Report No. 62082-MR Islamic Republic of MAURITANIA Public Expenditure Review - Update May 2011 Poverty Reduction and Economic Management 3 Country Department AFCW3 Africa Region Document of the World Bank __________________________ This document has a restricted distribution and may be used by recipient only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. Currency Equivalents (as of May 23 2011) Currency = Ouguiya 1$US = 274 MRO System of Measurement Metric System Abbreviations and Acronyms AfDB African Development Bank B/d Barrels per day BCM Central Bank of Mauritania BD Budget Directorate CCM Central Procurement Commission CIB Consolidated Investment Budget CNAM Caisse Nationale d’Assurance Maladie (National Health Insurance Fund) DAPBI Annual Initial Budget Planning Document DGB Directorate General of Budget DGTCP Directorate of Treasury and Public Accounting GDP Gross Domestic Product GNP Gross National Product EITI Extractive Industries Transparency Initiative ENA Ecole Nationale d’Administration EPA Public Administrative Establishment EPIC Public Industrial and Commercial Establishment HIV/AIDS Human Immunodeficiency Virus/Acquired Immune Deficiency Syndrome HRM Human Resources Management IDA International Development Association IFMIS Integrated Public Financial Management System IGF General Inspection of Finances IMF International Monetary Fund MAED Ministry of Economic Affairs and Development MAURIPOST National Post Company MDGs Millenium Development Goals MDRI Multilateral Debt Relief Initiative MoE Ministry of Education MoF Ministry of Finance MRO Mauritanian Ouguiya MTEF Medium Term Expenditure Framework MURITEL National Telephone company ODA Official Development Assistance OECD Organization for Economic Co-operation and Development OHADA Organization for the Harmonization of Business Law in Africa PEFA Public Expenditure and Financial Accountability PER Public Expenditure Review PFM Public Finance Management PIA Project Implementation Agency PIM Public Investment Management PIP Public Investment Program PRECASP World Bank technical Assistance project. PRSP Poverty Reduction Strategy Program PSI Special program of Intervention / Program Spécial d’Intervention RACHAD Automated Spending Chain Network RESEN Rapport de l’Etat sur le Systeme Educatif SOE State Owned Enterprise SOMAGAZ National Gas Company SOMELEC Electricity Company SOMINEX Food Import Company SNDE Company National Water Company SNIM National Mining Company TOFE Monthly government financial statistics UEMOA Union Economique et Monetaire Ouest Africaine (West African Econonomic and Monetary union) UN United Nations UNDP United Nations Development Programme USAID United States Agency for International Development VAT Value Added Tax WHO World Health Organization Vice President : Obiageli K. Ezekwesili Acting Country Director : Mary Kathryn Hollifield Sector Director : Marcelo Giugale Sector Manager : Jan Walliser Lead Economist : Eric Bell Task Team Leader : Dorsati Madani MAURITANIA Public Expenditure Review TABLE OF CONTENTS ACKNOWLEDGMENTS .......................................................................................................................... I EXECUTIVE SUMMARY........................................................................................................................II CHAPTER 1: EXPENDITURE PERFORMANCE AND FISCAL CHALLENGES...........................1 I. Expenditure performance 2004-2009...........................................................................................1 A. Introduction.............................................................................................................................................................1 B. Overview of economic developments in the Mauritanian economy: 2004-2009 ....................................................1 C. Budget execution by type of classification..............................................................................................................6 II. Fiscal Challenges: Creating Fiscal space ...................................................................................22 A. The importance of fiscal space..............................................................................................................................22 B. The cost of Mauritanian Public entities to the central budget ...............................................................................23 C. The impact of policy change and external shock on Mauritania’s fiscal space .....................................................28 III. Conclusions................................................................................................................................31 CHAPTER 2: ANALYSIS OF PUBLIC FINANCIAL MANAGEMENT...........................................37 I. Developments in Public Financial Management Reforms.........................................................37 II. Budget Planning, Execution and Accounting ............................................................................40 A. Budget Planning....................................................................................................................................................40 B. Budget Presentation – Budget Classification ........................................................................................................44 C. Expenditure Execution Circuit..............................................................................................................................45 D. Integrated Public Financial Management System .................................................................................................46 E. Accounting and Reporting ....................................................................................................................................47 III. Public Expenditure Execution Controls.....................................................................................48 A. Internal Control of public expenditure ..................................................................................................................48 B. Reform of the public procurement system ............................................................................................................49 IV. Conclusion .................................................................................................................................51 CHAPTER 3: EVALUATION OF PUBLIC INVESTMENT MANAGEMENT IN MAURITANIA .....................................................................................................................................................................52 I. Performance of the investment budget: .....................................................................................52 A. Trends in Investment Expenditures.......................................................................................................................53 B. Sectoral analysis of the implementation of the 2004-2009 consolidated investment budget ................................58 II. Challenges to Achieving PRSP-3 Objectives ............................................................................61 III. Diagnostic analysis of the public investment management framework.....................................67 IV. Political realities, human and institutional capacity and feasibility of reforms. ........................81 List of Tables Table 1.1 : Key economic indicators for Mauritania, 2004-2009 ................................................................2 Table 1.2 : Execution of the Government Budget (2004-2009) in billions MRO........................................6 Table 1.3 : Wage bill in some WAEMU countries and in Mauritania, 2004-2009......................................7 Table 1.4 : Evolution of the number of civil servants and non-permanent employees ................................8 Table 1.5 : Economic execution of the Budget by category: 2004-2009 in billion MRO..........................11 Table 1.6 : Evolution of share of financing sources in the health sector 2005-2010 .................................15 Table 1.7 : Health sector expenditures by strategic axes, in current MRO, 2005-2010............................16 Table 1.8 : Economic composition of government expenditures executed in............................................18 Table 1.9 : Execution of programmed credits based – overall budget and pro-poor expenditures in billions of MRO ..........................................................................................................................................21 Table 1.10 : Subsidies and transfers to EPAs in 2004-2009 ......................................................................25 Table 1.11 : Amounts to be collected from public establishments.............................................................26 Table 1.12 : Medium Term Macro-Economic Framework, 2007-2014 -- Base case.................................33 Table 1.13 : Medium Term Macro-Economic Framework, 2007-2014 – No reform scenario ..................34 Table 1.14 : Medium Term Macro-Economic Framework, 2007-2014 – Scenario with Lower Iron Ore Prices. ..........................................................................................................................................................35 Table 1.15 : Medium Term Macro-Economic Framework, 2007-2014 – Scenario with tax reform .........36 Table 2.1 : Difficulties in coordinating the implementation of PFM reforms............................................39 Table 2.2 : Indicative calendar for MTEF and Budget Law preparation ...................................................43 Table 3.1 : CIB - National and External Funding, in billions of Ouguiya (Central Government). ............54 Table 3.2 : 2004-2009 CIB: Foreign financing (executed by the central government)..............................57 Table 3.3 : 2004-2009 CIB, Execution - Domestic Financing ...................................................................58 Table 3.4 : 2004-2009 Consolidated Investment Budget ...........................................................................59 Table 3.5 : PRSP-2 sectoral programming and CIB (authorized payments)..............................................63 Table 3.6 : Comparison of PRSP, MTEF, and CIB (budgeted) and CIB (executed) .................................65 Table 3.7: Comparison of future programming (in billions of MRO)........................................................66 Table 3.8 : Alignment between PRSP, PIP, MTEF, CIB, and FL..............................................................67 Table 3.9 : Institutional framework for a successful investment program: Good practices and needed reforms in Mauritania. ...................................................................................................................................................69 List of Figures Figure 3.1 : CIB 2004-2009: Different expenditure categories included in investment spending .............55 Figure 3.2 : CIB 2004-2009: Share of foreign financing in the investment plan .......................................56 Figure 3.3 : Planned foreign aid and authorized payments: 2004-2010 by type of financing....................56 List of Boxes Box 1.1 : The 2007-2008 food crisis in Mauritania: costly measures..........................................................4 Box 1.2 : Income poverty in Mauritania ......................................................................................................5 Box 1.3 : SOMELEC: Employment and wage bill issues............................................................................9 Box 1.4 : Extra-budgetary expenditures and spending through special procedures....................................12 Box 1.5 : Some Highlights from the RESEN Report .................................................................................17 Box 1.6 : The main causes of the poor performance of Mauritanian public enterprises............................27 Box 2.1 : Historical developments in procurement reforms in Mauritania ................................................50 Box 3.1 : Definitions and links: PRSP, MTEF, PIP, budgeted CIB and executed CIB. ............................62 Box 3.2 : Investment Programming and Monitoring System in Mauritania (SYPSIM) .............................73 Box 3.3 : PIP: Project selection phases ......................................................................................................74 List of Annexes Annex 1: Millennium Development Goals, 1990-2015 .............................................................................84 Annex 2 : Country at a Glance ....................................................................................................................85 List of Graphs Graph 3.1 : Institutional framework of the PIM: stakeholders, actors, beneficiaries and roles .................72 List of Maps No table of figures entries found. Acknowledgments This joint Public Expenditure Review (PER) update is the result of close collaboration between the World Bank and the Government of the Islamic Republic of Mauritania. Useful information and insights were provided by government officials in many departments of the Ministry of Education; the Ministry of Finance; the Ministry of Infrastructure and Urbanism; the Ministry of Economy and Development; the Ministry of Agriculture and the Ministry of Health. The joint PER team extends its gratitude to his excellencies Mr. Sidi Ould Tah, Minister of Economic Affairs and Development and Mr. Thiam Diombar, Minister of Finance for their support of the PER update and for their guidance. The team is also very appreciative for the support and active collaboration provided by Mr. Fall Khayar, Director of Programming and coordinator of the national team for the PER-update; Mr. M. Mohamed Lemine Ould Dhehby, General Director of Treasury and Public Accounting; Mr. Papa Bocoum, Deputy General Director of investments projects and programs, Mr. Oumar Guey, Director of Projections and Economic Analysis; Mr. Stephen Luc, Advisor to the Director for Economic Policy and Strategies as well as the many technical staff who provided us with invaluable information and suggestions. For the World Bank, the task was managed by Ms. Dorsati Madani (Senior Economist, AFTP3), who was also the primary author of this report. The task team included Mr. Sid’Ahmed Ould Cheikhna (Economist, AFTP3), Mr. Tawfik Ramtoolah (Senior Public Specialist, AFTPR), Mr. Saidou Diop (Senior Public Finance Specialist, AFTFM) and Mr. Peter Osei (Research Analyst, AFPT3). Mr. Eric Bell (Lead Economist, AFTP3) provided valuable comments and suggestions during the development of the report. Mr. Cherif Diallo (Senior Implementation Specialist, AFTED) provided us with insight and advice about the issues related to expenditure in the education sector. Mr. Moulay Driss Zine-Eddine El- Idrissi (Senior Econonist in Health, AFTHE) and Dr. Eby Ould Cheikh (Focal Point for Health, Nutrition and Population at the World Bank in Mauritania) provided us with valuable sectoral information and preliminary analysis of the PER in Health. The report was prepared under the general supervision of Mr. Jan Walliser (Sector Manager, AFTP3) who offered overall conceptual guidance, and ensured quality control. Ms. Mary Kathryn Hollifield (Acting Country Director, AFCW3) and Ms. Katrina Sharkey (Country Program Coordinator, AFCCI) supported the process and provided the major guidelines. The peer reviewers—Mr. Vivek Srivastava (Senior Public Sector Specialist, PRMPS), Mr. Dino Leonardo Merotto (Senior Economist, ECSP3) and Mr. Tuan Minh Le (Senior Economist, (AFTPR, chapter on public investment management) provided valuable comments and suggestions. Finally, Ms. Fatima Cherif (Program Assistant, AFMMR), Ms. Maude Valembrun (Program Assistant, AFTP3) and Ms. Mariama Daifour Ba (Program Assistant, AFTP3) provided excellent support in helping with the missions, setting up the numerous video and audio conferences and formatting of the final document. i CHAPTER 1: Executive Summary 1. Mauritania is a West African country located on the western edge of the Sahara desert, with a population of approximately 3 million people that is mostly concentrated in the urban areas. The country is in part desert (3/4 of the 1,030,700 square kilometers of the territory). Since independence in the 1960s, Mauritania’s economy has been dependent on natural resources, iron ore first then combined with fisheries, and presently oil and other minerals. The severe droughts of the 1960s and 1970s, which generated migration from rural to urban areas and created pressures on the country’s administration through increased demand for education, housing, employment, health, administrative and other services, which continue up to today. 2. Since 2005, the Mauritanian economy weathered shocks from successive crises. During 2004–2009, Mauritania experienced two coups d’état followed by two transitions to elected governments. Mauritania’s eligibility to the multilateral debt relief initiative in 2006, the beginning of oil exports, the successive food, financial and political-institutional crises, as well as the mining sector boom, were the major factors in the changes in economic aggregates over this period. 3. Mauritania’s economic performance was relatively steady from 2004 to 2006, when GDP growth ranged from 4 to 5 percent despite the beginning of the political crisis in 2005. Inflation fell from 10.4% to 6.2% over the three years. Overall fiscal balance (excluding grants) shot to positive in 2006 to 46% of GDP from -4.8% of GDP in 2004 with the coming on line of petroleum production. The current account improved from -34.6% to -1.3% over the three years. 