manager Coralie Gevers Coralie Gevers Practice Manager/Manager: Mathew Verghis Mark Thomas Task Team Leader: Natasha Sharma Keiko Kubota ICR Team Leader: Natasha Sharma ICR Primary Author: Natasha Sharma 4 There can be a maximum of five Sector Codes (that have more than zero percent) and five Theme Codes (of which at least one must be “Primary�). ii F.Results Framework Analysis Program Development Objective1: There are two Program Development Objectives (PDO) for the Madagascar Resilience Development Policy Operation:  PDO1: Strengthening reporting and increasing availability of information relevant to assessing the effectiveness of public finance; and  PDO2: Improving payroll management and consolidating accounts in line with the treasury single account principle. Revised Program Development Objective: Not applicable (a) PDO Indicators Table 1: PDO Indicators Baseline ValueOriginal Target Formally Revised Actual Values (from approvalValues (from Target Values Achieved documents) approval at Completion or documents) Target Years Pillar I: Strengthening reporting and increasing availability of information relevant to assessing the effectiveness of public finance Overall score: 85.2 percent (Substantially met) PDO Indicator 1 Liter of diesel per kwh of thermic generation of electricity by JIRAMA (i): (including the private centers for which JIRAMA is responsible for the fuel supply)2 Value (quantitative 0.16 L/kwh 0.14 L/kwh 0.13 L/kwh or qualitative) Date achieved 08/31/2015 03/30/2017 07/31/2017 Comments (incl. % EXCEEDED (100 percent plus). The achievement of this indicator suggests achievement) there has been an improvement in the efficiency of JIRAMA’s operations, as (i) the use of diesel is being increasingly replaced by fuel oil, which is less expensive; and / or (ii) the use of diesel is more efficient. This target was achieved and exceeded. 1 The cost increased by 75 percent instead of declining by 12.5 percent, signaling The assessment scale used to assess progress against the results indicators is as follows: “Not met�: <50% completion vs. a reduction in the efficiency of thermal energy generation. This follows delays target, “Partly met�: >50% but < 75%, “Substantially met�: approximately 75%+, “Fully met�: approximately 100%; “Exceeded�: >100%. In to investment, case a percentage maintenance and installation rating is available of fuel based through the assessment generators. of the However, results target, the specific rating is used instead. this outcome indicator is arguably a step too far removed from the reform 2 supported This indicator is measured in consumption as: diesel this DPO, which is focused / (thermic on generation byimproving transparency diesel + thermic generationof bythe fuel oil). A the expected reduction of the ratio (tocosts 0.14) would of energy generation.mean improved efficiency in the use of diesel or increase in the use of fuel oil. iii PDO Indicator 1 Data on diesel consumption and electricity generation publicly available (ii): Value (quantitative No information is The database is Monthly data is or qualitative) available updated monthly available Date achieved 2014 03/30/2017 10/25/2017 Comments (incl. % FULLY MET (100 percent). Monthly statistics on diesel consumption and achievement) electricity generation are publicly available, which has helped to facilitate regular monitoring of the use of diesel by the Government and civil society. This includes the publication of civil society reports on JIRAMA’s efficiency and transparency, which has been used for wider public debate. Combined average score for PDO Indicator 1 = 100 percent PDO Indicator 2: The audited accounts of Air Madagascar are published within the time allotted by regulations Value (quantitative Publications are Publication is on Publication is on or qualitative) overdue time (in time compliance with the Law) Date achieved 2014 03/30/2017 11/16/2017 Comments (incl. % SUBSTANTIALLY MET (85 percent). The audited accounts of Air achievement) Madagascar are available on the website for the period 2010 to 2016. SOEs are subject to common commercial legislation (Loi 2003 -036 sur les sociétés commerciales), which states that audited financial statements should be produced no later than 6 months after the closing of the FY. While there is no legal requirement for the publication of SOE financial statements, this increased transparency measure was supported in the Resilience DPO. Since all available financial statements are published, but with a delay of five months for the publication of the 2016 financial statement, this target is considered substantially met. PDO Indicator 3: List of enterprises which receive tax expenditures through Law on Large Mining Investments (LMIL), Export Processing Zones (EPZ) Law and by Council of Ministers decisions publicly available. Value (quantitative No systematic The three lists, Two of the three or qualitative) information is updated annually, lists are published available are available on EDBM website publicly Date achieved 2014 03/30/2017 10/25/2017 Comments (incl. % PARTIALLY MET (66 percent). The EDBM publishes on its websites the achievement) list of companies that receive tax expenditures under the EPZ and the LMIL. The third list on the decision by the Council of Ministers is managed by the Prime Minister’s Office, and this has not been shared for publication. PDO Indicator 4: Length of time needed to submit the audited consolidated final accounts of the State Budget. iv Value (quantitative Over 5 years (for Within 24 months Within 24 months or qualitative) 2008 accounts) of end year of end year Date achieved 2014 03/30/2017 10/25/2017 Comments (incl. % FULLY MET (100 percent). The 2015 audited consolidated account was achievement) submitted to Parliament June 2017. Madagascar’s fiscal year runs from January to December. Therefore, the accounts were submitted within an 18-month period, addressing a significant weakness in financial management practices where the submission of accounts was historically delayed even before the political crisis period. PDO Indicator 5: Timeliness of the publication of the monthly budget execution report by the Treasury (OGT) Value (quantitative OGT is not Within 6 weeks of Within 8 weeks of or qualitative) publicly available end month end month Date achieved 2014 03/30/2017 10/25/2017 Comments (incl. % SUBSTANTIALLY MET (75 percent). The OGT is habitually published two achievement) months after the end of period, as it takes time to verify the information from other. The score of 75 percent was calculated as the ratio of the intended target (6 weeks) over the actual target achieved (8 weeks). This marks significant progress compared with the political crisis period, where the publication of the OGT ceased. Pillar II: Improving payroll management and consolidating accounts in line with the treasury single account principle Overall score: 74.7 percent (Partly met) PDO Indicator 6 The number of civil servants with date of birth information not on record at the (i): MFB Value (quantitative 30,000 civil 0 4,297 or qualitative) servants (approx. 25% of total) (2015); Date achieved 10/19/2015 03/30/2017 05/30/2017 Comments (incl. % SUBSTANTIALLY MET. 86 percent of civil servants’ date of birth were achievement) recorded by October 2017. PDO Indicator 6 The civil service rosters at MFB and MCS are reconciled every 6 months (ii): Value (quantitative Reconciliation has Reconciliation was No reconciliation or qualitative) not been done done within the last has been done (2014) 6 months Date achieved 12/31/2014 03/30/2017 05/30/2017 v Comments (incl. % NOT MET (0 percent). The Ministry of Civil Service (MCS) did not update achievement) the roster. Instead, payroll controls were pursued in another way, through the installation of a new software for human resource management (Application Unique pour la Gestion Uniforme des Ressources Humaines de l’Etat - AUGURE) at the MFB, where all agencies are required to self-declare staff details. An audit of the software and the self-declaration process is planned in 2018, which will facilitate an objective assessment of the effectiveness of the new approach. PDO Indicator 6 Amount of position-specific benefits paid to those who had already left the (iii): “high officer� positions Value (quantitative Approximately 0 0 or qualitative) 3.56 bn Ariary (2015) Date achieved 09/30/2015 03/30/2017 10/25/2017 Comments (incl. % MET (100 percent). In 2016 1,480 irregular payments were suspended. The achievement) recently installed software will ensure the sustainability of payment controls. A combination of reforms to update civil servants’ date of birth information, and cancelling irregular payments is estimated to have resulted in savings of US$10.5m (approximately 0.11 percent of GDP) by February 2017. Combined average score for PDO Indicator 6 = 62 percent PDO Indicator 7: The share of dedicated accounts for externally-financed projects held at the Central Bank Value (quantitative 0 % 90% of the 87.4% of project or qualitative) accounts, accounts are with explicit opened at the deadline for Central Bank closure of any accounts remaining at commercial banks Date achieved 2014 03/30/2017 10/25/2017 Comments (incl. % SUBSTANTIALLY MET (87.4 percent). Reforms to the TSA are ongoing, achievement) but the pace has been more moderate than expected. To facilitate transition to the TSA, a temporary decree was issued in August 2016, allowing for some project accounts to be held at commercial banks, as TSA reforms continued. Total average across all the PDO Indicators = 82.2 percent (substantially met) vi (b) Intermediate Outcome Indicator(s) - from Program Document Not applicable: There were no intermediate outcome indicators for the standalone DPO. G. Ratings of Program Performance in ISRs (all fields are entered by the system) No. Date ISR DO IP Actual Disbursements Archived (US$mil.) No formal ISRs were prepared, as this was not requirement for a standalone operation. However, during implementation regular monitoring and supervision activities were undertaken. The Government informally shared an ISR mid-way through implementation, which informed the preparation of this ICR. H. Restructuring (if any) (some fields are entered by the system) Not applicable vii 1. Program Context, Development Objectives and Design 1.1.Context at Appraisal: 1. Madagascar is an island nation, which is rich in assets and blessed with unparalleled natural wealth. Madagascar has a population of around 24 million, where 64 percent is under 25 years, and levels of literacy are comparatively high. Unparalleled biodiversity, including large swaths of agricultural lands, forest areas and a pristine coastline could provide sources of sustainable income and consumption for the Indian Ocean and beyond. Manufacturing could be bolstered by relatively strong capacity for producing garments, as well as mining-related activities for nickel and other minerals. While the scope of the private sector is small, it is diversified and has considerable potential to further thrive, particularly in areas such as construction, tourism, services, and transport. 2. However, the country’s ability to harness its growth potential has been repeatedly stalled by political fragility. Since independence in 1960, there have been four major political crisis periods, in 1972, 1991, 2001-2002, and 2009-2013. In each case, the political crisis followed a period of economic growth, and resulted in a contraction of the economy. During times of political stability, growth has been driven by the secondary and tertiary sectors, while the agricultural sector is characterized by low productivity and vulnerability to climatic variations. The end of the long political crisis between 2009 to 2013 has promoted a modest acceleration of economic growth, which is estimated to have averaged 3.7 percent between 2014 and 2017, with a growth projection of 4.1 percent in 2017. During the most recent political crisis, a coup d’état in January 2009 resulted in a change of Government. 