4. Mauritania experienced difficulties in the wake of a series of crises that beset it during 2007–2009. First, the sharp rise in the prices of staple foodstuffs on the international market led to higher costs of food. Second, the political-institutional crisis of 2008–2009 led to a 15 months freeze of external funding. Third, the financial crisis of 2009 and the fall in the prices of iron ore, Mauritania’s main mineral export, resulted in a drop in the contribution of the National Mining Company (SNIM) to the government budget. 5. Mauritania’s economic performance deteriorated sharply in 2008–09 on the back of these domestic and external shocks. Real GDP contracted from 5.9% in 2007 to -1.2% in 2009. The external positions weakened from a deficit of 9% of GDP in 2007 to 12.3 % of GDP in 2009, and international reserves only covered about two months of imports. The government launched the Special Intervention Programme (programme special d'intervention - PSI) in 2008 to reduce the impact of the food crisis on the population. Thanks to the PSI measures and the drop in foodstuff prices on the international market, inflation was brought down from 7.3% in 2007 to 2.2% in 2009. The country faces several key challenges in its recovery, including a narrow productive base, vulnerability to external shocks, a weak business climate, and persistent poverty levels in the rural sector. Commitment to achieve MDGs has brought success but faces challenges ii 6. Mauritania has adhered to the Millennium Development Declaration in 2000 and included the Millennium Development Goals (MDGs) in its development policies and programmes, including its recently adopted PRSP-3. In fact, Mauritania has already achieved a number of MDGs: (i) universal primary education (goal 2); (ii) gender equality in primary education (component of goal 3); (iii) access to drinking water (component of goal 7). It is in a position to control major diseases such as HIV/AIDS, malaria, and tuberculosis and stopping their spread (goal 6). 7. In other domains, important progress has been made, but they are not sufficient to reach the MDGs. For instance, the incidence of poverty in 2008 was 42%, lower than 2004 (46.7%), but still far from the goal of 25% in 2015. Poverty is still largely a rural phenomenon, with a poverty incidence of 59.4% in 2008, which is slightly higher than the 2004 59.0%. On the other hand, the incidence of poverty in the urban area has decreased from 28.9% in 2004 to 20.8% in 2008. While the extreme poverty has improved overall (from 28.8% in 2004 to 25.9% in 2008), the depth and severity of rural poverty has increased between 2004 and 2008 from 20.6% to 22.3% and from 9.6% to 11.1% respectively. 8. The country is also lagging with respect to other MDGs such as health, the environment, employment and some components of gender equality.1 For instance, infant mortality (MDG 4) stood at 122% in 2007 while the target is 45% by 2015, and maternal mortality (MDG 5) stood at 686 per 100,000 live births in 2007 while the target is 232 per 100,000 live births by 2015. The insufficient resource envelop dedicated to the health sector as well as the poor allocative distribution of these funds within the sector underscore the weak results obtained (chapter 1). 9. Several factors explain this mixed MDG score card. First, slow progress in some of the MDGs calls for further resources to be dedicated to reaching them, pointing to the need for the government to increase its fiscal space (chapter one). Second, the level of MDG achievement is closely linked to the implementation of the PRSP-2 whose second plan of action was seriously hampered by political, economic and institutional crises. The disconnect between the two PRSPs, the PIP, the budget and the various sectoral and global CDMTs, that were developed but never implemented, worsened the already weak strategic guidance provided by the PRSPs to the budgeting process (chapters two and three) and the weak monitoring and evaluation framework on outputs and outcomes. Finally the paucity of reliable public expenditure data did not permit the authorities to identify the lack of budgetary allocative and effective efficiency soon enough to impact the outcomes. This was exacerbated by the lack adequate collection, processing, storage and sharing of data that are supposed to provide baselines and progress indicators at the budget level, especially as it pertains to the investment budget (chapters one and three). Increasing fiscal space to improve results 10. On the fiscal front, the achievement of the MDGs faces a serious challenge. The country is recovering from a series of internal and external shocks that affected its fiscal stance. The global economic recession, the contraction in economic activity, and the reduction in donor 1 Government of Mauritania and United Nations, April 2010. “2010 Progress Report on achievement of Millennium Development Goals (MDGs) in Mauritania.� iii assistance led to lower revenues and a widening of the overall non-oil fiscal deficit. The deficit - excluding oil revenue, foreign-financed spending, and interest due on external debt – rose from 2.2% in 2007 to 7.8 % of non-oil GDP in 2008. In 2009, the non-oil fiscal deficit is estimated to have improved to -6.2 % of GDP. 11. Despite the weakening of the fiscal balances in 2008 and 2009, the authorities are committed to pursuing a fiscal policy consistent with macroeconomic stability and debt sustainability. The authorities prepared a macroeconomic program for 2010–12 to address the negative impact of the global economic downturn, preserve macroeconomic stability, accelerate growth, reduce poverty and consolidate the fiscal position. We replicate this baseline scenario as representing the present direction of the economy based on announced policies and on-going reforms. The overall GDP growth is projected at 5.2% in 2010 and projected at 5.1 % in 2011 and should increase regularly to about 5.7 % in 2014, supported by strong activity in nonoil sectors. On the whole, nonoil GDP growth will rebound to about 5.8 percent in 2014 following a contraction of 1.1 % in 2009. Preliminary projections suggest that the overall fiscal deficit will drop from 5.4% of GDP in 2009 to 2.0% in 2010, helped by the rising commodity prices. It rises again in 2011-2012 due to the new food crisis and the government response to slowly decrease to 2.7% of GDP by 2014. 12. If the government were to slacken its reform drive or backtrack on reforms its ability to affect MDG outcomes will weaken as growth falters and fiscal balances worsen. Chapter one of the PER-update looks at such scenario for 2011-2013, when the replication of economic behaviours and patterns from 2007-2009 brought about substantially negative economic and budgetary outcomes. Growth would be significantly lower and average about 1 percent per year over the medium-term. As a share of GDP, overall investment rates would also be lower than in the baseline, and expected to hover around 22 percent of GDP on average during 2012-14, compared with about 27 percent of GDP in the baseline. With lower fiscal revenues partly reflecting slower economic activity, if the government does not reduce its expenditures as well, the overall budget deficit including grants would average about 7.2 of non-oil GDP over the medium-term compared with 3.3 percent of non-oil GDP in the baseline scenario. This reduced fiscal space for further investment in reaching MDG goals. 13. On the other hand, if the government were to improve its tax base and collect back-taxes, we expect an improvement in the fiscal space the government can dedicate to reaching MDGs. In fact, the government is implementing such a reform, which was launched in 2009. Our projections assume an implementation period of 2012-2014. We expect a one-time overall improvement of some 10% of tax revenues over 3 years. This improvement fades in time as the number of individuals and companies owing back-taxes will shrink every year during the program. The gains from increasing the tax base are longer term, but not captured in our projections. In fact our projections find a slight improvement in the fiscal deficit over the 3 years (0.1-0.2% of GDP) over the period. 14. Three sources of rigidity in the fiscal profile of the authorities raise concerns. First, the tax effort remains a major constraint to effective government action. Revenues did not improve after the peak in 2006, when the government received the first tax revenues from oil (oil exports began in February 2006) and two non-recurring revenues. The non-oil primary fiscal balance iv (excluding grants) is characterized by a chronic deficit (8.6 % on average during the 2006-2009 periods). 15. Second, the civil service bill remains an important rigidity in the government fiscal management. The salary expenditure/tax revenue ratio is one of the highest in the region (60 % on average over the 2004-2009 period, compared with an average of 37 % in UEMOA countries during the same period). The government has tried to manage the wage bill with mixed results. The number official civil servants and non-permanent personnel continued to grow over 2004- 2009, resulting in a civil service that is over 46,000 strong in a country of 3 million people. These numbers do not include the military roster or the employees of autonomous public entities and enterprises, which raises further concerns about the size of a full wage bill. 16. The government should bring down the burden of the public service wages on the budget to create fiscal space for its development objectives. In concrete terms, several actions will help bring under control the expanding burden of the public sector salaries. First, the government needs to urgently finalize and validate the survey of civil servants launched in 2008. Second, the government should use the information obtained from the survey to rationalize the number of civil servants by removing the ghost workers, double payments, and the like. Third, the Ministry of public service and Ministry of Finance are encouraged to set-up an integrated human resources management system. This will allow for greater transparency of the civil service roster and clearer procedures of hiring non-permanent personnel. Fourth, the government needs also to examine and bring under control the number of employees in the public entities outside the central government. Fifth, the authorities would be best served to adopt a recruitment policy that (i) meets the needs of the Administration so as to avoid unreasonable and un-programmed recruitments; (ii) streamlines public service positions by maintaining a match between profiles and positions and (iii) addresses the status of the non-permanent employees as part of a civil service reform. 17. Third, the fiscal and financial status of the 123 State Owned Enterprises (SOE) and public administrative establishments (EPA) and their consequent impact on the budget are a serious cause for concern. We undertake a partial analysis of the costs of the public establishments and enterprises on the central budget. We focus on the costs data that could be obtained from the authorities. We therefore focus on two main aspects of the costs: (i) the costs of accumulated deficits to the authorities; (ii) the costs of non-paid taxes to the authorities. In depth assessment of the contingent liability of these entities for the government will require a much more data, which is not currently available. 18. The para-public subsector costs the central budget in terms of subsidies and in terms of deficits they have accumulated. The 85 EPAs received 3.2 % of GDP on average as subsidy during 2006-2009. Also, as of end December 2009, these EPAs had accumulated a deficit of 3 billion Ouguiya to balance their operating budgets (0.4 % of GDP for the year). The EPICs and the 20 national and mixed ownership companies, face many financial problems resulting largely from poor management, lack of a strategic vision and use of outdated technologies. The budget risks caused by the difficulties facing these companies are essentially related to the management of debt to the banking system and the financing of the recovery plans for these companies. Furthermore, many of the EPICs and 20 enterprises owe government taxes, which is an v alternative form of receiving subsidies. The balance to be collected - corresponding essentially to tax evasion - is very high (5.4 billion MRO). This is the equivalent of 5.1 % of 2009 tax revenues, the equivalent of 19 % of the education budget in Mauritania or 67.5 % of the health budget for the same year. 19. The study provides the following suggestions to start redressing the fiscal / financial situation with the EPAs and EPICs. In the short run, it is useful to undertake an audit of the financial situation of the largest state-owned enterprises SOMAGAZ, SONIMEX, MAURIPOST and SNDE.2 Also, effective supervision will need to be based on a systematic effort to record and analyze the performance of the public entities. The authorities need to set up a simple and transparent reporting system to regularly gather information on the economic and financial performance of all EPAs, including number of employees, investment, revenues and costs, subsidies they have received and financial commitments they have undertaken. In the medium term, the tax administration should scale up its supervision efforts to increase collection of past and current taxes. It will most likely require improved human and institutional capacity as well as stronger punitive authority to achieve this work. Also, the para-public sub-sector needs an adequate regulatory supervision by special directorates within the Ministry of Finance, Economic Affairs and Development and relevant sectoral ministries. This involves strengthening the human capacity, institutional capacity and the punitive authority of these directorates. Signing performance management contracts with the management of main SOEs are known to work well if they include clear and verifiable objectives and indicators, adequate supervision and adherence to the letter of the agreement. This includes reviewing the tariff structures of the SOEs periodically, and introducing - if needed –better targeted and more transparent subsidies to SOEs, which are then recorded in the national budget. 20. Aside from the fiscal space needed to increase funding for the MDGs and other PRSP-3 goals, the government needs to translate its strategic vision laid out in the PRSP-3 into its implementation tools, such as the central budget, the public investment budget, and the various sectoral and global CDMTs, and improve the monitoring and evaluation framework on inputs, outputs and outcomes (chapters two and three). Establishing this connection will depend heavily on continuing the public finance management reform agenda and launching a set of reforms to improve public investment management. Improving Public Finance Management to Improve results 21. While the government has an integrated reform agenda on Public Finance Management (PFM), chapter two of this report highlights a number of bottlenecks that affect the planning and execution of expenditures affecting development goals, and offers a set of prioritized, sequenced measures that mitigate them. To start with, chapter two identifies a dual budget preparation system, where the consolidated investment budget (BCI) is prepared separately from the current budget and is only an annex to the budget document. This disconnect harms programming and execution of the overall budget. The chapter recommends unifying the budget documents (general government budget and consolidated investment budget) under the full responsibility of the Ministry of Finance. In parallel, drawing up the global MTEF for 2012-2014 that should 2 An audit for SOMELEC has already been undertaken. vi underpin the preparation of the 2012 and 2013 budgets would be a major achievement in connecting the strategic view of the authorities with their budget implementation. 22. Third, it is recommended that the authorities operationalize the regulatory bodies responsible for implementing the new procurement code. This reform will improve a procurement process that was too centralized and too lengthy. The new law, once operationalized, is expected to facilitate expenditure processes for sectors, leading to more timely outputs and outcomes from current and investment expenditures. 23. Fourth, the paucity of reliable public expenditure data did not permit the authorities to identify the lack of budgetary allocative and effective efficiency soon enough to impact the outcomes. This was exacerbated by the lack adequate data collection, processing, storage and sharing to provide baselines and progress indicators at the budget level, especially as it pertains to the investment budget (chapters one and three). 24. To better monitor and evaluate the budget implementation process, the government should move ahead with its plans to finalize the interconnections between the different stand- alone PFM information systems. Currently, the absence of a genuine integrated public financial information system is a handicap to the production of reliable financial information and budgetary data. The financial information system of Mauritania is made up essentially of un- integrated information systems deployed at the Budget, Customs, and tax administration, reducing their scope and the visibility of the budget process. Data varies from source to source, sometimes in the same Ministry and often between Ministries. At the Customs, the SYDONIA++ monitors the collection of duties in the customs system. At the tax administration, Al Mouhassil monitors the collection of taxes but has no link with the Treasury system Beit El Mal. At the Budget Directorate, the RACHAD system handles budget allocation and commitment without dealing with payment related issues which are being monitored through Beit El Mal, the treasury and accounting system at the Treasury. The Automated Spending Chain Network (RACHAD) and its extension to the Treasury, Beit el Mal, function relatively well; thus, a file for the execution of expenditures may be processed in real time from commitment to payment, including the different levels at the Treasury, thereby enabling the compilation of the table of the four phases of budget execution. However, some expenditures are only integrated through a gateway, notably salaries and public procurement, while others, such as investment, are not integrated at all or if so, only partially. 25. Finally, to assess past budget performances, learn from them and improve upon them, regular independent audits are necessary. To this effect, it is recommended that the government submits the draft budget review laws for 2007, 2008 and 2009 to the Audit Court for jurisdictional control. Without this control, it is not possible to ensure that expenditure is carried out as planned. Transparency in the budget execution process and the Government accountability to Parliament can be compromised. Consequently, it is necessary to establish a calendar to produce the draft Budget Review Laws and to audit the Outstanding Payments Suspense Accounts, which stood at 7.257 billion Ouguiya at the end of 2009. Improving Public Investment Management to improve results 26. The Consolidated investment budget (CIB) is one of the government’s main tools for vii attaining PRSP-3 objectives. It is currently a stand-alone, relatively ineffective process that does not seem to be based on any strategic guidance. Its project selection criteria and process are not fully transparent or consistent. In fact, there is no unified appraisal methodology and no independent review of the appraisals. The CIB suffers from the duality of the budget procedures and an implementation approach heavily reliant on project implementation unit. The current implementation of the CIB does not include cancelling or unifying project and does not benefit from systematic project completion review and evaluation. Finally, the current system cannot fully track down the expenditures undertaken in different projects or their purported outputs and outcomes. 27. In reforming the CIB effectively, it will be necessary to align the PRSP, PIP, MTEF, CIB and the budget law. Chapter three of this report lays out a pragmatic and simple approach to achieving this goal, highlighting the role of sectoral ministries as beneficiaries but also as participants in the budget process. 28. To improve the CIB performance and its ultimate impact, the authorities may wish to consider the following measures. In the short term, MAED should implement, as quickly as possible, the integrated software under development that will monitor the project cycle from inception to impact. This software should include information on appraisal, approval, accountabilities and supervisory responsibilities, work time horizon, costs, results and impact indicators, among other indicators. In a second stage, this software could include a module directly installed in project implementation units and connected to the MAED and MoF to ensure regular information transfer. Also, the government should lead a biannual multilateral review of the investment budget, which could help alleviate bottlenecks in the implementation of projects and inform the preparation of the next year’s CIB. These biannual reviews can be placed under the aegis of the PRSP-3 Supervisory Committee and combined with PRSP-3 reviews. They can further be timed to coincide with the two annual IMF missions to ensure macroeconomic consistency of PRSP-3 and CIB plans. 29. In the medium term, the efficiency and visibility of the CIB portfolio will improve by streamlining the project portfolio. This can mean unifying similar projects affecting the same beneficiaries. It also means cancelling non performing projects in order to use resources allocated to them on performing projects. While MoF and MAED would be the natural champions of this effort, engagement at the higher political levels will lend credibility to the importance of this task. The authorities should ensure appropriate consultation with the sectoral beneficiaries of the non-performing projects, ensuring that their allocated resources are transferred to optimize the chances of reaching the development objectives of the PRSP-3, potentially to performing projects within the same sector. 30. Finally, MAED should adopt a much more rigorous and systematic methodology for the development and adoption of projects into the CIB portfolio. This includes a clear methodology for preliminary appraisal and cost-benefit analyses for any project/sector and criteria that are more rigorously linked to PRSP-3 and to sectoral strategies. Capacity building at MoF, MAED and sectoral levels will help facilitate implementation of the new methodologies and selection criteria. Encouraging donors to unify their data requirements, methodologies and processes for appraisal and their selectivity criteria will also facilitate improving the process of developing and viii choosing a project into the CIB portfolio. ix EXPENDITURE PERFORMANCE AND FISCAL CHALLENGES I. EXPENDITURE PERFORMANCE 2004-2009 A. Introduction 1.1. This chapter provides Mauritania’s economic context and its expenditure performance during 2004-2009, highlighting the need for more fiscal space. The first section of the chapter will briefly review the recent developments in the economy. Then, the chapter examines the profile of current spending and assesses current budget execution, comparing it to the initial programming. In doing so, the chapter looks at economic and functional expenditures as well as sectoral expenditures and extra-budgetary expenditures. The investment budget will be analysed in Chapter III. 1.2. The second section of this chapter will focus on the importance of creating fiscal space. It will highlight the three major factors that affect Mauritania’s fiscal space and examines the role of State-Owned Enterprises (SOEs) more specifically. Four indicative scenarios are provided to show the impact of keeping to the past status-quo, of the current macroeconomic baseline that includes reforms, of negative shocks and of a positive policy reform. 1.3. It is worth mentioning the difficulties encountered during data collection and reconciliation in order to analyse the execution of government expenditures. Data vary from source to source, sometimes in the same Ministry (Budget Directorate files are different from Treasury Directorate files), and often between two Ministries (Ministry of Economic Affairs and Development files on the Consolidated Investment Budget (BCI) are totally different from those of the Ministry of Finance). Furthermore, government changes have led to multiple mutations in the administrative structure of the ministries, making it impossible to do a proper administrative analysis of expenditures. Given such data disparities, the methodology used consisted in referring to Budget Directorate figures for initial budget programming, to Treasury and IMF files for executed expenditures, and to Ministry of Economic Affairs and Development (MAED) figures for the programmed BCI on external financing. B. Overview of economic developments in the Mauritanian economy: 2004-2009 1.4. In spite of a difficult context, the Mauritanian economy has been able to weather shocks from successive crises. During 2004–2009, Mauritania experienced two coups d’état followed by two transitions to elected governments. These political events had an impact on overall economic activities. Mauritania’s eligibility to the multilateral debt relief initiative in 2006, the beginning of oil exports, the successive food, financial and political-institutional crises, as well as the mining sector boom, were the major factors in the changes in economic aggregates over this period. 1.5. Mauritania’s economic performance was relatively steady from 2004 to 2006, when GDP growth ranged from 4 to 5 percent despite the beginning of the political crisis in 2005. Inflation fell from 10.4% to 6.2% over the three years. Overall fiscal balance (excluding grants) shot to positive in 2006 to 46% of GDP from -4.8% of GDP in 2004 with the coming on line of 1 petroleum production. The current account improved as well from -34.6% to -1.3% over the three years. 1.6. Mauritania experienced its share of difficulties in the wake of a series of crises that beset it during the 2007–2009 periods. First, the food crisis— caused by the sharp rise in the prices of staple foodstuffs on the international market—claimed some victims during demonstrations against the high cost of living in Nouakchott and throughout the country. Second, the political-institutional crisis of 2008–2009 led to a 15 months freeze of external funding. Third, the financial crisis of 2009 and the fall in the prices of some export commodities that are essential to Mauritania, such as iron, resulted in a drop in the contribution of the National Mining Company (SNIM) to the government budget. 1.7. Mauritania’s economic performance deteriorated sharply in 2008–09 on the back of these domestic and external shocks. Real GDP contracted from 5.9% in 2007 to -1.2% in 2009. The external positions weakened from a deficit of 9% of GDP in 2007 to 12.3 % of GDP in 2009, and international reserves only covered about two months of imports. The current account deficit rose to its highest point in 2008 following a sharp increase in currency-based imports under the Special Intervention Programme (programme special d’intervention - PSI), which was launched to reduce the impact of the food crisis on the population. Thanks to the PSI as well as the drop in foodstuff prices on the international market, inflation was brought down from 7.3% in 2007 to 2.2% in 2009. The country faces several key challenges in its recovery, including a narrow productive base, vulnerability to external shocks, a weak business climate, and rising poverty levels. Table 0.1 : Key economic indicators for Mauritania, 2004-2009 2004 2005 2006 2007 2008 2009 2010a Annual change in % Real GDP growth rate 5.2 5.4 4.1 5.9 3.5 -1.2 5.2 (excluding oil) Inflation rate 10.4 12.1 6.2 7.3 7.3 2.2 6.3 In % of non-oil GDP Total revenue -- -- 27.1 30.3 25.1 25.9 26.5 Total expenditure -- -- 36.4 32 32.9 32.1 29.7 Overall fiscal balance -4.8 -7.1 46 -2 -7.8 -6.2 -3.2 (excluding grants) Exports in million US$ 440 625 1367 1343 1787.6 1364.2 2073.5 Incl. Iron (million US$) 823.4 521.6 997.0 Imports (in million US$) -625 -781 -847 955 -1910.9 -1390.9 -1935.3 Current account balance -34.6 -47.2 -1.3 -9 -15.8 -12.3 -8.7 Memo Nominal GDP (in B of MRO) 397 493 729 724 854 794 997 Non-oil GDP (in B of MRO) 397 493 569 678 796 757.6 952.9 Per capita GDP (in US dollars) 541 658 938 952 1123 940 1099 Data: IMF. a/Projections Source: IMF. 1.8. On the fiscal front, the global economic recession, the contraction in economic activity, and the reduction in donor assistance led to lower revenues and a widening of the overall non-oil fiscal deficit. The deficit -excluding oil revenue, foreign-financed spending, and interest due on external debt – rose from 2.2% in 2007 to 7.8 % of non-oil GDP in 2008. In 2 2009, the non-oil fiscal deficit is estimated to have improved to -6.2 % of GDP. Despite the weakening of the fiscal balances in 2008 and 2009, the authorities are committed to pursuing a fiscal policy consistent with macroeconomic stability and debt sustainability. The authorities prepared a macroeconomic program for 2010–12 to address the negative impact of the global economic downturn, preserve macroeconomic stability, accelerate growth, reduce poverty and consolidate the fiscal position. Preliminary projections suggest that the overall fiscal deficit reached 3.2% in 2010, helped by the rising commodity prices. 1.9. The tax effort remains a major constraint to effective government action. Revenues have not improved since the peak in 2006, when the government received the first tax revenues from oil (oil exports began in February 2006) and two non-recurring revenues.3 The parallel jump in government expenditures in 2006 (36.4 % of GDP compared with an average of 32.2 % during the 2007-2009 period) is attributable to the decision taken by the transition authorities (2005-2007) to pay-off a large amount of arrears (26 billion Ouguiya, about 7 % of GDP).4 The non-oil primary fiscal balance (excluding grants) is characterized by a chronic deficit (8.6 % on average during the 2006-2009 periods). 1.10. The government opted to finance the fiscal deficit with resources from the banking system instead of resorting to external debt. Net claims on the government rose 16.3 % in 2008 compared 2007 levels (121.9 billion MRO). This reflects the increase in the stock of treasury bills in the portfolios of primary banks, which rose from 41.3 billion MRO in 2007 to 52.8 billion MRO in 2008 and 54.1 billion MRO in 2009. This situation was further worsened in 2008 by a 7.9 billion MRO decline of the government’s position at the Central Bank of Mauritania (BCM), due to the use of bank financing through the issue of treasury bills, but also to the withdrawal of deposits from the BCM to finance budget deficits. Resources from the banking system represented 42 % of deficit financing in 2009 compared with 31.9 % in 2008. Financing from the banking system had a crowding out effect on credit to the economy. 3 Revenues came from the third mobile telephone license and the results of the agreement concluded with the Australian Woodside in settlement of a dispute on contract amendments contested by transition authorities. 4 Arrears paid in 2006 resulted in the opening of a budgetary line(line 98) authorized by the Minister of Finance. No external control body was petitioned to audit the supporting documents. A summary verification was conferred on a commission coordinated by the IGF under the authority of the MF. 3 Box 0.1 : The 2007-2008 food crisis in Mauritania: costly measures. Mauritania, like most Sub Saharan countries, suffered the repercussions of the world food crisis. The erosion of the purchasing power of households due to unprecedented price hikes, coupled with nearly 30 % unemployment rate and a virtually non-existent social safety net system, led to social unrest. To mitigate the impact of this crisis, the Government set up a multidimensional emergency programme - Special Intervention Programme (PSI) in 2008. The PSI comprised three components. The first component included emergency measures to supply the market and cushion the effects of price hikes through removal of taxes on rice; payment of subsidies to SONIMEX, SOMELEC, SNDE, and SOMAGAZ; increasing civil servant salaries by 10 %; granting food aid; and setting up social safety nets, including an increase of the national food security stock, the sale of wheat at stabilised prices and the free distribution of foodstuffs. The second component included structural measures targeting primarily agriculture through financing the improvement of 30,000 ha of land (an unachieved objective);granting of a 6 billion MRO government guarantee for crop credit and medium-term credit; improving purchasing power; and reinforcing rural and pastoral water management by creating new water points. The third component was reserves for additional measures. PSI Content Amount programmed in B MRO Transfers to public enterprises (gas, electricity, and water) 5.4 Emergency Response (programme special d’intervention) – response to food crisis 19.1 Emergency reserves (Reserves pour depenses additionnelles d’urgence) 3.0 Wage increase 3.5 Total 31.0 Source: Mauritanian authorities and the IMF. The 2008 budget already had programmed a total of 23.3 billion MRO for transfers and subsidies. The government programmed an additional 31.0 billion MRO (US$ 124 Million) for the 2008 budget, of which 20.3 billion MRO (US$ 80 million) were executed. Under-execution of the PSI can be explained by the fall of commodity prices in the second half of 2008, hampering the authorities’ relief efforts. The table above does not include implicit guarantees included in the program, such as the 11.4 billion MRO support provided by the Central Bank to SOMINEX for food imports and the 6 billion MRO sovereign guarantees for crop credit and medium term credits, which apparently did not translate into budgetary spending. The PSI was not evaluated so the authorities do not know to what extend it achieved its stated objectives and mitigated the worst effects of the crisis. Status of MDGs 2004-2009 1.11. Mauritania has adhered to the Millennium Development Declaration in 2000 and included the Millennium Development Goals in its development policies and programmes, including in its recently adopted PRSP-3. The level of MDG achievement is closely linked to the implementation of the PRSP-2 whose second plan of action was seriously hampered by political, economic and institutional crises. A joint Government of Mauritania and United Nations evaluation of progress toward the achievement of the MDGs, conducted in 2010, concluded that Mauritania was likely to achieve several goals. Mauritania has already achieved the following goals: (i) universal primary education (goal 2); (ii) gender equality in primary education (component of goal 3); (iii) access to drinking water (component of goal 7). It is in a position to control major diseases such as HIV/AIDS, malaria, and tuberculosis and stopping 4 their spread (goal 6). In contrast, the country is lagging with respect to other goals such as poverty, health, the environment, employment and some components of gender equality.5 Box 0.2 : Income poverty in Mauritania An update of the poverty situation was undertaken in 2008 using the EPCV methodology.6 The poverty line is US$1/day in 1985 constant prices. Extreme poverty is measured by consumption is US$270 / per person / per year. The update finds that poverty continues to trend down since 1990. Furthermore, it concludes: • Incidence of poverty in 2008 was 42%, lower than 2004 (46.7%), however, still far from the goal of 25% in 2015. 