3. Repeated political crises are the result of deep-rooted governance challenges, which include:3 (i) a political system that is determined by networks of a few powerful political and business leaders who form and shift alliances periodically to preserve their access to rents; (ii) a society fragmented along ethnic, religious, wealth, and gender fault lines, giving rise to social tensions; (iii) the abundance of natural resources, coupled with weak governance, has contributed to the rapid growth of a trafficking economy around precious woods, gold and stones, cattle, and valuable biodiversity; and (iv) a nascent system of checks and balances because of a fragmented parliamentary system, weakness in civil society to become an organized voice, and the lack of an independent media. 4. The costs of political fragility have been high, where Madagascar is one of the poorest countries in the world.4 Over the 2001 to 2012 period, the poverty headcount rate decreased slightly but it still remains exceedingly high at 70.7 percent in 2012. The average Malagasy is 3 Further details on the sources of Madagascar’s political fragility can be found in the Country Partnership Framework, June 2017 4 World Bank (2016). “Recent Trends and Analytical Findings on the Causes of Madagascar’s Persistent Poverty.� It uses household survey data from EPM 2001, 2005, 2010, and ENSOMD 2012. Also see World Bank (2014). “Face of Poverty in Madagascar: Poverty, Gender, and Inequality Assessment.� The next household survey is expected to take place in 2018, following the census. 1 42 percent poorer today than she was in 19605, the year of Madagascar’s independence. In 20126, only 30 percent of Malagasy lived above the national poverty line and only 10 percent above the international poverty line. Inequality is defined not by an excessive concentration of wealth for a few but by deep poverty, where the average Malagasy consumes 46 percent less than a person living right at the national poverty line. 5. The return to constitutional order in early 2014 opened the door for expanded WBG engagement to pursue the goal of ending poverty. The Southern African Development Community helped to agree a Roadmap to end the political crisis, and strengthen democratic and security institutions. The Government has elaborated a National Development Plan 2015- 2019 (NDP) that aims to improve economic management and promote private-sector led growth. With support from the international community, municipal and senatorial elections were peacefully held in July and December 2015, the Independent National Commission for Elections was established, and security reforms are ongoing. Presidential elections are planned for the end of 2018 / early 2019. While elections are expected to be tense, it is hoped that this will be the first time since independence that a Presidential term starts and ends through democratic means. 6. The Systematic Country Diagnostic (SCD, 2015) argues for supporting fiscal policy reforms, which had been subject to poor governance practices during the 2009-2013 political crisis. The prolonged political crisis period gave way to deep-rooted governance practices, constraining fiscal space for financing public investments and social spending necessary for fighting poverty and pursuing sustainable development. As Madagascar exited the political crisis in 2014, economic growth was at 3.3 percent, barely high enough to keep up with population growth (see Table 1). 7. On the revenue side, Madagascar had one of the lowest tax-to-GDP ratios in the world, at 10.1 percent. While tax policy is generally sound7, low levels of revenue collection reflected the subdued economic activity during the political crisis period (which had deterred private sector investment) and poor tax administration practices. For example, ad hoc agreements that award generous tax exemptions to well-connected individuals and enterprises were awarded, undermining the credibility of the relatively sound tax policies. 8. Fiscal space had been further constrained by inefficient public expenditure management practices. The wage bill had been subject to weak fiduciary controls during the political crisis period, with ineligible recipients receiving payments, including at high levels of government. Two large state-owned enterprises (SOEs), JIRAMA (a water and electricity SOE) and Air Madagascar (the state’s national airline company) were characterized by inefficient financial and operational performance, reliant on large state transfers that were regressive in nature. The ability for the state to make payments, and avoid the accumulation of arrears was further constrained by poor cash management practices, including maintaining 5 Measured in real GDP per capita. Only DRC and Liberia fared worse than Madagascar over the 1960-2010 period. 6 Most recent data available drawn from the ENSOMD 2012 survey. 7 Selected Issues: Tax Revenue Mobilization in Madagascar, International Monetary Fund, 2014 2 Table 2: Madagascar Selected Economic and Financial Indicators accounts outside of the Treasury Single Account (TSA) mechanism. Furthermore, the ability to make effective decisions on public expenditure management, report to elected officials and share information with civil society was constrained by the lack of public accounts prepared and submitted to Parliament. 9. Against this backdrop, the Resilience DPO was prepared to address important fiscal policy reforms, on both the revenues and expenditures side. On the revenue side, the Resilience DPO aimed to improve transparency in how tax expenditures are awarded. On the expenditure side, the DPO supported several important reforms including advancing payroll controls, improving the transparency of operations for JIRAMA and Air Madagascar, ensuring the availability of public expenditure information for scrutiny by Parliament, civil society and development partners, and furthering TSA reforms. 10. At the time of appraisal, the macro framework for the operation was assessed as adequate. The Resilience DPO was important for supporting the Government in defining and executing a credible set of fiscal policies to serve as a basis for: (i) satisfactory performance under the IMF Staff Monitored Program; and (ii) continued policy-based support needed to close the financing gap. 11. The measures supported in the Resilience DPO were part of an integrated approach by the international community to improving public finances. The program aimed to reverse some of the long-standing poor financial management practices, which further deteriorated during the political crisis, as well as support further reforms, such as the TSA. The content of the program was complemented by reforms supported by other partners. For example, the IMF was supporting other aspects of fiscal policy, such as the recording of debt flows and improving the governance of EPZs; and the EU was supporting the management of human resources related reforms. 3 2013 2014 2015 2016 2017 2018 Estimate Projections (annual percent change, except noted otherwise) Real sector Real GDP growth 2.3 3.3 3.1 4.2 4.1 5.1 GDP per capita growth -0.5 0.5 0.3 1.4 1.3 2.3 Consumption (% GDP) 92.8 88.8 90.4 87.0 91.3 89.6 Investment (% GDP) 15.9 15.6 13.1 15.3 15.9 17.4 Net exports (% GDP) -8.7 -4.4 -3.5 -2.3 -7.2 -7.0 Inflation, consumer prices (annual %, end of year) 6.3 6.0 7.6 7.0 8.0 7.9 (in percent of GDP) Public Finance Revenues, excluding Grants 9.6 10.1 10.4 11.2 11.7 12.1 of which: Tax revenues 9.3 9.9 10.1 10.9 11.4 11.9 Grants 1.3 2.3 1.5 3.5 3.2 3.0 Total expenditures (commitment basis) 14.9 14.7 15.1 16.0 18.4 18.0 of which: Capital spending 3.1 3.9 3.5 5.2 6.3 8.1 Overall balance (cash basis) -2.0 -2.4 -3.7 -2.0 -4.3 -3.5 Gross external financing 1.3 1.7 2.6 1.4 3.0 3.6 Total public debt 33.9 34.7 41.3 38.4 36.2 36.2 (annual change) Monetary accounts Money supply (M2) 9.0 9.8 15.8 21.4 18.5 13.1 Net Foreign Assets -29.1 17.3 20.9 37.4 10.9 12.9 Net Domestic Assets 34.9 8.3 11.4 10.8 22.9 12.2 (in millions of US$) External Sector Exports of goods 1922 2194 2047 2165 2347 2456 Imports of goods 2773 2741 2376 2429 3143 3256 Current account balance -622 -34 -184 57 -388 -494 Foreign Direct Investment 551 312 416 451 411 411 Overall Balance -214 22 100 293 92 76 Foreign Reserves (months of imports) 2.2 2.5 2.9 3.9 3.7 4.0 Source: MFB, CBM, IMF and World Bank staff calculations 1.2 Original Program Development Objectives (PDO) and Key Indicators(as approved): 12. The Program Development Objectives for the Madagascar Resilience Development Policy Operation were: (i) Strengthening reporting and increasing availability of information relevant to 4 assessing the effectiveness of public finance, and (ii) Improving payroll management and consolidating accounts in line with the treasury single account principle. The original Key Indicators, used to measure progress towards expected outcomes were: Table 3: Summary of Key Indicators Pillar I: Strengthening reporting and increasing availability of information relevant to assessing the effectiveness of public finance PDO Indicator 1 (i): Liter of diesel per kwh of thermic generation of electricity by JIRAMA (including the private centers for which JIRAMA is responsible for the fuel supply) PDO Indicator 1 (ii): Data on diesel consumption and electricity generation publicly available PDO Indicator 2 The audited accounts of Air Madagascar are published within the time allotted by regulations PDO Indicator 3 List of enterprises which receive tax expenditures through LMIL, EPZ Law and by Council of Ministers decisions publicly available PDO Indicator 4 Length of time needed to submit the audited consolidated final accounts of the State Budget PDO Indicator 5 Timeliness of the publication of the monthly budget execution report by the Treasury (OGT) Pillar II: Improving payroll management and consolidating accounts in line with the treasury single account principle PDO Indicator 6 (i): The number of civil servants with date of birth information not on record at the MFB PDO Indicator 6 (ii): The civil service rosters at MFB and MCS are reconciled every 6 months PDO Indicator 6 (iii): Amount of position-specific benefits paid to those who had already left the “high officer� positions PDO Indicator 7 The share of dedicated accounts for externally-financed projects held at the Central Bank 1.3 Revised PDO and Key Indicators, and reasons/justification: Neither the PDO nor the Key Indicators were revised. 1.4 Original Policy Areas Supported by the Program: 5 13. This stand-alone DPF operation supported the reform program under two core Policy Areas: (i) Strengthening reporting and increasing availability of information relevant to assessing the effectiveness of public finance. Reforms to improve the transparency of public finance were supported as a precondition for informed decision-making to improve public service delivery. The reforms aimed at reinstituting certain public expenditure management practices such as the publication of monthly budget execution reports, and the timely submission of audited consolidated final accounts to Parliament. These practices were weak even before the political crisis, and sidelined thereafter, as poor governance procedures prevailed. The reforms also supported improvements to the transparency of JIRAMA and Air Madagascar’s operations, as well as tax expenditures. (ii) Improving payroll management and consolidating accounts in line with the treasury single account principle. The reforms supported under this objective aimed to take preliminary steps to increase fiscal space by cleaning up the payroll through removing ineligible beneficiaries from the roster, and promoting improved public expenditure management practices by passing a decree to facilitate the transition to a TSA. 1.5 Revised Policy Areas: 14. Not applicable: The policy areas remained unchanged throughout the program. 