25.9% of the population lived in extreme poverty compared to 28.8% in 2004. Independent farmers are the poorest group with a poverty incidence of 70% while a comparatively low 22% of public sector employees are categorized as such. • Inequality – as measured by the Gini Coefficient – has dropped very slightly from 39% in 2004 to 38% in 2009. • Poverty is still largely a rural phenomenon, with a poverty incidence of 59.4% in 2008, which is slightly higher than the 2004 59.0%. The depth and severity of poverty have increased between 2004 and 2008 from 20.6% to 22.3% and from 9.6% to 11.1% respectively. • On the other hand, the incidence of poverty in the urban area has decreased from 28.9% in 2004 to 20.8% in 2008. Rapid urbanization has created pockets of poverty in urban neighbourhood that have recently sprung up. • There is large disparity in regional rates of poverty: Some wilayas (regions) have rates of poverty incidence below 20% (Nouakchott, Nouadhibou, Tiris-Zemmour), while the incidence in other wilayas is above 60% (Tagant, Gorgol and Brakna). • Gender does not seem to play a major role as a determinant of poverty with 40.3% of households led by women living in poverty compared 42.6% of households led by men. 1.12. The current Government’s determination to achieve results with respect to the MDGs is evident in different policy communications and declarations. Work on the PRSP-3 (2011-2015) which started in 2010 was finalized in March 2011. The key objectives remain the same as in preceding PRSPs: (i) to reduce the poverty rate to 25 % by 2015; (ii) to achieve the various social objectives under different international agreements in the areas of education, literacy, health coverage, reduction in HIV/AIDS rates, access to drinking water and housing, etc; (iii) to reduce social and spatial disparities; (iv) to improve institutional capacities and governance, and (v) to strengthen the supervision, as well as the monitoring and evaluation, of programme implementation. 1.13. To attain the objectives above, the Government affirmed its determination to achieve the following MDGs by 2015: (i) halving poverty by 2015, compared to 1990 (poverty rate at end-period: 28.3 %); (ii) ensuring universal education by 2015 (100 % of boys and girls complete primary education); (iii) reducing gender inequality, especially in education; (iv) reducing infant and child mortality by two-thirds (2/3) and maternal mortality by three- 5 Government of Mauritania and United Nations, April 2010. “2010 Progress Report on achievement of Millennium Development Goals (MDGs) in Mauritania.� 6 Government of Mauritania, February 2010. “PRSP – Volume one: review of the implementation of the 2006-10 PRSP�. 5 quarters ( ¾) between 1999 and 2015; (v) stopping the spread of AIDS, and (vi) ensuring environmental sustainability by halving the population without access to drinking water. The extent of achievement of the MDGs will be closely linked to the implementation of the PRSP-3. C. Budget execution by type of classification 1.14. The list and reorganization of ministries change almost systematically with every new government. This makes analysis of the administrative classification difficult due to lack of legibility. However, it was possible to trace credits by economic and functional classification by crosschecking several data sources.7 1.15. Analysis of the overall budget for 2004-2009 shows a clear increase of programmed expenditures and executions. Budget entries have tripled with visible peaks in 2006 - due primarily to the exceptional conjuncture of revenues and events - and in 2008 due to spending under the special intervention programme drawn up by the authorities to address the effects of the food crisis in 2008 (Box 1). The execution rate of current expenditure for 2004-2009 stood at 94 %, led primarily by the wage bill. 1.16. The non-oil fiscal balance shows a structural deficit over the period, except in 2006. As Table 2 shows, revenues collected in 2006 were nearly twice the expenditures. This exceptional result reflects the start of oil exploitation, receipts from the third mobile telephone licence (27.5 billion Ouguiya) and receipts from the amicable settlement of the dispute between the Australian Oil merchant Woodside and Mauritania (26.7 billion Ouguiya), as well as foreign debt relief under the Multilateral Debt Relief Initiative (MDRI), which stood at 240.9 billion Ouguiya. On the expenditure side, 2006 captures the expenditures undertaken after the 2005 political crisis. Table 0.2 : Execution of the Government Budget (2004-2009) in billions MRO. Year Net expenditures and loans (including external Non-oil revenue and Non-oil balance % financing executed by the central administration) grants GDP 2004 149.4 130.4 - 4.7 2005 166.3 131.3 -7 2006 206.4 406 +35 2007 217.3 185.4 -4.7 2008 262.0 189 -9.1 2009 242.9 188.5 -7.1 Source: MF/IMF and World Bank calculations 7 The budget of Mauritania is tabled before Parliament under an administrative and economic classification. It is identified by 9 digit code (5 digits for the administrative part and 4 digits for the economic part. The credit specialization is indicated at the level of the article. Credit specialization is a principle of budgetary regulations. It means permission to spend is given not only globally but also at the specific budget lines. During the parliamentary control of the yearly budgets, this principle allows the parliament to check whether the government has remained within the framework of the budgetary limits. The introduction of RACHAD and payment authorization reform have given spending ministries an opportunity to break down credit at the sub-paragraph level. 6 Economic Classification 1.17. The breakdown of expenditure by economic classification shows consistent overspending in the salaries and wages category (see tables 3 and 4, category I). In fact, Mauritania’s salary expenditure/tax revenue ratio is one of the highest in the region (60 % on average over the 2004-2009 period, compared with an average of 37 % in UEMOA countries during the same period). This rate hit record levels in 2007, the year that saw the index point value of the basic salaries of civil servants rise from 217 in January 2006 to 434 in January 2007, amidst the euphoria of the oil era. Also, the government eliminated the category V classification entitled “Military Spending�, and distributed its resources to the other subcategories of the budget according to economic classification. This re-categorization of the military expenditures contributed to an increase in category I, which rose in absolute terms from 52 billion MRO in 2006 to 65 billion MRO in 2007 before hitting 77 billion MRO in 2009. As a consequence of this re-categorisation the 2007 wage bill rose by 6 percentage points of tax revenues compared to 2006. Table 0.3 : Wage bill in some WAEMU countries and in Mauritania, 2004-2009 (In % of tax revenues) 2004 2005 2006 2007 2008 2009 Benin 38 38.9 35.9 32.1 35.7 45.3 Burkina Faso 41.3 42.1 44.3 42.4 45.1 49.6 Mali 32.0 31 30.7 33.8 36.1 35.7 Mauritania 57.1 51 54.2 60.2 62.4 72 Niger 30.3 35 33.6 30.4 29.9 28.5 Senegal 30.5 31 30.9 30.3 32.2 33 Togo 31.4 30.2 35.1 32.9 32.5 37.2 Source: Authorities and World Bank 1.18. Several factors have played a role in increasing the volume of category I substantially. The payroll service is not computerized, making the calculation of the actual cost for category I during budget preparation difficult. In fact, the lack of a wage bill forecasting and management tool in the Ministry of Finance largely explains this persistent overrun in budget execution. Also, there has never been a serious attempt at limiting expenditure in this category, leading to a systematic overrun. This is exacerbated by the fact that the size of the military is unknown – owing to security reasons – but has increased the size of category I substantially. In fact, the 2006 decision to include the military expenditure in regular budget categories – instead of stand-alone category - had been considered progress in budget transparency at the time. However this has added to the opacity of budget execution since the military salaries are executed by the summary allocation of credits without feedback or supporting documentation. 1.19. The authorities have attempted to manage the wage bill, with mixed success. The table below provides an overall picture of the number of public servants on the payrolls of the central government.8 The table excludes the military and all employees of autonomous public 8 The mandate of the PER-update does not cover a full analysis of the civil service and its reform. We do not have the detailed information necessary to undertake a review of the wage and benefit structures. 7 entities and enterprises, which is a non-negligible number. Even with these exclusions, the number of regular civil servants is large and grew steadily until 2009, when an executive decision froze further hiring. The government notes however that health and education sectors captured 80% of these new civil servants. 1.20. Of great concern is the existence of some 10,500 non-permanent personnel, hired outside of the statutory framework, who work for the public sector, They were recruited on the basis of services rendered at a time when the government faced a freeze on official recruitment. Their numbers ballooned from 3000 to 10,500 between 2003 and 2009. The peak of this hiring boom was 2006, during the first political transition and was most likely related to it. Originally, these employees were paid from the goods and services and/or the investment budget lines - be it financed by domestic or foreign funds. Since then, the payment of their salaries has been reclassified into category I; however, their existence is in effect an unchecked public employment roster. The government does not provide any systemic oversight over this large pool of employees and their performance. Table 0.4 : Evolution of the number of civil servants and non-permanent employees during 2004-09 - Central Government (excludes military and public entities and enterprises employees) year Civil servants and Non-Permanent Retirements and (1)+(2)-(3) auxiliaries (1) employees (2) Deaths (3) Stock 2003 32299 3000 -- 35299 2004 1502 300 203 36898 2005 1363 1200 350 39111 2006 1881 3000 512 43480 2007 1105 2000 410 46175 2008 1386 1000 900 47661 2009 391 0 1600 46452 Total General 39927 10500 3975 46452 Source: MF and staff calculations 1.21. The table above does not include the employees of public entities and enterprises. As these entities and enterprises have increased in number in recent years (see discussion later in this chapter), so have the number of employees on government payrolls. This report does not attempt to analyse the employment issues of this subsector of the government, but a report on the national electricity company, SOMELEC is revealing of the challenges the government faces in tackling this bulging wage bill (see box ) and achieving a more efficient and professional civil service. To contain and eventually reduce this wage bill, a first step is to assess the number, composition and qualifications of these employees. A second step is to ensure that they are included in the civil service reform. 8 Box 0.3 : SOMELEC: Employment and wage bill issues9 The SOLEMEC report highlights the excessive number of staff employed at the enterprise. The national electricity company, SOMELEC, which serves a country of some three million, had a total of 1838 employees by end November 2009, an increase of 164 employees from Mai 2008. Between May 2008 and November 2009, the ranks of permanent staff grew by 6% while those of non-permanent staff grew by 13.2%. Of the 2009 total, 992 (or 54%) were non-permanent personnel (including subcontractors). The report also notes that while there are too many employees in administration, there is a lack of qualified personnel to do the technical work. For instance, the planning branch lacks technical competence and management skills and it is necessary to develop projection services within the enterprise. Also, the incentive structure and work/pay matrix need to be revised. The report finds that the reform program adopted by SOMELEC in 2007 is still valid. The program planned to improve productivity by reducing the overall number of staff. It intended to develop new organizational charts (defining the content of each position, surveying the staff, identifying the useful staff, and defining the role of each employee). One of the goals of this process was to integrate the permanent “temporary employees� deemed necessary. It also called for a three year training program for all the employees. 1.22. Civil service reform has been slow as it is politically and institutionally difficult to undertaken given the traditional role of the government as the main source of employment and the political instabilities the country has experienced since 2005. The current government appears interested in undertaking this reform within the country specific realities it faces. The authorities, through their PRECASP project10, have updated the legal and human resources management texts of the Ministry of Civil Service and reformed The National School of Administration (Ecole Nationale d’Administration, ENA). A census of civil servants was conducted in order to have an integrated database that would combine to two exiting databases: one that includes all the non-permanent personnel11; the other containing only the permanent staff hired through regular procedures. The validation of this census was delayed by the 2008 political events. However, more than 18 months after the resolution of these events in mid- 2009, the government has still to validate the census and move ahead with the reform of the civil service. 1.23. The government should bring down the burden of the public service wages on the budget to create fiscal space for its development objectives. In concrete terms for Mauritania, several actions will help bring under control the expanding burden of the public sector salaries. First, the government needs to urgently finalize and validate the survey of civil servants launched in 2008. Second, the government should use the information to rationalize the number of civil servants by removing the ghost workers, double payments, and the like. Third, the Ministry of public service and Ministry of Finance would need to set-up an integrated human resources management system. This will allow for greater transparency of the 9 The information on SOMELEC are from the June 2010 report ´Restructuration de la SOMELEC et du Secteur de L’electricité en Mauritanie – Rapport final actualisé » by NOVADIS Conseil and BSD & Associés. The report was financed by the Public Private Infrastucture Advisory Facility and the World Bank Group. Pages 115-120. 10 Funded by the World Bank. 11 This database includes also includes the non-permanent personnel that were managed by Ministry of Finance. 9 civil service roster and clearer procedures of hiring non-permanent personnel. Fourth, the government needs also to examine and bring under control the number of employees in the public entities outside the central government. Fifth, the Administration needs to adopt a recruitment policy that (i) helps it avoid unreasonable and un-programmed recruitments; (ii) streamlines public service positions by maintaining a match between profiles and positions and (iii) addresses the status of the non-permanent employees as part of a civil service reform. 1.24. The goods and services category (category III) of the budget has a 91% rate of execution (on average) over the period (Table 4). The rate of execution of the debt service in categories 3 and 8 remained unchanged during the first two years. However, debt repayment in 2006 stood at 27 % of the budgeted amount. This low rate is explained by Mauritania becoming eligible for the Multilateral Debt Relief Initiative at end-2005. 1.25. Execution of goods and services spending has been affected by non-transparent practices. On one hand, it has been executed through non-competitive bidding practices, depriving the administration of opportunity to secure the best quality/price ratio. On the other hand, it includes the catch-all sub-category “other expenditures�, which during the 2004-2009 period, was executed systematically without competitive bidding, with the staff resorting routinely to contract splitting to avoid the public procurement procedures. 1.26. The authorities cut 20 % the goods and services budget in the 2008/2009 fiscal year and directed the amount to the investment budget. The almost insignificant impact of this cut on the smooth operation of public services suggests that there was lack of transparency in execution as discussed above, and that initial forecasts and budget estimates were largely overestimated. 1.27. No in-depth analysis can be undertaken on the subsidies and transfers category (category 4) as the execution of this category is not sufficiently disaggregated, either at the commitment level (payment authorizer) or at the payment level (accountant). We can only note that the 2008 spike in programming and execution reflect the significant amounts transferred to households through subsidies to public entities and companies under the PSI (See box 1). The lack of detailed information on this category raises concerns about the quality of budget reporting and the absence of traceability in the execution of budget allocations to public enterprises. Also, the difficulty of tracking the exact amounts transferred to households makes it difficult for the government to properly target vulnerable people during crisis periods. Given this reality, the government must develop a mechanism to ensure that transfers to households through public enterprises reach their intended destination. This situation highlights the inefficiency of the expenditure forecast mechanism and explains the tendency to exceed budgetary authority. 10 Table 0.5 : Execution of the Budget by economic category: 2004-2009 in billion MRO. 2004 2005 2006 2007 2008 2009 Category BL Exec BL Exec BL Exec BL Exec BL Exec BL Exec 1-Processing of 16.15 14.06 19.39 18.76 26.87 28.17 60.32 63.97 70.39 70.23 73.61 75.35 Salaries etc. 2- Expenditure on 29.19 24.8 65.07 62.19 65.87 61.26 65.62 62.02 56.89 53.74 56.58 49.58 goods and services 3- Interests on debt 9.15 7.26 13.8 9.59 17.06 11.46 11.9 9.4 12.3 7.34 16.83 15.12 4- Current subsidies 5.7 5.52 9.09 8.71 19.25 18.38 15.76 15.38 44.73 42.51 22.22 17.66 and transfers 5- Unallocated 5.43 3.87 19.21 17.32 23.36 23.26 0 0 15.17 15.15 17.27 16.6 Credits 6- Acquisition fixed 23.3 21.28 23.8 22.3 30.51 26.09 34.65 30.32 52.24 44.67 60.72 36.08 assets and non- product assets 8-Debt repayment 13.55 4.75 166 3.61 40.39 11.74 9.4 1.72 8.9 8.89 7.61 6.5 9-Loans, advances 3.57 3.26 4.39 1.11 15.92 5.94 13.48 10.06 11.5 6.88 16.51 9.43 and participation Total 106.08 84.8 171.37 143.59 239.27 186.3 211.15 192.87 272.5 249.7 271.39 226.32 Source: MoF/IMF and World Bank calculations 11 1.28. Overall, the execution of investment expenditures is much lower than that of current expenditures. This is due to the small investment absorption capacity, slow procurement procedures, and the absence of rigorous monitoring of investment budget execution. The uncertainty that characterizes mobilization of external resources, coupled with the above factors, has often resulted in the government using the investment budget (share sourced from national resources) as an adjustment tool, frequently to the benefit of the operating budget. 1.29. Chapter III analyzes the public investment management system and provides recommendations for reform. Box 0.4 : Extra-budgetary expenditures and spending through special procedures Budget execution in 2004 and 2005 was marked by a high volume of extra budgetary expenditures, but the practice of using automatic debit letters (LDA) was discontinued in 2006. The practice of executing some expenditure through LDA was a fairly current practice until 2006. LDAs, which have no legal or regulatory basis are de facto requisitions issued by the payment authorizer and paid by the central government accountant, without the authorization or endorsement of the financial controller. Thus, of the expenses executed in 2004 and 2005, close to 50 billion ouguiya were executed through LDAs. This category of extra-budgetary expenditures resulted in the signing on 30 December 2004 of an agreement between the Central Bank of Mauritania (BCM) and the Ministry of Finance under which the BCM would pay compensation for government commitments and assets in its cash books to the tune of 81 billion MRO, to be reimbursed over a 33- year period, with a 3-year grace period and at a 4.65 % annual interest rate. Another category of expenditures that covers exchange losses on amounts executed in Mauritanian Embassies through LDAs also resulted, after the agreement, in the regularization of the BCM’s financial claims on the government through the reimbursement of 49 billion Ouguiya, without interest, over 40 years with a 10-year grace period.. The 2006 reform of the chancellery rate ended the haemorrhage caused by assumption of exchange loss. Besides the fact that LDAs were illegal, this practice poses serious problems for accounting and reclassification of the expenditures in question. Adjustment takes place much later and in a rather perfunctory manner. However, discontinuing the use of extra-budgetary expenditures through LDAs in 2006 has resulted in attempts to find alternative processes to undertake rapid disbursement. The LDAs were replaced with two instruments: (i) systematic use of credits upon submission to the treasury of a simple immediate request for payment (demande de réglement immediate or DRI); and (ii) régies d’avance (procedure for credit authorities provided by the Mauritanian law for minor expenditures. 12 In this case, the tendency has been excessive use of DRIs and of régies d’avance. In the case of DRIs, although the practice is provided for by law subject to its regularization at the next session of Parliament, its wide use poses the risk of bypassing legislative authorization and making Parliamentary votes on the budget meaningless. In fact, the PEFA Report compiled a list of expenditures executed in 2007 through the abnormal procedure of immediate request for payment (DRI) or of credit notification; that is, without using the usual procedure of commitment note and the result was clear, 78 % of expenditure was executed without the commitment note. In the case of régies d’avances, there has been credit authorized with amounts higher than 760 million MRO (3 million dollars), as was the case with the election authority in 2006. Also, some works contracts have even been paid for with advance credits. 12 Order No. R-165 of 12 December 1993 concerning advance and receipt authorities, article 3 : �. The creation of an advance authority is reserved for the payment of expenditures of minor importance or for expenditures of a particular and urgent nature�. 12 Sectoral Approach 1.30. An in-depth analysis of sectoral expenditure was not possible due to lack of data. Clearly a thorough understanding of the quality of spending in the major sectors is critical to ensure better inter and intra-sectoral budget allocations as well as improved efficiency of budget executions, especially as the government wishes to achieve MDGs within a limited fiscal space. A health PER is underway independently of this exercise and is expected to be finalized by end 2011. Preliminary results from this health PER are reported below, however, since the analysis is not finalized, this section will refrain from making policy recommendations. Given the government’s commitment to education and infrastructure development, it is also recommended that PERs be launched in these sectors to help illuminate the efficiency of their expenditures and provide recommendations for their improvements. Health 1.31. In health, a sectoral note, developed as part of the elaboration of the 2010-2012 action plan, highlights the three principal issues of the system: the sanitary problems of the population, the problems of the system of health and the external factors affecting the sector.13 1.32. The sanitary problems of the population are linked to the epidemiological profile of the country, dominated by infectious diseases (severe respiratory infections, diarrhoea, malaria, tuberculosis, VIH/SIDA), and peri-natal pathology affecting the health of mothers and children. The nutritional imbalances affect particularly mothers and children. The emerging diseases include cardio-vascular diseases, cancers, mental diseases, diabetes, and the sanitary effects of the consumption of tobacco and of addictive drugs and the new wave of broncho--pulmonary diseases. These diseases are worsened by external factors such as poverty, ignorance and the environment, translating themselves into: (i) maternal mortality that was among the highest in the sub-region in 2007 (686 maternal deaths for 100 000 living births) , (ii) mortality of children under 5 years of 122 for 1000 and (iii) infantile mortality of 77 for 1000. 1.33. The problems of the health system are associated with the different levels of the sanitary pyramid (primary, intermediary and tertiary services): ¾ The sanitary coverage remains insufficient with 33% of the population living beyond five km of a functional health unit. This is linked to the shortages of infrastructures and equipments, and the logistics and maintenance problems that continue to affect the system at all levels. ¾ The human resources have a number of shortcomings: insufficient quantity and quality of professional staff (nurses, midwifes, doctors, and the like), weak capacity building (basic and continuous education), and problems related to personnel management (motivation, career plan, promotion criteria, and the like). 13 « Note sectorielle de la Santé », Ministry of Health, Islamic Republic of Mauritania, January 2011. 13 ¾ Despite government efforts, the health system cannot guarantee the continuous supply of medicines and vaccines and other essentials at reasonable prices. The pharmaceutical sector continues to have a weak legislative and statutory framework, and suffers from lack of qualified personnel.14 The overlap between the public and private sectors continue to hamper the normal functioning of the pharmacy services. ¾ The growth of the hospital sector is hampered by such factors as lack of certain specialized cares, generally inadequate quality of the hospital cares, unsuitable statutes for the sub-sector, the overlap of the public/private care, and an inadequate patient reference system. ¾ Financially, the access of several fringe groups within the population to the health care is limited due to lack of social programs to cover the poorest. ¾ The quality of service provision needs to be unified across the private sector and para- public sector. ¾ The National System of Sanitary Information does not cover the hospital sector, the private or the para-public structures. The data collected are not reliable and the delays of their publication long. 1.34. The health system is also negatively affected by a myriad of external factors. These include problems linked to hygiene (purification, drinkable water and domestic garbage disposal); problems linked to human behaviour related to lack of education; the macroeconomic problems that affect financing and allocations to the sector; the problems linked to the nutrition and food security and natural disasters. 1.35. Preliminary results from the health PER paint a mixed picture15. While funding has improved between 2005 and 2010, the sector is still marked by insufficient funding and inefficient use of the available resources. The sector does not receive sufficient resources from any of its sources of funding to achieve key MDG goals. These sources of funding are: the national budget allocation, the external resources, national health insurance company (Caisse Nationale Assurance Maladie, CNAM) and the contributions made by the population into the cost-recovery scheme. The efficiency of resource use has been affected by two sets of factors. The delays and difficulties in implementing the budget and weak capacities in managing the health structures reduce efficiency of expenditures in the health sector. Also, the intra-sectoral use of resources is not well aligned with the goals and objectives outlined by the authorities. 14 for the quality control of medicines, the management of the pharmaceutical supplies and pharmaceutical services inspection techniques 15 From – « Revue des Depenses Publiques en Republique Islamique de Mauritantie, 2005-2010 - Note Technique sur l’Analyse quantitative (pour le processus du CDMT) « by Driss M. Zine-Eddine El-Idrissi, World Bank, May 20, 2011. 14 1.36. Public expenditure in the health sector (from all funding sources and all institutions), grew substantially over 2005-2010. Public expenditure in the health sector increased from 10.8 B MRO in 2005 to 27.4 B MRO in 2010. This growth translated into an increase in per capita expenditure from 3709 constant MRO to 6171 constant MRO, a 10.7% average annual increase. The public expenditure in health has thus increased from 2.3% of GDP in 2005 to 2.8% of GDP in 2010. 1.37. The share of health expenditure in the government budget slowly increased from 4.1% in 2006 to 10.7% of total public expenditures in 2010. In a comparison with neighbouring countries using 2009 data from WHO data, Mauritania’s share of public expenditures dedicated to health compared to overall public expenditures government ranked second from the bottom, besting only Guinea.16 1.38. The financing gains discussed above are mainly due to the efforts of the donor community. The table below highlights the falling share of the general budget in funding the health sector while the role of international partners has increased. While the donors’ 2010 contributions may be singularly high due to re-engagement by partners after the August 2009 elections, the role of the partners remains substantial. Furthermore, the national health insurance company (CNAM) and households together contribute an average of over 20% over the period. Table 0.6 : Evolution of share of financing sources in the health sector 2005-2010 (In percent of total financing) 2005 2006 2007 2008 2009 2010 2005-2010 General budget 72.7 70.7 59.6 55.0 50.6 45.9 55.9 a/ CNAM 10.5 13.6 20.3 15.2 11.9 households 8.4 13.4 11.4 11.6 11.1 8.4 10.4 Partners 14.5 11.6 15.4 14.8 10.9 27.5 17.3 Others 4.4 4.3 3.0 5.0 7.1 3.0 4.4 Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Source : data base DPS. Table from preliminary results of the PER on health, Word Bank, May 2011. a) is Caisse National Assurance Maladie (national health insurance company) 1.39. The resources are spent in accordance to five policy axes of the National Health Policy. These include: (i) improving supply of health services; (ii) improving the fight against diseases; (iii) supporting the financing of the health system; (iv) improving supervision, control and evaluation; (v) strengthening performance of the sector. The table below shows the strong concentration of expenditures in axe 1. 16 The comparator countries included Burkina Faso, Niger, Chad, Senegal, Mali, Benin, Togo, and Guinea. From – « Revue des Depenses Publiques en Republique Islamique de Mauritantie, 2005-2010 - Note Technique sur l’Analyse quantitative (pour le processus du CDMT) « by Driss M. Zine-Eddine El-Idrissi, World Bank, May 20, 2011. 15 Table 0.7 : Health sector expenditures by strategic axes, in current MRO, 2005-2010. Data from database DPS. Graph from preliminary results of the PER on health, World Bank, May 2011. 1.40. There is a serious mis-alignment between resource distribution within the sector and the stated objectives of the sector. In a country suffering from severe problems with transmissible diseases and high maternal and infantile mortality, the structures providing primary care, health centers and health posts together with public health programs only captured an average 21% of the resources over 2005-2010. On the other hand, tertiary services, such as hospitals capture an average 41% of total resources and the administration captures an average 18% over the same period.17 1.41. Despite the increase in funding, the performance of the system is still very mixed and much remains to be done. While there are notable improvements, the main health indicators have not improved apace of the funding increase and some have even worsened. The indicators related to HIV/AIDS and tuberculoses are improving steadily, with the first dropping from approximately 0.8% in 2005 to 0.5% in 2009, and the second falling from over 100 infections per 100,000 in 2005 to approximately 80 infections per 100,000 in 2010. However, Malaria infections have increased again from some 223,400 in 2005 to over 244,300 in 2010. Also, among the MDGs, especially those related to infantile and maternal mortality, there are improvements, but the objectives will not be reached at the current pace. For instance, between 2000 and 2007, maternal mortality has only decreased from 747 to 686 per 100,000 live births while infantile mortality has only improved from 127% to 122%. For Penta-3 vaccination (children under one year old who have received the complete course of vaccines needed for their age), the rate has dropped from 71% in 2005 to 64% in 2010. Education 1.42. For education, in the absence of a PER, the government finalized a study of the sector (Rapport d’Etat sur le Systeme Educatif National, RESEN) in March 2010. The report outlined the budget performance and assessed the internal and external efficiencies of the system 17 See reference 13. 