1.6 Other significant changes (in design, scope and scale, implementation arrangements and schedule, and funding allocations): 15. Not applicable: No changes were made to the design, scope and scale, implementation arrangements and funding allocations. 2. Key Factors Affecting Implementation and Outcomes 2.1 Program Performance: 16. The Program was supported by a stand-alone, single-tranche Development Policy Financing, disbursed upon effectiveness, in the total amount of US$55 million. All required Prior Actions were completed prior to the operation’s approval. 2.2. Major Factors Affecting Implementation: 17. While all prior actions were achieved, the overall implementation for this DPO was moderately satisfactory. Major factors that affected implementation are summarized below: (i) Uneven stakeholder engagement throughout the preparation, implementation and supervision of the DPO. While the MFB was the main implementing agency for the reforms supported under the DPO, several other stakeholders were involved to a lesser extent, including the EDBM, the BCM, the Office of the Prime Minister, the 6 Ministry of Public Service, JIRAMA, Air Madagascar and development partners (concerning the TSA reform). Where the MFB has been responsible for reform implementation, progress has been largely on-track. However, where other agencies are concerned, such as EDBM and MCS, progress has been uneven, suggesting different stakeholders should have been more actively consulted and engaged throughout the DPO, at an earlier stage. (ii) There was good alignment with other development partners regarding the content of the reform program. The reforms supported in the DPO complemented well the actions of other partners, reflecting convergence of views on the criticality 8 of the policy changes required. For example, under the IMF’s Rapid Credit Facility and the Staff Monitored Program complementary reforms were supported such as improving transparency of the stocks and flows of public and publicly guaranteed debt (building on the progress made to the debt management framework, supported under the Reengagement DPO in 2014), mobilizing fiscal revenues from companies located in EPZs, and establishing audit teams to work on ex-post highly suspect fraud operations. The EU and the African Development Bank supported civil service reforms, while the US Treasury supported TSA reforms through Technical Assistance (TA) in the form of a Resident Advisor. Discussions around the criticality of reforms were framed under Cadre de Partenariat, which is co-presided by the WB and the MFB. (iii) Good progress, where reforms are integrated with Investment Projects and supported with technical assistance (TA). Complementary reforms related to Air Madagascar and JIRAMA were supported by the Integrated Growth Poles Project (P113971) and the Energy Sector Operations and Governance Improvement project (P151785) respectively, which were being prepared in parallel with the DPO. This complementarity helped to reinforce the policy dialogue. (iv) However, in some instances the reforms supported by other partners were pursued in an alternative way, or with a different timeline, which affected the implementation of the Resilience DPO. These divergences in approach also highlight that there are risks when implementation is being supported by parallel projects and / or other partners. Some examples are provided below:  The Program Document makes reference to EU support to payroll reforms when explaining the content of the reform program. However, the reform was described at a relatively early stage of development, and since then the EU’s support evolved and adopted a different approach.  In the case of the publication of the OGT, the Government had agreed to an 8- week timeframe for publishing the information, compared to a 6-week timeframe with the WB. This suggests that there should have been greater harmonization between the institutions regarding the periodicity of reporting the OGT. 8 Criticality refers to the need to focus on the key elements to support the transition from fragility to resilience (as discussed in the Good Practice Note on DPL in Situations of Fragility, World Bank, 2011 7  The US Treasury Resident Advisor for TSA reforms was in place only in March 2016, which has affected the pace of reform implementation. While the Program Document does not make reference to this TA, the speed at which the Government was able to find appropriate TA has affected implementation. (v) Soundness of background analysis. The DPO was prepared in parallel with the SCD, with the same teams involved. The analysis is well reflected in the choice of reforms supported. The DPO was also supported by the Reengagement Policy Notes prepared in 2014 (which had recommendations related to improving JIRAMA’s performance, implementing civil service controls, and strengthening tax administration), a Tax Administration Diagnostic Assessment Tool which was under preparation and published in October 2015, an analysis of tax revenue mobilization undertaken by the IMF in January 2015, a Public Expenditure and Financial Accountability (PEFA) assessment, a Debt Management Performance Assessment (DeMPA), and the reforms in the Country Policy and Institutional Assessment (CPIA). The reforms had a strong governance component, which was appropriate given that activities were normalizing after the political crisis. The importance of addressing governance challenges was underscored in the SCD, as prerequisite for realizing other reforms. (vi) For a country emerging from a deep political crisis, overall there has been strong commitment and ownership for the reform agenda. As the country reengaged with the international community following the return to constitutional order, there has been increasing reform momentum. The reforms supported by this DPO are well-aligned with the priorities the Government had identified as requiring institutional reforms. The gradual but important progress is demonstrated in different ways:  Madagascar qualified for exceptional financing under the IDA Turnaround Regime for countries exiting a period of fragility or conflict;  Following successful completion of a Staff Monitored Program and Rapid Credit Facilities, a three-year Extended Credit Facility was approved by the IMF’s Executive Board of Directors in 2016, setting a medium -term basis for structural reforms;  Improvements have been made to the institutional and policy environment, as highlighted by a gradual but important improvement to t he country’s Country Policy and Institutional Assessment score, which has increased from 3.0 in 2014 to 3.2 in 2016, underpinned by improvements to budgetary management and the social protection framework;  In recognition of the progress above, in December 2016, the international community increased its support for Madagascar at a Donors and Investors Conference, where US$6.4 billion was pledged for the 2017-2020 period, including US$2.1 billion of undisbursed commitments and US$4.3 billion in new commitments. This represents more than a doubling of overseas development assistance per capita, from an average of US$28.60 over the period 2007 to 2015 to an estimated US$66.60 on average over the period 2017-2021. 8 2.3. Monitoring and Evaluation (M&E) Design, Implementation and Utilization: (a) Design 18. Overall, the PDOs are well-articulated, although improvements could be made to the formulation of PDO Objective 1: “Strengthening reporting and increasing availability of information relevant to assessing the effectiveness of public finance,� could have been strengthened to also include the purpose of enhancing transparency, which is to improve the accountability and effectiveness of public finance. This would have been a stronger articulation of the PDO and a better reflection of the reforms supported in the DPO, which are in the vein of improvements in PFM rather than just an assessment of PFM effectiveness. Furthermore, the rationale for the PDO could have provided further information on how decision makers, parliamentarians, civil society, and development partners are expected to use the available information. 19. In contrast, PDO 2 is concise and clear: “Improving payroll management and consolidating accounts in line with the treasury single account principle.� The selection of the prior actions clearly support the attainment of this PDO. 20. The design of the Resilience DPO built on certain public finance reforms supported in the Reengagement DPO in 2014. In December 18, 2014, the Reengagement DPO was approved, which had the PDO “Improving the Efficiency and Transparency of Public Service Delivery.� The reforms supported in the Resilience DPO continued the public finance reforms in the Reengagement DPO, in the areas such as JIRAMA’s operational management, and the public debt law, which enabled the continuation of reforms supported through two stand-alone operations, as presented in Table 4. Other reforms in the Reengagement DPO relate to the adoption of a supplemental budget to increase pro-poor expenditures and the clearance of arrears, which were relevant to improving the effectiveness of public finance, as supported in the Resilience DPO. Table 4: Link between the Reengagement DPO (2014) and the Resilience DPO (2015) Reform in the Follow-up Reform in the Comment Reengagement DPO Resilience DPO The Recipient has instituted JIRAMA has started to Continued measures to a system of coloring diesel publish monthly statistics on improve JIRAMA’s delivered to JIRAMA with the delivery of diesel to, and operational efficiency the aim of reducing leakages generation of electricity at, as a first step towards major diesel-based electricity reducing operating losses of generation centers. the company 9 The Recipient’s National The Recipient has issued an Resilience DPO supported an Assembly has adopted a law implementing decree for the implementing decree for the on public debt which law on public debt public debt law supported in provides a legal framework concerning opening the Reengagement DPO to ensure public borrowing is dedicated accounts for conducted strategically and externally-financed projects prudently and stipulates for at the Central Bank of the preparation of a medium Madagascar. term debt strategy 21. However, support for certain reforms in the Reengagement DPO were not continued in the Resilience DPO, for which further explanation would have been useful. The reforms in the Reengagement DPO related to anti-corruption measures and the weakening of governance during the political crisis period, such as asset declaration by Cabinet members, and the management of the seized stockpile of precious woods were not continued in the DPO series, although they are being supported by other WB engagements.9 Nevertheless, it would have been useful to explain the reason for not continuing to support these reforms in the Resilience DPO. (b) Implementation 22. Monitoring was carried out through continuous dialogue with the Government, specifically with the dedicated Economic Council established by Decree number 34316/2015 in November 2015. The Economic Council is a high level and multi sector group, representing different ministries and agencies that are responsible for achieving development objectives in line with the National Development Plan. Members of the Economic Council include the Governor of the Central Bank (relevant for TSA reforms), a Presidential Advisor responsible for following up externally financed public investments, and the Secretary General of the MFB. The Economic Council has a direct reporting line to the President and has the authority to convene other implementing agencies to discuss reforms with a wider group of sectors. 