16 (see box below).18 Taking RESEN as a starting point, an education PER should analyze the relevance of resource allocation to the education sector compared to other sectors, the intra- sectoral allocation of resources and how to optimize the use of these resources to improve the efficiency of sectoral expenditures. Box 0.5 : Some Highlights from the RESEN Report RESEN reports that expenditures on education appear stagnant, going from 3.8% of GDP in 2002 to 3.9% in 2008. The share of current expenditure (excluding debt) in total expenditures allocated to the sector rose from 13% in 2005 to 18.3% in 2008, still below the 20% suggested by the Fast-Track Initiative. Nominal overall expenditures and current expenditures grew faster (2.5 times from 2004 to 2008) than nominal investment expenditures financed by national funds. The international community made 19 up for this difference by tripling its investment in the sector during the period. As of 2008, primary education captured 52.4% of all current intra-sectoral expenditures, compared to 5.2% directed to higher education.20 Also, overall, the investment budget reflects the priorities given to the national policy on Education for All, which is supported by the international community, and which focuses on school infrastructure for the primary and first cycle of secondary education. The launch of the sector-wide approach, entitled PNDSE in Mauritania, has benefitted from the investments of a numerous donors. Salaries and wages capture approximately ¾ of the current expenditures in 2008. The weight of the wage bill decreases as a function of the educational pyramid: at the primary level, the wages constitute nearly 94% in 2008 of the current budget while they constitute only 27% at the university level. The cost of primary education per student stood at 11.3% of per capital GDP in 2004 and at 13.7% of per capital GDP in 2008, rising sharply for other educational levels. This cost stood slightly below the regional average of 14.8% per capita GDP, above Mali’s 10.9% per capita GDP (2008), but below Ghana’s 17.8% per capita GDP (2007).21 Internal efficiency (based usually on rate of repeating a year and drop-out rate for a grade) improved from 54.4% in 2003/04 to an estimated 63.7% in 2007/08 for primary education, mostly due to better retention and lower drop-out rate. This means that in 2007/2008 almost 36% of the resources invested are lost due to students dropping out or having to repeat a year.22 As of 2007, the probability that adult women (age 22 to 44) could read easily (after 6 years of studying) was slightly below 50% in Mauritania, results that are close to Mail and Burkina Faso, but clearly below Benin and Cote D’Ivoire. 10 years of studies are needed to increase this probability to 80%. The few studies of the external efficiency (how much the education system prepares students for employment) find that the rate of insertion in the job market is low. The RESEN concludes that youth leaving the education system captured only 20% of the 23 modern jobs in 2008, while unemployment among the same group is approximately 50%. 18 Rapport d’Etat sur le Systeme Educatif National. March 2010, by MAED and the Ministries and basic and secondary and higher education. Government of Mauritania. 19 IBID RESEN, table III.1, pg. 42. 20 According to RESEN, Mauritania sits at the higher end of the spectrum compared to other countries in the region. The average for 10 comparator countries is at 43.3%. Mauritania has a rate superior to 7 out of 10 countries, with Benin and Burkina Faso surpassing it slightly (table III.8, pg 50). 21 IBID RESEN, table III.10 22 These percentages include students who have to repeat a class but stay in the system. 23 RESEN is a March 2010 government led analysis of the education system (Rapport d’état sur le système éducatif national). Does not include a PER. 17 Economic Classification of Sector Expenditure 1.43. Despite the lack of in-depth data, this section provides a brief overview of the sectoral expenditures by economic classification for four sectors: health, education, infrastructure and rural development. 1.44. The education sector spends ¾ of its budgetary allocations on salaries. This situation is normal as the education sector alone employs about 50 % of civil servants. 1.45. Rural development and the health sectors use 39% and 32.5% on average of their respective budgets on goods and services. This is explained by the volume of vaccines, drugs and health consumables purchased, as well as the purchase of fertilizers, seedlings, farm equipment and tools during crop years. 1.46. Transfers and subsidies cannot be analysed in depth as we could not obtain the necessary detailed information. This is because the transfers and subsidies to public institutions under the 2004 and 2005 budgets were not entered in ministry budgets. Instead, they are entered under title 99 relating to joint expenditures authorized by the Minister of Finance, for which we could not get details. Table 0.8 : Economic composition of government expenditures executed in the education, health, rural development, and transport sectors, 2004-2009 (Share in total expenditure of the sector, excluding external financing, in percentage) Sector Category Sector Category Education Rural Development Staff 74.13 Staff 18.48 Goods and services 10.39 Goods and services 39.22 Transfers 13.05 Transfers 8.6 Investments 2.4 Investments 33.71 Health Transport and Equipment Staff 40.27 Staff 5.77 Goods and services 32.51 Goods and services 15.68 Transfers 12.76 Transfers 4.45 Investments 14.47 Investments 74.09 Source: MF and World Bank calculations 1.47. The small share (2.4%) of government budget resources allocated to investment in the education sector is attributable to the strong mobilization of foreign donors in this sector in recent years. Seven donors have financed the sector through a sector-wide approach (called PNDSE) to the tune of 170 million dollars during the 2004-2009 periods. On the other hand, 74% of the expenditures in the infrastructure sector go to investment. As noted above, this sectoral budget line as other investment budget lines is affected by inclusion of salaries and benefits and other current expenditures. Functional Classification 18 1.48. The Budget of Mauritania is not prepared in accordance with the functional classification; however, an ex-post categorization is currently in use. A functional presentation of the budget will respond to the needs for policy, economic or statistical information. For now, the Directorate General of the Treasury and of Public Accounting has made a significant effort to reclassify budgetary expenditures ex-post. 1.49. Table 6 highlights the relatively low prioritization of pro-poor expenditures in the overall budget. The overall pro-poor expenditures during the period 2004-2009 constitute an average of 44-46% of total budget execution, with a slight dip in 2005 (28.2%) and 2006 (36.3%) due to the political crisis. The low prioritization of pro-poor expenditures is due to the government facing budgetary rigidities associated with high recurrent expenditure (see above discussion) and the increasing number of public entities that capture budgetary resources (see discussion below). Also, the administration has not been able to operationalize previous poverty reduction strategies (PRSP-1 and PRSP-2), leading to a disconnect between the development objectives of country and its budget planning and execution practices. This failure undermines the chances of achieving the MDGs and highlights the ineffectiveness of the budget allocation tool. These issues are discussed in more details in chapter 3. 1.50. It is important to note that the above ratios are optimistic as spending on defence and security forces have been excluded from table 6. Spending by defence and security forces (categories 102 and 103) is not included as it does not provide information to determine what proportion of their budget contributes to poverty reduction. That said, the failure to take into account amounts allocated under these categories improves the execution rate of pro-poor expenses in relation to the overall expenditures. These ratios will increase further if class 111 (unclassified) were to be excluded from the baseline calculations below. Clearly, it would be best to have the detailed expenditure data from all budget categories to enlighten the policy analysis and decision making processes. 1.51. Reviewing the methodology used for the categorization and calculation of the categories of the functional classification may bring more transparency about the true magnitude of pro-poor expenditures in Mauritania. For instance, all health expenditures and community development and housing are categorized as pro-poor, which may not be true. On the other hand, education expenditures have increasingly become pro-poor: from 53.6% in 2004 to 77.2% in 2009. There are alternative explanations for this increase. It may truly be due to a re- casting of expenditure priorities due to the increased awareness of PRSP and MDG goals in education. However, it can also be that over the years as the awareness about pro-poor expenditures has increased so have the expenditures classified as pro-poor, without revision of the earlier years. The same can be said of category 104 (economic affairs). 1.52. To better understand and eventually correct the low execution of pro-poor expenditures, the authorities should consider the following measures. First – the authorities need to create more fiscal space for pro-poor expenditures by bringing under control the wage bill and subsidies to public entities. Second, the Treasury together with the authorities planning the budget should review the pro-poor categorization and calculations to verify and update the pro- poor nature of expenditures, and update the data for all the years to ensure data and analytical consistency. Third, the authorities need to elucidate the pro-poor share of defence and security 19 forces budget to have a more complete picture of overall pro-poor expenditures. Finally, in the process of operationalizing PRSP-3, the authorities need to ensure executed budgets mirror planned budgets closely through tight coordination of the MoF and MAED (further discussed in chapter 3). 20 Table 0.9 : Execution of programmed budgets based – overall budget and pro-poor expenditures in billions of MRO 2004 2005 2006 2007 2008 2009 Class Category Total pro-poor Total pro-poor Total pro-poor Total pro-poor Total pro-poor Total pro-poor Amount Amount Amount Amount Amount Amount 101 General Public Services 35.78 0.17 81.9 0.63 94,3 2.76 61.84 4.37 84.84 4.61 51.15 4.47 104 Economic affairs 25.97 17.54 23.7 13.65 36,30 24.52 21.76 13.99 32.78 22.2 24.18 19.77 105 Environmental protection 0.21 0.21 0.16 0.16 0,43 0.41 1.01 0.96 2.44 0.71 2.33 0.7 106 Community development and 5.35 5.35 5.05 5.05 10,68 10.68 5.85 5.85 11.09 11.07 6.46 6.46 housing 107 Health 4.89 4.89 3.97 3.97 6,79 6.79 8.36 8.58 8.97 8.97 9.19 9.19 108 Religious affairs, culture and 1.23 0.45 1.26 0.57 4.5 1,5 6.82 1.55 6.32 1.84 5.36 0.86 leisure 109 Education 14.44 7.74 15.25 8.71 30.81 14,86 36.56 24.21 36.97 26.54 35.08 27.09 110 Social action and protection 6.79 6.79 5.9 5.9 8.23 8,23 9.51 9.32 18.69 18.69 18.70 18.70 111 Unclassified 7.28 0 36.17 0 Total 94.66 43.14 137.19 38.64 192.04 69,75 151.71 68.83 209.9 94.72 188.62 87.24 Source: MF and World Bank calculations 21 II. FISCAL CHALLENGES: CREATING FISCAL SPACE A. The importance of fiscal space 1.53. The Government needs to create fiscal space if it wishes to allocate additional resources to achieve its MDG and PRSP-3 goals. It can create this space by increasing tax revenues (broadening the tax base or reforming the tax structure), by obtaining additional foreign grants (for example, MDRI), by increasing expenditure efficiency (through reallocation or eliminating wastage), or by borrowing from national or foreign creditors. However, any foreign and domestic borrowing needs to be handled in a sustainable and non-inflationary manner so as to sustain stable macroeconomic conditions. 1.54. Shifting spending from non-priority sectors to priority sectors such as health, education and infrastructure can also enlarge the existing fiscal space to achieve the PRSP- 3 development goals. In fact, Mauritania needs to create additional fiscal space to achieve some MDGs (or at least reduce the gaps between actual MDG indicators and their targeted values). For example, as discussed above, closing gaps under MDG 4 (reducing infant mortality) and MDG 5 (improving maternal health) will require the mobilization of additional funds that may be obtained by significantly reducing non-priority expenditures, such as government transfers to public enterprises. Also, the government’s goal to improve infrastructure will require significant financing as the country is poorly equipped in tarred roads, electricity infrastructure, dams, and airports.24 1.55. Improving the efficiency of the government’s existing sectoral expenditures will also play a key role to making “one Ouguiya go farther�. The efficiency gains of these sectors can be identified by undertaking sectoral analyses of the patterns of inter and intra-sectoral resource allocation and executions (PERs). The analyses will permit identifying whether the money is allocated to the subsectors according to development goals and sectoral strategies and whether they are spent according to their allocation. As noted earlier, a health PER is underway and specific education and infrastructure PERs are recommended as follow-up to this study. 1.56. The analysis of budget categories reveals some rigidity in budget margins over the 2004-2009 periods, limiting the government’s policy choices and its ability to secure the requisite funding to execute projects to achieve its PRSP-3 and MDG objectives. The wage bill, public enterprises and the relatively small tax effort are the three main constraints that weigh on the fiscal space in Mauritania. Also, the economy remains highly susceptible to external shocks and needs to improve the fiscal space to counter them. While resilience to external shocks will take time to develop, it is important to take measures that will help promote efficient mobilization and allocation of resources in areas where the government has the ability to act. Payroll consolidation and eventual civil service reform, transparency in the relations between the government and public enterprises through reform of government portfolio, and the institution of performance management contracts are some of the measures that may bring concrete results. On 24 The investment policy must take into consideration recurrent costs that may result in unsustainable financing needs. Furthermore, infrastructure investments must be results-oriented under a selective approach with a view to ensuring that it does not worsen the national debt ratio due to their relatively low productivity. 22 the taxation front, it is possible to improve collection given the untapped potential that exists, due to tax avoidance and the lack of an information management system in the Directorate General of Taxation. In fact, the government has launched a tax collection program that has met with a degree of success in 2009. B. The cost of Mauritanian public entities to the central budget 1.57. The situation of the State Owned Enterprises and public administrative establishments (EPA) is particularly worrisome. We undertake a partial analysis of the costs of the public establishments and enterprises on the central budget. We focus on the costs data that could be obtained from the authorities: (i) the costs of accumulated deficits to the authorities; (ii) the costs of non-paid taxes to the authorities. In depth assessment of the contingent liability of these entities for the government will require a much more data, which is not currently available. 1.58. There are a total of 123 public entities in Mauritania: 85 public administrative establishments (EPA), 18 public industrial and commercial establishments (EPIC), 9 national companies, and 11 mixed ownership companies. The EPAs may be classified into six categories (Education, Health, Research, Training, Information, and Other). 1.59. Among these 123 public entities, there are strong performers, such as SNIM (mining) and MURITEL (telephone) which have performing management and produce solid results. The stronger performance of these few companies can be attributed to: 1) their ownership structure (they are both mixed-ownership companies); 2) their operation in competitive markets; and 3) access to the expertise of strategic and foreign investors. However, a majority of the 123 public entities face managerial and financial problems. Some important public service providers in Mauritania are technically if not legally insolvent. These 123 entities absorbed 11 % of the government budget during 2004-2009. The financial burden of these entities on the government budget induces a crowding out effect on the financing of social sectors and the building of infrastructure. 1.60. The number of these public entities has increased substantially in the past decade. Some of the recent additions include a new air transport company, two new public property management companies, and a company dedicated to agricultural equipment. Two main factors play a role in this increase: the experience with failed privatizations in the 1990s and early 2000s and the use of public enterprises as an additional means to provide safe employment and financial resources, with an aim to secure political stability. ¾ The experience of the 1990s and early 2000s was one of structural adjustment and privatizations. While the idea of privatization was well received and agreed to, the actual privatization process was often poorly executed, with the perception that it benefitted those were connected to former regimes. It also left serious voids in specific sectors and services that where vital to the country. For instance, for a while there was no provider of urban public transport and demand for training in critical skills for the country, such as mining, remained unsatisfied. The government stepped in again to fulfil some of the public services that affected a large number of citizens, such as public transportation. However, not all interventions are necessarily justifiable ones as they do not benefit a large enough share of 23 the population. For instance, the cost of reviving the national airline must be carefully considered in the context of other options for attracting air connectivity.. ¾ The second factor in the increase of public entities is the spread of clientelism to secure political stability. The authorities of different governments have tried to ensure allegiances from different constituencies, by providing them economic resources through the public budget. This has occurred in an environment where the public sector provides the most reliable source of employment and income in the country. The creation of myriad public entities with allocated budgets and little supervision provides such an avenue. In this sense, these public entities are potential sources of rent-seeking. 