23. Mid-way through implementation, the Government shared an informal information status report, which while not formally required, demonstrated their commitment to using their monitoring and evaluation processes to collect relevant data. The Government was able to demonstrate evidence and the sources of information. Of the ten indicators included in the PD, for six the information is publicly available, in line with the program development objective and prior actions that seek to improve transparency. For the other four indicators, further information is needed from the Government. A summary of the availability of information is presented below (Table 5). 9 The World Bank has been encouraging and supporting the authorities in their efforts to establish a special tribunal for the traffic of precious woods, to finalize an audit of precious woods stocks, and to develop transparent ways for possibly liquidating those stocks with the agreement of Convention on International Trade in Endangered Species of Wild Fauna and Flora (CITES). 10 Table 5: Availability of Information to Monitor Completion of Reforms Indicator Available publicly Pillar I: Strengthening reporting and increasing availability of information relevant to assessing the effectiveness of public finance PDO Indicator 1 (i): Liter of diesel Yes per kwh of thermic generation of http://www.jirama.mg/index.php?w=scripts&f=Jirama- electricity by JIRAMA (including page.php&act=statgo the private centers for which JIRAMA is responsible for the fuel Source: JIRAMA website supply); PDO Indicator 1 (ii): Data on Yes diesel consumption and electricity http://www.jirama.mg/index.php?w=scripts&f=Jirama- generation publicly available page.php&act=statgo Source: JIRAMA website PDO Indicator 2: The audited Yes accounts of Air Madagascar are http://www.airmadagascar.com/fr/compagnie- published within the time allotted by aerienne/organigramme-chiffres-clefs regulations Source: Air Madagascar website PDO Indicator 3: List of Yes enterprises which receive tax http://www.edbm.gov.mg/Actualites/Actualites/Liste-des- expenditures through LMIL, EPZ Zones-et-Entreprises-Franches-inventoriees-au-04-aout-2017 Law and by Council of Ministers decisions publicly available. http://www.edbm.gov.mg/Actualites/Actualites/Liste-des- societes-gerees-actuellement-a-la-DGE-beneficiant-du-regime- LGIM Source: EDBM website PDO Indicator 4: Length of time Yes – can be determined by monitoring the frequency of needed to submit the audited information published consolidated final accounts of the http://www.mefb.gov.mg/index.php?option=com_content&vie State Budget. w=article&id=647&Itemid=810 Source: MFB website Pillar II: Improving payroll management and consolidating accounts in line with the treasury single account principle PDO Indicator 5: Timeliness of the Yes publication of the monthly budget http://www.tresorpublic.mg/?page_id=214&content=temp&typ execution report by the Treasury e=ogt (OGT) Source: MFB website PDO Indicator 6 (i): The number No, although information on progress is regularly of civil servants with date of birth communicated through press releases. information not on record at the http://www.dggfpe.mg/index.php/2016/02/10/solde-les- MFB operations-dassainissement-se-poursuivent/ http://www.newsmada.com/2017/02/04/gestion-des-soldes-et- pension-le-controle-des-etats-de-paie-lance/ 11 http://matv.mg/assainissement-des-soldes-et-des-pensions-une- economie-de-7-milliards-ariary-pour-2016/ PDO Indicator 6 (ii): The civil No, but the alternative approach to implementing this reform service rosters at MFB and MCS are through the reconciliation of data held at MFB and the self- reconciled every 6 months declared human resource data in the recently implemented software will be published. PDO Indicator 6 (iii): Amount of No position-specific benefits paid to those who had already left the “high officer� positions PDO Indicator 7: The share of No dedicated accounts for externally- financed projects held at the Central Bank 24. Nevertheless, the periodicity of the reports shared by the Government could have been improved. The Resilience DPO was prepared one year after the country emerged from a deep political crisis, and the Government’s team was still learning procedures and reporting requirements associated with DPOs. The change in focal point beyond the MFB meant that a new team had to become acquainted with the processes required for the DPO, where greater emphasis was placed on identifying reforms for new operations. As a result, while the Government has prepared monitoring reports for the WB, the periodicity of the reports, and the necessary follow-up to ensure that reforms remain on track could have been strengthened. 25. Implementation monitoring was enhanced with dialogue from complementary investment projects, analytical and TA activities, as well as activities from other partners, as summarized below: Table 6: Implementation Activities Supporting the Achievement of the Reforms Prior Action Supporting Implementation Activities JIRAMA has started to publish monthly The World Bank Project Energy Sector statistics on the delivery of diesel to, and Operations and Governance Improvement generation of electricity at, major diesel based project (P103950) has a major governance electricity generation centers component, seeking to improve the operational / financial performance of JIRAMA’s performance. Air Madagascar has (i) appointed a new general As part of the restructuring of Air manager through a competitive process, and (ii) Madagascar, which includes searching for published its audited financial accounts for the a strategic partner, supported by the World 2010 to the 2013 calendar year on its website Bank’s Integrated Growth Poles Project (P113971) there has been analysis of the financial situation. The follow-up dialogue on the publication of audited financial statements was 12 supported by an analytical piece on SOEs in Madagascar, led by the World Bank’s Governance team in FY17. The Recipient has published, through the The dialogue on the reform on increasing Economic Development Board of Madagascar, transparency on tax benefits has been the list of enterprises that are granted benefits continued through the publication of a tax under the Free Zone regime (Zone franche). expenditures statement, supported by the World Bank’s Governance team, as part of the Public-Sector Performance Project The Recipient has submitted the audited final Complementary reforms supported by the accounts of the State Budget for the fiscal years IMF, such as the timely submission of the 2009, 2010 and 2011 to the Parliament for state budget approval Starting May 2015, the Ministry of Finance and Complementary reforms supported by the Budget is publishing the monthly in-year budget IMF execution results in no later than 60 days of month end The Ministry of Finance and Budget of the A European Union project is supporting Recipient has eliminated ineligible persons from civil service reforms and payroll controls its payroll roster of high office holders through the NFD project (Ny Fanjakana ho an'ny Daholobe – one administration at the service of all). Following the installation of the AUGURE database, the next step is for the Government to decide on the process for the co-management of civil service controls between the MFB and the MCS. The Recipient has issued an implementing A US Treasury advisor is supporting cash decree for the law on public debt concerning management reforms, including the opening dedicated accounts for externally transition to a treasury single account financed projects at the Central Bank of Madagascar (c) Utilization 26. The DPO promoted and supported the use of Government M&E arrangements. The Government undertook monitoring of the program, and shared information with the WB’s team upon request. Furthermore, through the Cadre de Partenariat, the Government shares information on a quarterly basis on the implementation of high level reforms (which were initially supported through the Resilience DPO), such as the improved governance of Air Madagascar, civil service reforms, as well as progress in realizing public finance reforms that aim to enhance fiscal space. 2.4 Expected Next Phase/Follow-up Operation: 27. Efforts to deepen and strengthen the reforms supported under the Resilience DPO led to the preparation of a follow-up programmatic DPO, Public Finance Sustainability 13 and Investment (PFSI) DPO1 and DPO2, implemented in 2016 and 2017. This programmatic DPO presented a continuation of many of the reforms supported under the Resilience DPO, as envisaged in the program design, which sought to create the building blocks for further reforms. Specifically, the follow-up programmatic PFSI continued supporting reforms related to strengthening governance of JIRAMA, tax expenditures management and an extension of payroll-related reforms. The agreement of a three-year Extended Credit Facility program with the IMF in August 2016 also helped to set the pathway for a programmatic DPO series. The continuation of reforms is summarized in Annex 4. 28. Selected reforms were followed-up outside of the follow-on PFSI DPO, such as those related to civil service reforms, Air Madagascar and the TSA, as presented in Table 6. However, the reforms related to the timely publication of the State Budget and the OGT have been fully accomplished. Complementary public finance reforms as outlined in the PFM Strategy are being supported by the WB and other partners. 3. Assessment of Outcomes 3.1 Relevance of Objectives, Design and Implementation (to current country and global priorities, and Bank assistance strategy): Overall rating: moderately satisfactory (a) Relevance of Objectives: substantial 29. The overall Program Development Objectives remain highly relevant to Madagascar and is fully aligned with the Government’s program, as described below:  The DPO directly supports the Government’s National Development Program, most notably through Pillar I on governance and Pillar II on macroeconomic stability preservation. Specifically, the macroeconomic stability pillar has the objective of improving economic efficacy and transparent and efficient public finance. These two pillars were selected since the ability of the Government to deliver pro-poor policies depends on having adequate fiscal space, for which improved transparency and better public financial management are prerequisites.  The more recently approved Economic Development Paper (EDP – April 2017) further underscores that the objectives continue to remain relevant. The EDP is as strategic document that updates the National Development Plan by documenting the achievements made, and specifying major priorities going forward. The EDP emphasizes the need to enhance fiscal space for spending on development priorities. While revenue collection has been improving in Madagascar, from 10.4 percent of GDP when the DPO was prepared in 2015 to an estimated 11.2 percent in 2017, the tax-to-GDP ratio still falls far behind the average for Sub-Saharan African countries, which is 16.8 percent in 2017. Low levels of revenue collection have constrained the possibility to spend on Madagascar’s development priorities, including social sectors and public investments. 14 30. The Resilience DPO was also fully supportive of the SCD, and the subsequently discussed Country Partnership Framework (CPF) 2017 – 2021. The SCD underscores the importance of “establishing and reinstituting budget oversight mechanisms to create greater accountability and better performance of public expenditure.� The DPO also supports the objectives of the CPF, which was discussed by the Board in June 2017, underscoring the continued importance of the challenges and the reform measures. The objectives of the Resilience DPO were well-aligned with the SCD, as well as the CPF. Specifically, the objectives are most aligned with (i) objective 4 of the CPF, which is to enhance transparency and accountability, where specific reference is made to improving the availability of information to support strategic and informed decision-making; and (ii) objective 5 of the CPF, which is to increase fiscal space to finance priority social and infrastructure spending, where reference is made to transforming PFM practices. Furthermore, the relevance of the objectives is reflected in the fact that the subsequent programmatic PFSI DPO strengthened the public expenditure management reforms supported under the Resilience DPO. (b) Relevance of Design: modest 31. The relevance of the selection of prior actions is clear and justified. The prior actions are supportive of the Program Development Objectives, and are well-aligned with the SCD and the CPF. Since Government procedures were still normalizing after the political crisis period, the rationale for having a fewer number of pertinent prior actions is justified. However, further explanations could have been provided on why certain reforms were not continued from the Reengagement DPO, such as the asset declaration measures. 32. In general, the prior actions are well-aligned with the end of program target, but there are some moderate shortcomings. The prior actions related to improving transparency are well-aligned with the end of program targets on improving the timeliness of publication. However, the end of program target on improving the efficiency of JIRAMA’s thermic generation of electricity was too far removed from the prior action which is on enhanced transparency. Improvements to the efficiency of electric thermic generation is largely dependent on the type of contracts that JIRAMA secures with Independent Power Producers, which goes beyond the action supported in the Resilience DPO; and whether the infrastructure in place is being properly maintained. In the case of the payroll reforms, the relevance of a results target is questionable, since it measures the reconciliation of the MCS civil servants roster with the MFB payroll roster, when there was no financing (and therefore demonstrable commitment) for the MCS roster. 33. The specificity of certain results targets could have been improved, to facilitate adequate monitoring by both the Bank and the Client. For example, further information could have been provided on the timeframe for publishing Air Madagascar’s financial statements. And the rationale for having a different target for publication of the OGT, compared with the IMF, could have been explained. A summary of all observations related to the relevance of the prior actions, results indicators and targets are summarized in Table 8. Table 7: Comments on Prior Actions and Indicators – Alignment and Specificity 15 Prior Action Indicator Comment The Ministry of (i) The number of civil The relevance of sub-indicator (ii) Finance and Budget of servants with date of birth for prior action 1 is questionable, as the Recipient has information not on record at the MCS did not have the resources eliminated ineligible the MFB, (ii) the civil service to update its roster to enable persons from its payroll rosters at MFB and MCS are reconciliation to take place with the roster of high office reconciled every 6 months, roster maintained by the MFB. holders and (iii) amount of position- Instead, payroll management specific benefits paid to those reforms were pursued through the who had already left the implementation of a new human “high officer� positions. resources software (AUGURE), which was supported by the EU. The Government provided feedback that the formulation of sub- indicator (iii) could have been strengthened by measuring the delay in removing position-specific benefits for those that had left a position of high office, since the Government was already fully committed to the reform. JIRAMA has started to (i) Liter of diesel per kwh of Sub-indicator (i) for prior action 2 publish monthly thermic generation of is too removed from the prior statistics on the electricity by JIRAMA action. While the prior action on the delivery of diesel to, (including the private centers publication of monthly statistics and generation of for which JIRAMA is may contribute to informed electricity at, major responsible for the fuel decision-making, it is arguably too diesel-based electricity supply); (ii) data on diesel far removed from the target of generation centers. consumption and electricity improved efficiency, which generation publicly available depends on a number of other factors such as the integrity of procurement practices for establishing new contracts with Independent Power Providers. Air Madagascar has (i) The audited accounts of Air The indicator on the publication of appointed a new Madagascar are published audited financial statements, states general manager within the time allotted by that the timing of the statement through a competitive regulations should be in line with the process, and (ii) regulations, but it would have been published its audited better to specify what that the financial accounts for timeframe is to avoid any scope for the 2010 to the 2013 confusion with the Client. calendar year on its website. The Recipient has Length of time needed to For the indicator on the timely 16 submitted the audited submit the audited publication of the budget execution final accounts of the consolidated final accounts of report, there was a misalignment State Budget for the the State Budget. with the IMF, which had agreed fiscal years 2009, 2010 with the Government that given and 2011 to the current levels of capacity, the report Parliament for approval should be produced within eight weeks. (c) Relevance of Implementation: satisfactory 34. The implementation arrangements continue to remain highly relevant, with the MFB as the main interlocutor, given the nature of the reforms that are being pursued. Since then, the Government’s implementation arrangements have evolved through the establishment of the Economic Council, which was formed by Presidential decree in November 2015 and has high level Government decision makers across a number of agencies that are relevant to the reforms. The Economic Council was established partly in response to a request from the World Bank to have a centralized focal point for the policy dialogue, given that the programs in 2014 and 2015 involved agencies outside of MFB and required high-level coordination. Examples of specific reforms that were supported by the Economic Council include:  Publication of JIRAMA’s financial information. Starting the process of publishing potentially sensitive financial data has required political will of different stakeholders, including the Ministry of Energy. The Economic Council undertook the necessary follow- up to ensure the data was published.  Publication of Air Madagascar’s financial statements. While there was a delay in publication by five months, this lag would have been arguably longer had the Economic Council not undertaken follow-up measures. Notably, members of the Economic Council have also been involved in Air Madagascar’s restructuring deal, which has involved detailed analysis of the SOEs financial statements. 3.2 Achievement of Program Development Objectives (including brief discussion of causal linkages between policy actions supported by the operation and outcomes): 35. The PDOs for the Madagascar Resilience DPO were: (i) strengthening reporting and increasing availability of information relevant to assessing the effectiveness of public finance, and (ii) improving payroll management and consolidating accounts in line with the treasury single account principle. Objective 1: Strengthen reporting and increase availability of information relevant to assessing the effectiveness of public finance Rating: Moderately satisfactory 36. Overall, progress is rated as moderately satisfactory, since there were moderate shortcomings in the attainment of certain targets, although it is worth highlighting that 17 additional progress has been made to improve the transparency of public finance. Shortcomings refer to the lack of publication of the tax expenditures awarded by the Council of Ministers, and the delay in publishing Air Madagascar’s 2016 financial statement. However, progress has been made in improving the transparency of public finances, such as the regular publication of Citizen Budget Reports and Citizen Execution Reports in three languages (Malagasy, French and English), where the quality and level of detail of the documents is comparable with other countries in Africa. To this end, the MFB now has a dedicated website to budget transparency: http://www.dgbudget.mg/transparencebudgetaire Outcome indicator 1: (i) Liter of diesel per kwh of thermic generation of electricity by JIRAMA (including the private centers for which JIRAMA is responsible for the fuel supply); Baseline: 0.16 L/kwh; Target: 0.14 L/kwh by 03/30/201710; Actual achievement: 0.13 L/kwh. 37. The ratio achieved was 0.13 in July 2017, meaning that progress has exceeded the target. This indicator is intended to measure to the extent to which (i) JIRAMA has replaced the use of diesel by fuel oil, which is less expensive, and (ii) the use of diesel is more efficient [= diesel consumption / (thermic generation by diesel + thermic generation by fuel oil)]. A reduction of the ratio (to the expected 0.14) implies improved efficiency in the use of diesel or increase in the use of fuel oil. 38. It is worth noting that JIRAMA is making ongoing improvements to their governance procedures following complementary reform measures, which are being supported by the World Bank and the IMF. A new business plan has been adopted (structural benchmark in the IMF’s ECF program), which has a number of actions that aim to make JIRAMA financially operational by 2020, and thereby no longer requiring subsidies from the state. The implementation of JIRAMA’s governance-related reforms is being supported by the WB’s PAGOSE project. (ii) Data on diesel consumption and electricity generation publicly available Baseline: No information is available; Target: The database is updated monthly; Actual achievement: monthly data is available which is updated periodically. 39. The second outcome indicator has been substantially achieved, where statistics are published on a monthly basis, with a delay of 45 days to allow for the data to be verified. Between March and July 2016, the production of monthly statistics was suspended to improve the quality of data reporting. Information on the cost of producing thermic energy is publicly available, regularly updated, and accessible. Data is published on indicators such as the stock of diesel available, the quantity of diesel consumed, the quantity of electricity generated, thermic production, the liter of diesel consumed per kwh, and the liter of diesel consumed per kwh of thermic production. 10 Unless otherwise noted in the Indicators that follow, all targets were to be achieved by March 30 th 2017 18 40. The data has led to an improved flow of information within JIRAMA as basis for informed decision-making when agreeing new contracts, and has also been used by civil society organizations in their analysis. For example, the civil society organization, Collectif des Citoyens et des Organisations Citoyennes, makes reference to the statistics in their reports on JIRAMA’s performance, which discusses transparency and efficiency: https://www.fichier- pdf.fr/2017/05/08/donnees-cles-jirama/ Outcome indicator 2: The audited accounts of Air Madagascar are published within the time allotted by regulations. Baseline: Publications are overdue; Target: Publication is on time (in compliance with the Law); Actual: financial statements up to 2015 have been published. 