1.61. Most of the 85 EPAs are not well managed. First of all, they are not managed on the basis of a results-oriented action plan or performance contracts. Secondly, the accounting/audit office is run by civil servants (often from the Ministry of Finance) chosen in a non-competitive process. These accountants do not have enough job security and independence to be immune to pressures from the management of the EPAs regarding their accounting/audit reports. Finally, the management style of EPAs ensures their autonomy from the government which provides some financing for each, without proper supervision. This supervision should come from technical ministries and the Ministry of Finance in its capacity as the government’s portfolio manager. Neither the technical ministries nor the Ministry of Finance have the technical and institutional capacity to properly monitor the EPAs. 1.62. EPAs cost the central budget in terms of subsidies and in terms of deficits they have accumulated. Table 7 shows that during 2006-2009 3.2 % of GDP on average was used to subsidize these entities. Also, as of end- December 2009, these EPAs had accumulated a deficit of 3 billion Ouguiyas to balance their operating budgets (0.4 % of GDP for the year). 24 Table 0.10 : Subsidies and transfers to EPAs in 2004-2009 Year Subsidy (in billion of MRO) % GDP 2004 6.4 -- 2005 5 -- 2006 15.1 2.7 2007 15.2 2.3 2008 42.,8 5.3 2009 20.3 2.6 Source: MF and World Bank calculations 1.63. The EPICs and the 20 national and mixed ownership companies, face many financial problems (tight cash flow and costly access to bank financing) resulting largely from poor management, lack of a strategic vision and use of outdated technologies. The managerial shortcomings are related to a strong politicization of their Boards of Directors. The outdated technologies used (such as in the case of SOMELEC) ended up making many of these companies financially non-viable and have created risks to the central budget. 1.64. The budget risks caused by the difficulties facing these companies are essentially related to the management of debt to the banking system and the financing of the recovery plans for these companies. The recovery plans did not ensure the sustainability of the solutions implemented. This has led to yearly allocation of subsidies and financial accommodation by the banking sector, with the understanding that the central government would honour the debts of the enterprises. In fact, the government budgeted 8 billion MRO in subsidies in 2009 while overdrafts granted to these entities reached 40 billion MRO in principal as of 31 December 2009 (the equivalent of 170 million dollars). 1.65. Furthermore, many of the EPICs and 20 enterprises owe government taxes, which is an alternative form of receiving subsidies. Table 11 shows that the balance to be collected - corresponding essentially to tax evasion - is very high (5.4 billion MRO). This is the equivalent of 5.1 % of 2009 tax revenues, the equivalent of 19 % of the education budget in Mauritania or 67.5 % of the health budget for the same year. 1.66. The Tax evasion takes several forms and could be stopped by strengthening the supervisory capacity and punitive authority of the tax administration. Some enterprises collect VAT and the IMF from their customers but do not deposit the proceeds into the Public Treasury. Others who have to pay the tax due under “transactions with third-parties� either declare a lower turnover or avoid their commitments to the Directorate General of Taxation. Furthermore, the practice of compensating cross debts (fiscal debt v. government debt) - as is generally the case with SOMELEC - and the reprocessing of public enterprises’ heavy liability to the banking system are not generally carried out in line within the standards of financial transparency. 25 Table 0.11 : Amounts to be collected from public establishments ESTABLISHMENT BALANCE AS AT 05/12/2010 MAURIPOST 75,121,648 ENS 21,621,289 ENER 83,009,855 PARC DAWLING 1,071,229 AMI 8,499,376 CNAM 1,047,772 CFPE 530,127 CUN 721,437,368 CDHLC 8,917,197 CNERV 576,627 SOMELEC 2,033,045,275 SNDE 936,246,154 TVM 12,330,104 CRM 9,103,922 SONIMEX 522,045,530 SOCOGIM 764,641,229 CSA 3,295,786 FNSVA 3,853,949 FI AWQAF 2,126,084 Sté ABATTOIRES 6,825,252 SAM 57,165,640 CFPF 1,981,082 SONADER 11,828,852 ONT 129,053,627 ONS 2,723,658 AMEXTIPE 49,198,214 TOTAL 5,467,296,846 Source: MF/DGI/DGE 26 Box 0.6 : The main causes of the poor performance of Mauritanian public enterprises The Main causes of poor performance are as follows: Poor business models. These models often include old technologies, excess labor, costly and selective access to financing and excessive reliance on implicit and/or explicit government support (lack of hard budget constraints). Weak boards of directors. The boards of directors are usually not effective as an oversight body or in defending the shareholder interest of the state. They are often bypassed by higher level authorities and have insufficient influence to stand up to state-appointed chief executives. In theory, boards are legally obliged to consider all questions of importance that influence the function of the SOE. In practice, they dispense perfunctory approval of executive decisions with little consideration of the merits. Weak audit practices. The quality of the audits is generally lacking. This is partially due to the fact that the audits are carried out by national audit firms recruited based on an ineffective selection process. Weak or non-existent financial reporting. Financial statements are not forwarded systematically to the supervisory authority or are not analysed to justify corrective actions. The Ministry of Finance does not have adequate information on the volume and details of Government accounts.25 The weak taxation approach applied to public enterprises and establishments has led to clear tax evasion and potentially serious governance breaches that cause huge tax losses (as discussed above). Weak supervisory authority and capacity. The Ministry of Finance is the “ownership entity� that oversees SOEs. The specific role of the supervisory directorate within the MoF needs to be reviewed and adjusted to better fit best practice. Its authority over public entities reasserted, its supervisory capacity strengthened. It needs better access to information and also needs to have the ability to analyze a broader spectrum of factors that impact SOE performance. Its say on board member nominations needs to be enhanced. Its role should be well-defined in the government’s policy statement on SOE governance. 1.67. Even by the simple and incomplete calculations above, the para-public sector of the Mauritanian economy proves to be a costly burden on the national budget, limiting the fiscal space available to the authorities to implement their development goals. Adding to this burden are the potential contingent liabilities associated with some of these public entities. In order to redress this situation the government could consider the following two-pronged approach: ¾ In the short run: 1. Undertake an audit of the financial situation of the largest state-owned enterprises SOMAGAZ, SONIMEX, MAURIPOST and SNDE.26 The authorities are committed to undertaking these audits and are developing terms of reference for this work. 25 Officials of the Directorate General of Treasury in charge of drawing up the table of financial operations of the government (TOFE) claim that they are just waiting for public enterprises’ data to produce a comprehensive TOFE. Currently, this statistical document of great value only covers the government’s budgetary operations. 27 2. Effective supervision will need to be based on a systematic effort to record and analyze the performance of the public entities. The authorities need to set up a simple and transparent reporting system to regularly gather information on the economic and financial performance of all EPAs, including number of employees, investment, revenues and costs, subsidies they have received and financial commitments they have undertaken. ¾ In the medium run: 1. The tax administration should scale up its supervision efforts to increase collection of past and current taxes. It will most likely require improved human and institutional capacity as well as stronger punitive authority to achieve this work. 2. The public entities need to be regulated appropriately by the relevant directorates within the Ministry of Finance and Economic Affairs and Development. This involves strengthening the human capacity, institutional capacity and the punitive authority of these directorates. 3. Sign performance management contracts with the management of main SOEs that include clear and verifiable objectives and indicators. 4. Review of the tariff structures of the SOEs, and introduce - if needed – of better targeted and more transparent subsidies to SOEs, which are then recorded in the national budget. C. The impact of policy change and external shock on Mauritania’s fiscal space 1.68. The World Bank’s RMSM-X macroeconomic forecast model offers the possibility to do simulations on the impact of policy measures and shocks and to suggest indicative results to enrich the government’s policy decision making process. We develop a base scenario that is very similar to the macro-economic scenario of the government, which is supported by the IMF and assumes a series of ongoing reforms. We then develop three alternative scenarios that we compare to our baseline: (i) a scenario where the commitment of the authorities weakens, impacting growth and public finance; (ii) a downside scenario where the government experiences a negative shock to its revenues due to a drop in the prices of mining products, which it taxes; (iii) an upside scenario where the government improves its tax collection by 20 % over 2011-2014 (with 10% collected in 2011 in the baseline scenario, and another 10% collected over 2012-2014). 1.69. The RMSM-X model (extended minimal revision Standard Model) is a macroeconomic simulation tool to analyse macroeconomic policies and monetary flows in developing countries. It models the economy by using a comprehensive flow of funds framework. A basic model includes national accounts, the public sector, the monetary sector, and an account for the rest of the economy. The model also projects detailed accounts on trade and external debt stocks and 26 An audit for SOMELEC has already been undertaken. 28 flows. It is used to analyse policies and for standard World Bank analyses such as annex reports on country assistance strategies. Detailed information on the model is available on the World Bank Group portal for data development (DECDG). 1.70. Medium-term prospects should remain encouraging if Mauritania is able to efficiently implement its proposed reform programme and its investment programme. The main medium-term challenge is to achieve strong and sustainable growth to boost employment and further reduce poverty. However, this requires budget consolidation, a cautious monetary policy, and the acceleration of structural reforms, notably in public financial management and in the financial sector. Baseline scenario 1.71. The baseline medium-term projection represents an assessment of the present direction of the economy based on announced policies and on-going reforms. Overall GDP growth is estimated at 5.2% in 2010 and projected at 5.1 % in 2011 and should increase regularly to about 5.7 % in 2014, supported by strong activity in nonoil sectors. On the whole, nonoil GDP growth will rebound to about 5.8 percent in 2014 following a contraction of 1.1 % in 2009. Nonoil growth will be stimulated partly by the launch of a major project (estimated at about 1.1 billion dollars) by the National Mining Corporation (SNIM). The project comprises the following four components: (a) increasing capacity; (b) building a new port terminal to export mining products; (c) modernizing the rail link between the mine and the port, and (d) building a maintenance base. When implementation will start, construction will include increased activity in construction-related sectors such as industry and services. With proper management of monetary and tax policies, and with mild inflationary pressures in spite of the strong economic activity expected in the coming years, the nonoil GDP deflator will be reduced from about 19 % in 2010 to zero in 2014. 1.72. As regards the budget, the government is determined to review the scope and efficiency of the public service remuneration system in the medium-term. Possible measures include decompressing the salary structure, setting up a system geared towards performance- based remuneration, improving recruitment policies, and implementing new public finance management measures. Transfers to public enterprises should also be reduced by revising tariff structures and introducing better targeted subventions. Once these measures are implemented, the overall budget deficit (including grants) will drop from 5.4 % of nonoil GDP in 2009 to about 2.7 % in 2014. This will help control inflationary pressures, free up resources for public and private investment, and ensure sustainability of public debt. 1.73. The external position should improve in the medium-term. Growth in exports will be driven by the recovery of the world economy and improvement of local capacity. Export volumes, of iron ore in particular, should increase significantly by an average of about 12 % per year in the medium-term. With higher import growth associated with rapidly growing economic activity, the current account deficit will rise moderately from about 7.5 % in 2011 to about 10.9 % of GDP in 2014. Foreign exchange reserves are projected to accumulate to an equivalent of about 2.3 months imports of goods in 2014, compared to 1.8 months imports in 2010. Alternative scenario I: no implementation of reforms. 29 1.74. If the Government’s commitment to continue to reform the economy waivers and implementation efforts are not intensified, we assume a scenario where growth is faltering. We use the average of the 2007-2009 economic performance to mimic this scenario. The indicative consequences on the economic and budgetary outcomes can be substantial. Growth would be significantly lower and average about 1 percent per year over the medium-term. As a share of GDP, overall investment rates would also be lower than in the baseline, and expected to hover around 22 percent of GDP on average during 2012-14, compared with about 27 percent of GDP in the baseline. With lower fiscal revenues partly reflecting slower economic activity, if the government does not reduce its expenditures as well, the overall budget deficit including grants would average about 7.2 of non-oil GDP over the medium-term compared with 3.3 percent of non-oil GDP in the baseline. Reserve accumulation will continue albeit a slower pace than in the baseline. 1.75. Clearly, in this scenario, the fiscal space that could be dedicated to PRSP-3 expenditures has shrunk, negatively affecting the prospects of reaching the MDG goals. Alternative scenario II: short-term negative external shock to the iron ore prices. 1.76. Ceteris paribus, we assume a strong short term shock to Iron ore prices -a drop by 50 % in 2011- and then slightly slower recovery, with a rise by 40 % in 2012, 5 % in 2013, and 5 % in 2014. As export earnings drop significantly, the current account deficit will equally worsen in 2011 to about 28 % of GDP (compared with 8 % of GDP in the baseline scenario) before improving gradually to about 15 % of GDP by 2014. Additional foreign financing to cushion the balance of payments should be made available to the government on concessional terms by development partners. 1.77. The shock would worsen the public deficit, restricting the government’s fiscal space over the medium term. The semi-public mining company SNIM that mines the iron ore pays taxes and dividends to the government. Lower iron ore prices would lead to a lower tax and non tax revenues (dividends). The scenario assumes that while the price drop occurred in 2011, the loss of revenues will be felt in 2012 and beyond as SNIM usually pays taxes and dividends for one year’s activities in the following year. With nominal public consumption assumed to remain at its baseline level, the medium-term overall budget deficit would worsen in 2011 compared to the baseline scenario and remain on a slightly higher path than the baseline scenario over the next 3 years, declining from about 4.8 % of nonoil GDP in 2011 to about 3.2 % in 2014. To compensate for this negative shock, three actions can be considered: (i) reduce the recurrent expenditures to avoid increasing deficits; (ii) secure grants and concessional lending all the while ensuring a sustainable debt profile; (iii) improve revenue collections from other sources, including public EPAs, EPICs and mixed ownership enterprises. The scenario discussed above did not consider the impact of this price shock on overall growth, lower imports and lower consequent taxes. We only limited the impact on the first rounds of effects and therefore underestimated the impact of such a shock on the whole economy and secondary impacts on the budget. Alternative scenario III: implementation of a tax collection effort over 3 years. 