41. This outcome target has been substantially met, where there has been a marked improvement in the publication of Air Madagascar’s financial statements. Commercial legislation (Loi 2003 -036 sur les sociétés commerciales) is applicable to SOEs, and states that all limited companies are required to produce financial statements no later than six months after the closing of the FY. However, there is currently no legal requirement that the financial statements are published. Following the reform supported in the Resilience DPO, financial statements up to 2016 are publicly available, although there was a five-month delay in publishing the 2016 statement. It should be noted, that this reform provided an important building block for measures related to the restructuring of Air Madagascar in 2017, including clearing liabilities to secure a strategic partnership with another airline, with the aim of making the company commercially viable. The availability of audited financial statements provided important information to secure the terms of Air Madagascar’s restructuring agreement. However, the accessibility of the financial statements on the Air Madagascar website could be improved. Outcome indicator 3: List of enterprises which receive tax expenditures through LMIL, EPZ Law and by Council of Ministers decisions publicly available. Baseline: No systematic information is available; Target: The three lists, updated annually, are available; Actual: Two of the three lists are published on EDBM website. 42. Overall, this reform is considered to be partially met as two out of three lists were published. The reform has resulted in significant improvements to the transparency of tax expenditures awarded under special exemption regimes. The companies receiving exemption benefits under the LMIL and the EPZ Law have been published on the website of the EDBM. The reform has helped to systemize data on the enterprises that receive benefits, particularly in Economic Processing Zones, as previously there was no single source of information. 43. However, the exemptions awarded following the decision by the Council of Ministers has not been published, as this information remains undisclosed by the Prime Minister’s Office. The non-publication of this list underscores the need to more closely engage all the 19 stakeholders involved in the program at the time of design, to ensure they are fully committed to the policy action. Furthermore, when achievement was off-track this could have been brought to the attention of the Economic Council in more formal terms. Currently, information on the value of tax expenditures included in this non-disclosed list is not available. 44. This reform has also helped to set the foundation for further work on tax expenditures, including the publication of a tax expenditures statement on the MFB’s website, which was supported in the follow up programmatic PFSI DPO. The reform around improving transparency of tax expenditures continues to be highly relevant as further legislation has been passed in 2017, granting additional tax benefits for Special Economic Zones, Industrial Zones and potentially Agriculture Zones. This evolution of events suggests that further policy dialogue is warranted on the rationale for awarding tax expenditures, and that continued efforts are needed to promote transparency to inform decision-making, including for the list of decisions by the Council of Ministers. Outcome indicator 4: Length of time needed to submit the audited consolidated final accounts of the State Budget. Baseline: Over 5 years (for 2008 accounts); Target: Within 24 months of end year; Actual: Within 24 months of end year. 45. This outcome indicator has been fully met, where the reform supported a much- needed improvement to the timely production of audited state accounts. Madagascar has historically experienced challenges in producing timely audited accounts; a situation which is existed ahead of the political crisis. In 2015, the MFB undertook serious efforts to regularize the situation. A Technical Committee for the Completion of Public Accounts was established to reassemble all state accounts. The MFB intends to deepen the reforms by preparing a Performance Report to accompany the audited state accounts (Loi de Règlement) that is submitted to Parliament, to show the gap between the initial commitments and expenditures. The Court of Accounts (Cour de Comptes) publish an opinion on the quality of the accounts. The achievement of this outcome indicator has directly contributed to the objective of strengthening reporting and increasing availability of information relevant to assessing the effectiveness of public finance. CSOs have actively use the available information to undertake budget analysis, which generally receives wide coverage in the local and regional press: https://ccocmadagascar.wordpress.com/category/loi-de-finances-madagascar/ Outcome indicator 5: Timeliness of the publication of the monthly budget execution report by the Treasury (OGT) Baseline: OGT is not publicly available; Target: Within 6 weeks of end month; Actual: Within 8 weeks of end month. 46. This target has been substantially met, as the availability of the OGT has improved from not being published to being available within 8 weeks at the end of the year. The timeliness of the publication depends on other actors, including the Central Bank, and sufficient time is needed to verify the information. The 8-week deadline is included in the 20 IMF’s Extended Credit Facility program, implying that further efforts could have been at the design stage of the DPO to align with the IMF. Objective 2: Improving payroll management and consolidating accounts in line with the treasury single account principle Rating: Moderately satisfactory 47. Overall progress is rated as moderately satisfactory, since there were moderate shortcomings in the achievement of targets. In particular, the reconciliation of the payroll rosters at MFB and MCS has not been completed, and there are still accounts outside of the TSA. Outcome indicator 1: (i) The number of civil servants with date of birth information not on record at the MFB Baseline: 30,000 civil servants (approx. 25% of total) (2015); Target: 0: Actual: 4,297 48. The first component of the outcome indicator has been substantially achieved. An estimated 86 percent of civil servants have their birth date recorded. It is expected that all civil servants will have their birth date recorded by the end of 2017. (ii) The civil service rosters at MFB and MCS are reconciled every 6 months Baseline: reconciliation has not been done (2014); Target: reconciliation was done within the last 6 months; Actual: No reconciliation has been done 49. The second component of the outcome indicator has not been achieved, as the MCS was not able to secure financing to update the civil servants roster. Again, the non- achievement of a reform by an institution external to MFB highlights the importance of engaging closely with external actors in the design and supervision of the program. 50. However, progress has been made in strengthening payroll controls, through the installation of a new human resources software (AUGURE), where all line ministries are required to declare staff in active service. The AUGURE software can be used to monitor recruitment, reassignments, and the end of contract. The first round of self-declaration was completed in June 2017, and this exercise is expected to be updated every six months. There has been a first reconciliation of the “administrative� roster produced through AUGURE and the payroll roster held at the MFB, which uncovered that an estimated 5,000 non-declared individuals are receiving pay. The MFB is undertaking further audits before cancelling payments to ensure all anomalies are first resolved. There is a plan to undertake an audit of the system in 2018, although financing still has to be secured for this task. (iii) Amount of position-specific benefits paid to those who had already left the “high officer� positions Baseline: Approximately 3.56 bn Ariary (2015); Target: 0; Actual: 0 21 51. The third component of the results target was achieved, as 1,480 irregular payments were regularized. The recently installed AUGURE software will be used to sustain the control processes. Since all cabinet members are nominated by decree it will be possible to track payments to those in ‘high officer’ positions. A combination of reforms to update civil servants’ date of birth information, and cancelling irregular payments is estimated to have resulted in savings of US$10.5m (approximately 0.11 percent of GDP) by February 2017. Outcome indicator 2: The share of dedicated accounts for externally-financed projects held at the Central Bank. Baseline: 0%; Target: 90% of the accounts, with explicit deadline for closure of any accounts remaining at commercial banks; Actual: 87.1% of project accounts are opened at the Central Bank. 52. Overall, progress toward this indicator was substantially achieved. As part of its Public Financial Management Reform plan, the MFB has committed to implementing a TSA, which should consolidate all Government cash balances into a single account.11 However, a transition Decree No. 2016-1160 was approved on August 30, 2016 requiring project accounts to be opened in the BCM, but allowing for the continued operation of secondary accounts in commercial banks for current projects. The approval of this transition decree reflects concerns amongst partners on the integrity of the fiduciary control environment, again suggesting that further consultations could have been undertaken during the design phase with different stakeholders. The transition period will end December 31, 2017. From January 1, 2018 and onward the October 2015 decree will be in full effect. To date 54 project accounts have been opened in the BCM, which is around 87 percent of total foreign accounts.12 53. Other steps are also being taken to advance the TSA reforms, which are being supported by a full-time Resident Advisor financed by the US Treasury, including:  The TSA structure, scope, and phased implementation was approved by MFB management in July, 2017. The initial structure will consist of an international currency account, a central account (TSA), and ledger accounts. All revenues and expenditures will flow through the central account including those for central spending agencies, public administrative entities, public commercial and industrial entities, and international programs and projects. Funds will be converted to local currency and transferred from the international account to the central account as needed. 11 All cash balances will be fungible and thereby available to the Treasurer at any time to make government payments. The advantages of the TSA are to: (i) ensure that adequate cash is available to pay expenditures when they are required; (ii) minimize the need for short-term borrowing; (iii) allow for the short term investing of surplus cash; and (iv) eliminate or reduce the need for costly accounts in commercial banks and other financial institutions. The system should also support account to account automated payment for payroll and payment of invoices for vendors. 12 As of December 18th. 22  Currently under development is a timeline for execution of the TSA as well as guidelines for the implementation and operation of the TSA.  A Regulatory Framework Work Group is operational and is finalizing a new law on cash management as well as decrees for dematerialization of payment system and for operation of the TSA. 3.3 Justification of Overall Outcome Rating (combining relevance, achievement of PDOs): Rating: Moderately satisfactory PDO 1: Strengthening reporting and increasing availability of information relevant to assessing the effectiveness of public finance: Moderately satisfactory PDO 2: Improving payroll management and consolidating accounts in line with the treasury single account principle: Moderately satisfactory 54. Overall, the Resilience DPO delivered relevant reform progress in the areas chosen. The DPO supported the normalization of a number public expenditure management issues following the prolonged political crisis period, including increased transparency of SOE operations, payroll management, transparency of tax expenditures, and the timeliness and availability of public expenditure information. New reforms to modernize public expenditure management were also initiated such as the development of the TSA. These reforms have to a certain extent built on the progress achieved under the 2014 Reengagement DPO, including improving the governance of JIRAMA’s procedures, adopting a law on public debt and clearing arrears. The program also set the building blocks for further reforms supported through the programmatic PFSI DPO, as detailed in table 5. 55. However, while the reforms remain relevant progress in implementation has been uneven, particularly where the main implementing agency has not been the MFB. Of the two outcome indicators that were not achieved, the one related to improving JIRAMA’s thermic energy production was too far removed from the actions supported, and the second one related to reconciliation of the payroll was pursued through alternative means but with the same overall objective of improving controls. Given the overall progress made, which was further strengthened in many areas with the programmatic PFSI DPO, an overall rating of moderately satisfactory has been applied. 3.4 Overarching Themes, Other Outcomes and Impacts (if any, where not previously covered or to amplify discussion above): (a) Poverty Impacts, Gender Aspects, and Social Development 56. The overall impacts of the DPO on poverty and social development are likely to be minimal, and very difficult to measure. While it is challenging to prove the counterfactual, the reforms supported through this DPO have served as a building block for future reforms 23 that increased fiscal space for higher priority social expenditures and public investment. Furthermore, the steps taken to increase transparency of different aspects of public expenditure management have served as a basis for undertaking further reforms, such as the restructuring of Air Madagascar, which is likely to contribute to a revival of the tourism industry and stimulate growth in the future. There were no gender aspects to this DPO. (b) Institutional Change/Strengthening (particularly with reference to impacts on longer- term capacity and institutional development): 57. The DPO’s objectives and respective prior actions supported institutional strengthening. In many instances, the reforms supported a normalization of public expenditure management practices following the prolonged political crisis, helping to foster greater stability. The emphasis of improving transparency of public finances are important steps toward supporting improved decision-making, reporting to elected officials and citizens, and helping to build a social contract between the state and citizens. (c) Other Unintended Outcomes and Impacts (positive and negative): Not applicable 3.5 Summary of Findings of Beneficiary Survey and/or Stakeholder Workshops (optional for Core ICR, required for ILI, details in annexes): Not applicable 4. Assessment of Risk to Development Outcome Rating: Substantial 57. While important progress has been achieved, the overall risk to development outcome remains substantial. Although the Government remains committed to the reform agenda, political tensions and weak governance continue to present a serious risk, particularly as preparations are underway for the Presidential elections in late 2018 / early 2019. A number of reforms supported in the DPO such as reducing the cost of diesel per kwh for thermic generation of electricity supplied by JIRAMA, improving transparency of tax expenditures, and improving payroll controls may affect those who have benefited from the poor levels of governance exacerbated during the political crisis period. Furthermore, political instability could reverse the progress in achieving reforms such as the timeliness and transparency of public expenditure information, as already demonstrated by the previous political crisis. 58. The risk of political instability is closely related with the macroeconomic situation. The fiscal framework is based on an upward trajectory of growth, where domestic revenue collection is increasing and public investments are expected to be scaled up, largely supported 24 by external financing. The macro framework also assumes that important governance related reforms will materialize for the two large SOEs, JIRAMA and Air Madagascar. Slippages on governance reforms, or political instability could compromise the reforms supported through this operation. Institutional capacity risks are largely mitigated through the provision of targeted advisory support. Therefore, while important progress has made, the risks remains substantial, largely related to the quality of governance. 5. Assessment of Bank Borrower Performance 5.1 Bank Performance (a) Bank Performance in Ensuring Quality at Entry (i.e., performance through lending phase): Bank Performance in Ensuring Quality at Entry was moderately satisfactory. 59. The design of the operation was appropriate, timely and responded to the country’s priorities. The reforms supported in the operation helped to normalize procedures following a protracted political crisis, as well as address weaknesses that existed long before, such as the delay in submitting accounts to Parliament. The prior actions set an important foundation for further reforms that were supported in other IPFs and a programmatic PFSI DPO series. This was supported by solid dialogue with the Client, which laid the basis for the Government to define its reform program with strong levels of ownership, particularly from the MFB. The reforms were also well-aligned with efforts from other partners, particularly the IMF. Given the political situation, the choice of a single tranche, standalone operation at the time was appropriate. 60. However, there were some moderate shortcomings, including:  The need to better define in the Program Document how PDO Indicator 1 (i) would be calculated on the liter of diesel per kwh of thermic generation of electricity, which was not clearly specified.  The need to clearly specify in the Program Document PDO Indicator 2 the time allocated in the regulations. The indicator makes reference to publishing the accounts allocated in the regulations, when in fact, this is not required by law, and was rather a timely and important transparency measure supported in the Resilience DPO. The Program Document would have benefitted from further explanation in this area.  Risk ratings were in some cases lower than would have been expected, or recorded in a follow up programmatic series, particularly with respect to stakeholders and fiduciary.  Provide further information on the choice of the title as ‘Resilience DPO.’ The use of the term ‘resilience’ would imply that the Government is taking steps to address the causes of its fragility. However, this justification or explanation does not come out clearly in the choice of the title and its relevance to the Program Development Objectives. 25 (b) Quality of Supervision (including M&E arrangements): Rating: Satisfactory 61. During implementation, the Bank's performance was satisfactory. The Bank focused on the program’s development impact and allocated sufficient budget and staff resources to carry out adequate supervision by closely monitoring program activities. The Government prepared an interim status report showing progress against the achievement of all of the results indicators, provided reasons for divergences as well as progress that was expected by the end of the program. The task team worked closely with the Country Management Unit and other Global Practices, notably the Governance GP, Trade and Competitiveness GP and Energy GP. While there was an active policy dialogue, there was arguably scope to formalize the outcome of the ongoing supervision activities, for example through writing official communication to the MFB where a reform needed further attention. (c) Justification of Rating for Overall Bank Performance: Rating: Moderately satisfactory 62. With a moderately satisfactory rating for quality at entry, and a satisfactory rating for quality of supervision, the overall Bank performance is rated as Moderately satisfactory. This rating considers that while the WB has ongoing policy dialogue with the Client, the stakeholder consultations and identification of risks could have been much stronger. 5.2 Borrower Performance Rating: Moderately satisfactory 63. For purposes of this review, the Government’s performance refers to the MFB, which was directly responsible for implementing most of the reforms. Other agencies were also involved in the reforms, including the MCS, the EDBM, JIRAMA, Air Madagascar, and the Prime Minister’s Office. (a) Government Performance Rating: Moderately satisfactory 64. Overall, the MFB’s commitment to and ownership of the program were adequate but there were moderate shortcomings. The MFB, as a member of the Economic Council, could have taken further steps to ensure the timely publication of the Air Madagascar financial statement, the publication of the Council of Ministers decision on tax expenditures, and the reconciliation of the MFB and MCS rosters. 65. The MFB also took responsibility for the overall supervision and monitoring of the program and furnished relevant documents to the Bank to validate the program’s progress. Implementation monitoring was further enhanced through the establishment of the Economic Council in 2015. Regular interaction and policy dialogue with the Economic Council has facilitated an active policy dialogue and demonstrated commitment to the reforms 26 at the highest level (please see the section of relevance of implementation for specific examples). (b) Implementing Agency or Agencies Performance Rating: Moderately Satisfactory 66. Progress achieved by other implementing agencies was more uneven. The other implementing agencies, JIRAMA and Air Madagascar demonstrated commitment to the reforms supported in the DPO, which have been strengthened through complementary IPFs. However, certain reforms that involve the Prime Minister’s Office and the MCS have not been achieved. 6. Lessons Learned  The need to adequately consider capacity and risks. While the design of the DPO had the stated objective of balancing ambition with realism, this was still a relatively complex operation involving several different stakeholders in the post political crisis period, with a one-year time frame for implementation. In that regard, there could have been further realism in the formulation of targets, and/or or a better identification of risks in Madagascar’s political economy environment that would have informed monitoring and supervision activities. For example, in the case of the delay in publication of the Air Madagascar financial statement, the OGT, and the list of tax expenditures awarded by the Council of Ministers, the target could have considered the internal clearance structures of different institutions in the Government, and that working toward shortening them requires changes in work flow management. Furthermore, the risks section could have more comprehensively elaborated ‘stakeholder’ risk, with a view to encouraging dialogue with different agencies throughout implementation.  Continuity in the policy dialogue can be maintained even with stand-alone operations. While the DPO was prepared on a standalone basis, the continuity of the policy dialogue has been maintained through strengthening the reforms, by building on previous progress in the Reengagement DPO and future support through the programmatic PFSI DPO. Having a follow-up programmatic DPO provided an important avenue for the WB to follow up on the policy dialogue.  Promoting reforms that are supported by the WB (through IPFs and TA) as well as by other partners has both advantages and risks. The benefits are that the policy dialogue is reinforced, there is continuity and that challenging reforms receive implementation support, for example, through technical assistance. However, the risk is that the approach to implementing a reform may be subject to change (as was the case with payroll reforms supported by the EU) and that the timeliness may be affected (such as the TSA reforms). These factors are not entirely in the control of the WB, but could have been elaborated in the risks section.  Early stakeholder consultation is essential, particularly in an environment with 27 weak institutions. Reforms where progress was uneven involved multiple stakeholders moving beyond the MFB, such as EDBM and MCS. The Government’s Economic Council should be further leveraged to support reforms with these other implementing agencies, particularly as the WB continues to pursue multi-sectoral DPOs in Madagascar, with both the PFSI DPO programmatic series and the Inclusive and Resilient DPO, two- year programmatic DPO, which was sent to the Board’s consideration at the end of November 2017.13  While the WB maintained an active policy dialogue with the Client throughout the DPO preparation and supervision, key messages could have been reinforced through more formal mechanisms. For example, Aide Memoires or formal letters could have been delivered to counterparts to bring their attention to the implementation of reforms, particularly when progress was slower or more uneven than expected.  The progress made by the Government in using its own M&E systems for reporting is commendable. Moving forward, these procedures could be further institutionalized by the WB requesting progress reports on a quarterly or bi-annual basis.  Some of the prior actions related to enhancing transparency could have been more specific, to emphasize the accessibility of the information, particularly related to Air Madagascar and the list of tax expenditures. For example, the publication of statistics and reports could be accompanied with a press release to inform the public of its availability, and these processes could be outlined in the Letter of Development Policy. 13 The Inclusive and Resilient DPO supports structural reforms aiming at building the resilience of the rural poor. 28 7. Comments on Issues Raised by Borrower/Implementing Agencies/Partners Not applicable 29 ANNEXES ANNEX 1. Bank Lending and Implementation Support/Supervision Processes (a) Task Team members Name Title Unit Lending Hajarivony Andriamarofara Consultant GGODR Faniry Razafimanantsoa Economist GMFDR Quentin Gouzien Economist GMFDR Enrique Blanco Armas Lead Country Economist GMFDR Keiko Kubota Lead Economist, Task Team Leader GMFDR Mark Thomas Practice Manager GMFDR Rondro Rajaobelison Program Assistant AFMMG Julio Revilla Program Leader GMFDR Nathalie Munzberg Senior Counsel LEGEN Nancy Benjamin Senior Country Economist GTCDR Theresa Osborne Senior Economist GPVDR Vonjy Miarintsoa Rakotondramanana Senior Energy Specialist GEEDR Aissatou Diallo Senior Finance Officer CTRLA Hugues Agossou Senior Financial Management Specialist GGODR Enagnon Ernest Eric Adda Senior Financial Management Specialist GGODR Eneida Fernandez Senior Private Sector Development GTCDR Specialist Madeleine Chungkong Senior Program Assistant GMFDR Anne-Lucie Lefebvre Senior Public Sector Specialist GGODR Noroarisoa Rabefaniraka Senior Transport Specialist GTIDR Anne-Lucie Lefebvre Senior Public Sector Specialist GGODR Supervision Mark Lundell Country Director AFCS2 Coralie Gevers Country Manager AFMMG Faniry Razafimanantsoa Economist GMFDR 30 Heriniaina Mikaela Andrianasy Public Sector Specialist GGODR Hajarivony Andriamarofara Consultant GGODR Fanjaniaina Prisca Mamitiana Consultant GMFDR Abdelmajid Aluiz Consultant GGODR Keiko Kubota Lead Economist, Task Team Leader GMFDR Mark Thomas Practice Manager GMFDR Rondro Rajaobelison Program Assistant AFMMG Natasha Sharma Senior Economist GTCDR Vonjy Miarintsoa Rakotondramanana Senior Energy Specialist GEEDR Aissatou Diallo Senior Finance Officer CTRLA Anne-Lucie Lefebvre Senior Public Sector Specialist GGODR Maharavo Ramarotahiantsoa Financial Management Specialist GGODR (b) Staff Time and Cost (from SAP) (the system pulls data available for all fields) Staff Time and Cost (Bank Budget Only) Stage No. of Staff Weeks US$ Thousand Lending s FY15 5.38 (including travel and 30.736 FY16 25.28 149.970 consultant costs) FY17 1 0.625 TOTAL: 31.66 181.331 Supervision/ICR FY16 0 0.542 FY17 13.08 30.110 TOTAL 13.08 30.652 31 ANNEX 2. Summary of Borrower’s ICR and/or Comments on Draft ICR The draft ICR has been shared with the borrower. The borrower provided additional information on the PDO indicator, which has been incorporated in the main text. 32 ANNEX 3. List of Supporting Documents International Monetary Fund 2016. Madagascar - Request for an Arrangement Under the Extended Credit Facility: First Review Under the Staff Monitored Program-Press Release; Staff Report; and Statement by the Executive Director for Republic of Madagascar, Country Report No. 16/273 International Monetary Fund 2015. Madagascar - Staff-Monitored Program and Request for Disbursement Under the Rapid Credit Facility-Press Release; Staff Report; and Statement by the Executive Director for Republic of Madagascar, Country Report No 15/325 International Monetary Fund 2015. Madagascar, Selected Issues Paper; Country Report No 15/25. International Monetary Fund 2015. Madagascar - Staff Report for the 2014 Article IV Consultation, Country Report No. 15/24 International Monetary Fund 2014. Request for Disbursement Under the Rapid Credit Facility, Country Report No. 16/377 World Bank 2014. Program Document, Madagascar – Reengagement Development Policy Operation, Report No.90809 – MG, dated November 18, 2014. Worldbank 2015. Program Document, Madagascar – Resilience Development Policy Operation, Report No.100436 – MG, dated November 11, 2015. Worldbank 2016. Program Document, Madagascar – Public Finance Sustainability and Investment Development Policy Operation I, Report No. 108936– MG, dated October 25, 2016. Worldbank 2016. Program Document, Madagascar – Electricity Sector Operations and Governance Improvement project, Report No. PAD1147 – MG, dated March 1, 2016 Worldbank 2016. Program Document, Madagascar – Second Integrated Growth Poles and Corridor Program, Report No. PAD1090 – MG, dated November 25, 2014. World Bank. 2017. Madagascar - Country partnership framework for the period of FY17 - FY21. Washington, D.C.: World Bank Group. http://documents.worldbank.org/curated/en/725881498788115661/Madagascar-Country- partnership-framework-for-the-period-of-FY17-FY21 World Bank Group. 2015. Madagascar - Systematic Country Diagnostic. Washington, D.C.: World Bank Group. http://documents.worldbank.org/curated/en/743291468188936832/Madagascar-Systematic- country-diagnostic 33 JIRAMA. 2017. Statistics http://www.jirama.mg/index.php?w=scripts&f=Jirama-page.php&act=statgo AIR MADAGASCAR. 2017. Chiffres clés – En bref http://www.airmadagascar.com/fr/compagnie-aerienne/organigramme-chiffres-clefs Economic Development Board of Madagascar – list of inventoried “entreprises franches� as of October 31, 2017 http://www.edbm.gov.mg/Actualites/Actualites/Liste-des-Zones-et-Entreprises-Franches- inventoriees-au-31o Economic Development Board of Madagascar – list of companies currently managed under DGE and taking advantage of LGIM - 2017 http://www.edbm.gov.mg/Actualites/Actualites/Liste-des-societes-gerees-actuellement-a-la- DGE-beneficiant-du-regime-LGIM Malagasy Treasury Department – Monthly Treasury Operations report - 2017 http://www.tresorpublic.mg/?page_id=214&content=te Direction Générale de la Gestion Financière du Personnel de l’Etat – 2017 http://www.dggfpe.mg/index.php/2016/02/10/solde-les-operations-dassainissement-se- poursuivent/ NewsMada – Press release -2017 http://www.newsmada.com/2017/02/04/gestion-des-soldes-et-pension-le-controle-des-etats- de-paie-lance/ Ma TV - Press release -2017 http://matv.mg/assainissement-des-soldes-et-des-pensions-une-economie-de-7-milliards- ariary-pour-2016/ Ministry of Finance – Madagascar.2017. Budget Transparency http://www.dgbudget.mg/transparencebudgetaire Collectif des Citoyens et des Organisations Citoyennes (CCOC) - report on JIRAMA’s performance https://www.fichier-pdf.fr/2017/05/08/donnees-cles-jirama/ Madagascar – Presidency of Republic - Ordinance N.34316/2015 related to the establishing and organizing the Economic Council 34 ANNEX 4: Follow-up of the reforms in the PFSI programmatic DPO 2017-2018 Reform Prior Action – Public Finance Public Finance area Resilience DPO Sustainability and Sustainability and Investment DPO1 Investment DPO2 Governance JIRAMA has started to The Recipient has, through The Recipient has of JIRAMA publish monthly its Ministry of Finance and caused JIRAMA to (water and statistics on the Budget, issued an Arrêté adopt standard electricity delivery of diesel to, précisant le mode de contracts for all new SOE) and generation of computation des seuils des Power Generation electricity at, major marchés publics et fixant Agreements. diesel based electricity leurs montants, that generation centers imposes a priori review of all JIRAMA procurement by the CNM in accordance with the provisions of the 2004 Public Procurement Code dated March 31, 2016 (N°7275 /2016/MFB). The Recipient has caused JIRAMA to begin publication of quarterly statistics on the cost of electricity per KW.h for each existing power supply contract beginning with the first two quarters of 2016. Improved The Recipient has The Recipient has, through The Recipient has, transparency published, through the its Ministry of Finance and through its Ministry of of tax Economic Budget, issued an Arrêté Finance and Budget, expenditures Development Board of portant exonération des published the first / Madagascar, the list of droits et taxes à annual tax expenditure rationalizing enterprises that are l’importation that restricts statement. exemptions granted benefits under the scope of the customs the Free Zone regime exemption regime in the (Zone franche). Customs Code dated May 4, 2016 (N° 10416/2016/MFB/SG/DGD). Payroll The Ministry of The Recipient has completed The Recipient has (i) management Finance and Budget of the verification of pension completed a reforms the Recipient has beneficiaries in the large verification of the eliminated ineligible urban centers to remove roster of pension persons from its ineligible beneficiaries from beneficiaries in rural payroll roster of high the Pension Register. collectives to remove office holders. ineligible beneficiaries; and (ii) issued 35 Circulaire no. 475- 2017 du 9 août relative à la Standardisation des procédures de versement de cotisations à la Caisse de Prévoyance des Retraites et à la Caisse de Retraite Civile et Militaires, establishing the obligations and process related to the payment of contributions, and sanctions in the case of non-compliance, in order to reduce the transfers to the Pension Fund. 36 MAP 37