30 1.78. We model an increase in nonoil revenues from direct taxes that shadows a current effort on the ground to collect back taxes from contributors and enlarge the tax base. We modelled direct taxes to increase 5 % in 2012, 3 % in 2013, and 2 % in 2014, above their nominal levels in the baseline scenario, which we assume already contains the impact of the reform in 2011. The authorities confirmed that the reform has already been underway in 2010 and that our hypothesis of decreasing gains from the tax collection effort is a realistic one. For 2012, the scenario predicts a reduction of the overall budget deficit from 4.1 % of non-oil GDP in the baseline scenario to 3.9 % of GDP in our tax scenario, creating marginal fiscal space. The overall budget deficit reduction will amount to about 0.1 percent of non-oil GDP per year during the subsequent two years. 1.79. We did not consider the potential impact of such a tax effort on the activities of the private sector in our model. It is therefore important to note that while efforts at collecting back taxes and enlarging the tax base key to sustained tax collection in the medium to long run, it is also critical to review and revise as needed the tax rates to ensure a simple, transparent and attractive fiscal environment for the private sector. III. CONCLUSIONS 1.80. The chapter discussed the economic context of Mauritania, examined the profile of current expenditures, comparing them to initial programming. It also looked at two major drivers of budget rigidities (salaries and SOEs) and the indicative impact of four different scenarios on the fiscal spaces. A number of recommendations were outlined in the chapter. We reiterate a priority list of reforms that would help the authorities enlarge fiscal space in the medium term. These include actions to address the public sector wage bill and the SOEs situation: ¾ In the short term: 1. Validate the survey of civil servants launched in 2008. 2. Using the validated survey, rationalize the number of civil servants by removing ghost workers, double payments, and the like. 3. Set up an integrated human resources management system. 4. Undertake an audit of the financial situation of the largest state-owned enterprises SOMAGAZ, SONIMEX, MAURIPOST and SNDE. 5. Set up a simple and transparent reporting system to regularly gather information on the economic and financial performance of all EPAs, including number of employees, investment, revenues and costs, subsidies they have received and financial commitments they have undertaken. ¾ In the medium term: 1. Adopt a public sector recruitment policy with clear guiding principles (as outlined in paragraph 23). 2. Identify and bring under control the number of employees in the public entities outside the central government. 3. Give the tax administration and regulatory and supervisory directorates at MoF and in 31 sectoral ministries clear institutional and punitive authority to actively supervise the EPAs. This will require institutional and human capacity building. 4. Sign performance management contracts with the management of main SOEs that include clear and verifiable objectives and indicators. 5. Review of the tariff structures of the SOEs, and introduce - if needed – of better targeted and more transparent subsidies to SOEs, which are then recorded in the national budget. In summary, Reforms to increase fiscal space in the medium term Responsible Time Frame Validate the survey of civil servants launched in 2008. Ministry of Civil Service, Prime 1 year minister, council of ministers Using the validated survey, rationalize the number of civil Ministry of Civil Service, Prime 1 year servants by removing ghost workers, double payments, minister, council of ministers and the like. Set up an integrated human resources management MoF, Ministry of Civil Service, 1 year system. sectoral ministries Undertake an audit of the financial situation of the largest MoF, Prime Minister 1 year state-owned enterprises SOMAGAZ, SONIMEX, MAURIPOST and SNDE. Set up a simple and transparent reporting system to MoF, Sectoral Ministries, EPAs 1 year regularly gather information on the economic and financial performance of all EPAs, including number of employees, investment, revenues and costs, subsidies they have received and financial commitments they have undertaken. Adopt a public sector recruitment policy with clear Ministry of Civil Service, Prime 1-3 years guiding principles (as outlined in paragraph 23). Minister, council of ministers Identify and bring under control the number of employees Ministry of Civil Service, Prime 1-3 years in the public entities outside the central government. Minister Give the tax administration and regulatory and MoF, Sectoral Ministries, Prime 1-3 years supervisory directorates at MoF and in sectoral ministries Minister, council of ministers clear institutional and punitive authority to actively supervise the EPAs. This will require institutional and human capacity building. Sign performance management contracts with the MoF, Sectoral Ministries, SOEs 1-3 years management of main SOEs that include clear and verifiable objectives and indicators. Review of the tariff structures of the SOEs, and introduce MoF, Sectoral Ministries, Prime 1-3 years - if needed – of better targeted and more transparent Minister subsidies to SOEs, which are then recorded in the national budget. 32 Table 0.12 : Medium Term Macro-Economic Framework, 2007-2014 -- Base case 2007 2008 2009 2010 2011 2012 2013 2014 Est Projected National income and prices Real GDP growth (%) 1.0 3.5 -1.2 5.2 5.1 5.7 6.3 5.7 of which: Nonoil 5.9 3.9 -1.1 5.7 5.3 5.8 6.4 5.8 GDP deflator (% change) 0.2 12.4 -5.9 19.4 9.4 5.9 1.3 -0.8 Non-oil GDP deflator (% change) 13.4 13.0 -3.9 19.0 9.2 6.2 1.6 0.0 External sector Exports of goods (nominal; % change) 2.5 27.5 -23.7 52.0 37.5 7.4 -2.4 -4.8 of which: Nonoil 45.2 37.5 -21.6 55.0 46.5 8.2 -2.4 -5.0 Imports of goods (nominal; % change) 36.7 21.8 -27.2 39.1 27.5 10.7 -0.9 -0.4 Current account balance (% of GDP) -18.3 -14.8 -10.7 -8.7 -7.5 -9.2 -9.6 -10.9 Gross international reserves (US$ million) 208.8 194.9 238.0 287.3 395.4 442.3 493.8 518.8 Gross international reserves (months of imports goods) 1.6 1.2 2.1 1.8 1.9 1.9 2.2 2.3 Money M oney and quasi-money (% change) 18.9 13.7 15.2 12.9 13.3 13.4 15.3 13.0 Credit to private sector (% change) 19.0 28.9 3.7 14.2 12.8 11.7 12.8 12.3 Investment and savings Gross investment (% of GDP) 22.3 27.3 24.6 24.5 25.7 27.3 27.0 26.3 National savings (% of GDP) 11.0 12.5 13.9 15.7 18.2 18.1 17.5 15.5 Public Finance Government revenue and grants (% of non-oil GDP) 1/ 30.3 25.9 26.7 27.7 27.4 25.8 25.3 25.3 Oil 2.9 2.2 1.8 1.4 1.2 1.1 1.0 0.9 Non-oil 24.9 22.9 24.1 25.1 24.9 23.7 23.3 23.5 Tax 2/ 15.8 14.4 14.1 16.3 14.9 15.9 15.9 16.2 Nontax 9.2 8.5 10.0 8.7 10.0 7.8 7.4 7.3 Grants 2.4 0.8 0.8 1.2 1.4 1.0 1.0 0.9 Government expenditure (% of non-oil GDP) 32.0 32.9 32.1 29.7 30.3 29.9 28.5 28.0 Current expenditure 23.4 25.9 24.6 21.5 22.2 20.4 19.3 18.8 Capital expenditure 3/ 8.6 7.0 7.4 8.2 8.1 9.4 9.2 9.2 Overall balance including grants (% of non-oil GDP) -1.8 -7.0 -5.4 -2.0 -2.8 -4.1 -3.2 -2.7 External debt Debt outstanding to GDP ratio: PPG (%) 82.9 80.8 93.5 67.9 47.5 48.6 52.6 54.5 Memo items Exchange rate (LCU/US$, end-period ) 252.2 261.5 262.0 282.0 .. .. .. .. Nominal GDP (LCU billion ) 734 854 794 997 1147 1284 1382 1450 Nominal non-oil GDP (LCU billion ) 679 797 758 953 1096 1232 1331 1409 Nominal GDP (US$ million ) 2822 3536 3031 3629 3999 4322 4526 4642 Population (million ) 3.1 3.1 3.2 3.2 3.3 3.3 3.4 3.5 GDP per capita (US$ ) 918 1123 940 1141 1228 1296 1326 1328 Nonoil direct taxes (lcu billion) 33 35 35 43 50 59 65 73 Nonoil direct taxes (% non-oil GDP) 4.9 4.4 4.6 4.6 4.6 4.8 4.9 5.2 Source: IM F/WB staff estimates and projections 1/ including net revenue from oil. 2/ 2010 includes tax arrears from national electricity company. 3/ including net lending 33 Table 0.13 : Medium Term Macro-Economic Framework, 2007-2014 – No reform scenario 2007 2008 2009 2010 2011 2012 2013 2014 Est Projected National income and prices Real GDP growth (%) 1.0 3.5 -1.2 5.2 5.1 1.1 1.1 1.1 of which: Nonoil 5.9 3.9 -1.1 5.7 5.3 2.9 2.9 2.4 GDP deflator (% change) 0.2 12.4 -5.9 19.4 9.4 2.3 2.3 2.3 Non-oil GDP deflator (% change) 13.4 13.0 -3.9 19.0 9.2 2.2 2.3 1.0 External sector Exports of goods (nominal; % change) 2.5 27.5 -23.7 52.0 37.5 -3.3 -4.9 -0.4 of which: Nonoil 45.2 37.5 -21.6 55.0 46.5 -5.9 -8.3 -3.9 Imports of goods (nominal; % change) 36.7 21.8 -27.2 39.1 27.5 -5.1 -4.5 -1.3 Current account balance (% of GDP) -18.3 -14.8 -10.7 -8.7 -7.5 -6.5 -6.0 -3.4 Gross international reserves (US$ million) 208.8 194.9 238.0 287.3 395.4 390.0 419.7 435.7 Gross international reserves (months of imports goods) 1.6 1.2 2.1 1.8 1.9 2.0 2.2 2.4 Money M oney and quasi-money (% change) 18.9 13.7 15.2 12.9 13.3 8.1 16.5 12.6 Credit to private sector (% change) 19.0 28.9 3.7 14.2 12.8 9.4 16.4 17.0 Investment and savings Gross investment (% of GDP) 22.3 27.3 24.6 24.5 25.7 22.0 22.0 21.9 National savings (% of GDP) 11.0 12.5 13.9 15.7 18.2 15.5 16.0 18.5 Public Finance Government revenue and grants (% of non-oil GDP) 1/ 30.3 25.9 26.7 27.7 27.4 25.7 24.6 24.9 Oil 2.9 2.2 1.8 1.4 1.2 1.1 1.0 0.9 Non-oil 24.9 22.9 24.1 25.1 24.9 23.6 22.6 22.9 Tax 2/ 15.8 14.4 14.1 16.3 14.9 15.9 15.3 15.8 Nontax 9.2 8.5 10.0 8.7 10.0 7.7 7.2 7.2 Grants 2.4 0.8 0.8 1.2 1.4 1.1 1.1 1.0 Government expenditure (% of non-oil GDP) 32.0 32.9 32.1 29.7 30.3 32.3 32.3 32.2 Current expenditure 23.4 25.9 24.6 21.5 22.2 24.7 24.7 24.7 Capital expenditure 3/ 8.6 7.0 7.4 8.2 8.1 7.6 7.6 7.6 Overall balance including grants (% of non-oil GDP) -1.8 -7.0 -5.4 -2.0 -2.8 -6.6 -7.7 -7.3 External debt Debt outstanding to GDP ratio: PPG (%) 82.9 80.8 93.5 67.9 47.5 52.6 52.9 52.1 Memo items Exchange rate (LCU/US$, end-period ) 252.2 261.5 262.0 282.0 .. .. .. .. Nominal GDP (LCU billion ) 734 854 794 997 1147 1186 1226 1268 Nominal non-oil GDP (LCU billion ) 679 797 758 953 1096 1152 1212 1254 Nominal GDP (US$ million ) 2822 3536 3031 3629 3999 3992 4014 4058 Population (million ) 3.1 3.1 3.2 3.2 3.3 3.3 3.4 3.5 GDP per capita (US$ ) 918 1123 940 1141 1228 1198 1176 1161 Nonoil direct taxes (lcu billion) 33 35 35 43 50 54 58 64 Nonoil direct taxes (% non-oil GDP) 4.9 4.4 4.6 4.6 4.6 4.7 4.8 5.1 34 Table 0.14 : Medium Term Macro-Economic Framework, 2007-2014 – Scenario with Lower Iron Ore Prices. 2007 2008 2009 2010 2011 2012 2013 2014 Est Projected National income and prices Real GDP growth (%) 1.0 3.5 -1.2 5.2 5.1 5.7 6.3 5.7 of which: Nonoil 5.9 3.9 -1.1 5.7 5.3 5.8 6.4 5.8 GDP deflator (% change) 0.2 12.4 -5.9 19.4 9.4 5.9 1.3 -0.8 Non-oil GDP deflator (% change) 13.4 13.0 -3.9 19.0 9.2 6.2 1.6 0.0 External sector Exports of goods (nominal; % change) 2.5 27.5 -23.7 52.0 -1.7 24.7 3.2 0.0 of which: Nonoil 45.2 37.5 -21.6 55.0 0.8 28.2 3.9 0.2 Imports of goods (nominal; % change) 36.7 21.8 -27.2 39.1 27.5 10.7 -0.9 -0.4 Current account balance (% of GDP) -18.3 -14.8 -10.7 -8.7 -27.8 -22.1 -17.8 -15.1 Gross international reserves (US$ million) 208.8 194.9 238.0 287.3 395.4 442.3 493.8 515.3 Gross international reserves (months of imports goods) 1.6 1.2 2.1 1.8 1.9 1.9 2.2 2.3 Money M oney and quasi-money (% change) 18.9 13.7 15.2 12.9 13.3 13.4 15.3 13.0 Credit to private sector (% change) 19.0 28.9 3.7 14.2 13.0 11.5 12.7 12.7 Investment and savings Gross investment (% of GDP) 22.3 27.3 24.6 24.5 25.7 27.3 27.0 26.3 National savings (% of GDP) 11.0 12.5 13.9 15.7 -2.2 5.2 9.2 11.2 Public Finance Government revenue and grants (% of non-oil GDP) 1/ 30.3 25.9 26.7 27.7 25.5 24.5 24.4 24.8 Oil 2.9 2.2 1.8 1.4 1.2 1.1 1.0 0.9 Non-oil 24.9 22.9 24.1 25.1 23.0 22.4 22.5 23.0 Tax 2/ 15.8 14.4 14.1 16.3 13.0 14.8 15.1 15.7 Nontax 9.2 8.5 10.0 8.7 10.0 7.6 7.4 7.2 Grants 2.4 0.8 0.8 1.2 1.4 1.0 1.0 0.9 Government expenditure (% of non-oil GDP) 32.0 32.9 32.1 29.7 30.3 29.9 28.5 28.0 Current expenditure 23.4 25.9 24.6 21.5 22.2 20.4 19.3 18.8 Capital expenditure 3/ 8.6 7.0 7.4 8.2 8.1 9.4 9.2 9.2 Overall balance including grants (% of non-oil GDP) -1.8 -7.0 -5.4 -2.0 -4.8 -5.4 -4.1 -3.2 External debt Debt outstanding to GDP ratio: PPG (%) 82.9 80.8 93.5 67.9 56.3 78.7 92.5 100.5 Memo items Exchange rate (LCU/US$, end-period ) 252.2 261.5 262.0 282.0 .. .. .. .. Nominal GDP (LCU billion ) 734 854 794 997 1147 1284 1382 1450 Nominal non-oil GDP (LCU billion ) 679 797 758 953 1096 1232 1331 1409 Nominal GDP (US$ million ) 2822 3536 3031 3629 3999 4322 4526 4642 Population (million ) 3.1 3.1 3.2 3.2 3.3 3.3 3.4 3.5 GDP per capita (US$ ) 918 1123 940 1141 1228 1296 1326 1328 Nonoil direct taxes (lcu billion) 33 35 35 43 29 45 55 66 Nonoil direct taxes (% non-oil GDP) 4.9 4.4 4.6 4.6 2.7 3.6 4.2 4.7 Source: IM F/WB staff estimates and projections 1/ including net revenue from oil. 2/ 2010 includes tax arrears from national electricity company. 3/ including net lending A/ Iron ore prices drop by 50% in 2011; rise by 40% in 2012, 5% in 2013 and 5% in 2014. 35 Table 0.15 : Medium Term Macro-Economic Framework, 2007-2014 – Scenario with tax reform Est Projected National income and prices Real GDP growth (%) 1.0 3.5 -1.2 5.2 5.1 5.7 6.3 5.7 of which: Nonoil 5.9 3.9 -1.1 5.7 5.3 5.8 6.4 5.8 GDP deflator (% change) 0.2 12.4 -5.9 19.4 9.4 5.9 1.3 -0.8 Non-oil GDP deflator (% change) 13.4 13.0 -3.9 19.0 9.2 6.2 1.6 0.0 External sector Exports of goods (nominal; % change) 2.5 27.5 -23.7 52.0 37.5 7.4 -2.4 -4.8 of which: Nonoil 45.2 37.5 -21.6 55.0 46.5 8.2 -2.4 -5.0 Imports of goods (nominal; % change) 36.7 21.8 -27.2 39.1 27.5 10.7 -0.9 -0.4 Current account balance (% of GDP) -18.3 -14.8 -10.7 -8.7 -7.5 -9.2 -9.6 -10.9 Gross international reserves (US$ million) 208.8 194.9 238.0 287.3 395.4 442.3 493.8 518.8 Gross international reserves (months of imports goods) 1.6 1.2 2.1 1.8 1.9 1.9 2.2 2.3 Money M oney and quasi-money (% change) 18.9 13.7 15.2 12.9 13.3 13.4 15.3 13.0 Credit to private sector (% change) 19.0 28.9 3.7 14.2 12.8 11.7 12.8 12.3 Investment and savings Gross investment (% of GDP) 22.3 27.3 24.6 24.5 25.7 27.3 27.0 26.3 National savings (% of GDP) 11.0 12.5 13.9 15.7 18.2 18.1 17.5 15.5 Public Finance Government revenue and grants (% of non-oil GDP) 1/ 30.3 25.9 26.7 27.7 27.4 26.0 25.4 25.4 Oil 2.9 2.2 1.8 1.4 1.2 1.1 1.0 0.9 Non-oil 24.9 22.9 24.1 25.1 24.9 23.9 23.5 23.6 Tax 2/ 15.8 14.4 14.1 16.3 14.9 16.1 16.0 16.3 Nontax 9.2 8.5 10.0 8.7 10.0 7.8 7.4 7.3 Grants 2.4 0.8 0.8 1.2 1.4 1.0 1.0 0.9 Government expenditure (% of non-oil GDP) 32.0 32.9 32.1 29.7 30.3 29.9 28.5 28.0 Current expenditure 23.4 25.9 24.6 21.5 22.2 20.4 19.3 18.8 Capital expenditure 3/ 8.6 7.0 7.4 8.2 8.1 9.4 9.2 9.2 Overall balance including grants (% of non-oil GDP) -1.8 -7.0 -5.4 -2.0 -2.8 -3.9 -3.1 -2.6 External debt Debt outstanding to GDP ratio: PPG (%) 82.9 80.8 93.5 67.9 47.5 48.6 52.6 54.5 Memo items Exchange rate (LCU/US$, end-period ) 252.2 261.5 262.0 282.0 .. .. .. .. Nominal GDP (LCU billion ) 734 854 794 997 1147 1284 1382 1450 Nominal non-oil GDP (LCU billion ) 679 797 758 953 1096 1232 1331 1409 Nominal GDP (US$ million ) 2822 3536 3031 3629 3999 4322 4526 4642 Population (million ) 3.1 3.1 3.2 3.2 3.3 3.3 3.4 3.5 GDP per capita (US$ ) 918 1123 940 1141 1228 1296 1326 1328 Nonoil direct taxes (lcu billion) 33 35 35 43 50 62 67 74 Nonoil direct taxes (% non-oil GDP) 4.9 4.4 4.6 4.6 4.6 5.0 5.0 5.3 Source: IM F/WB staff estimates and projections 1/ including net revenue from oil. 2/ 2010 includes tax arrears from national electricity company. 3/ including net lending. A/ Non-oil direct tax collections are projected to increase by 5% in 2012; 3% in 2013 and 2% in 2014 above their nominal levels in the base case. 36 CHAPTER 2: Analysis of Public Financial Management 2.1. This chapter analyses the mechanism and functioning of the public financial management (PFM) system with a view to updating public expenditure review in the Islamic Republic of Mauritania (IRM). The analysis will be based on the different evaluations and diagnoses of the PFM system carried out in recent years by different international partners. This chapter will focus on public finance reforms in connection with the shortcomings and findings of the analysis made in chapter I above, and not on all the shortcomings of the Mauritanian PFM system that would be brought out clearly in a PEFA evaluation. Part I of this chapter will review recent developments in public financial management reforms as well as the underlying legal and institutional framework, while Part II will make a diagnosis and recommendations on the specific issues relating to the programming and execution of the budget and public accounting. Part III will examine the internal and external control of Government expenditures. I. DEVELOPMENTS IN PUBLIC FINANCIAL MANAGEMENT REFORMS Legal and Institutional Framework 2.2. The legal and institutional framework governing public finance in the Islamic Republic of Mauritania is built around the Constitution of 1991 as amended in 2006, and mainly on the Public Finance Organic Law No. 78-011 of 19 January 1978 ; Order No. 89-012 of 23 January 1989 on government public accounting amended in 2006; Law No. 93-19 of 26 January 1993 on the Audit Court; Decree No. 2005-122 of 19 September 2005 creating a State General Inspectorate (SGI); and Law No. 2010 on the Public Procurement Code. This legal framework lays down the rules governing the preparation and voting of budget laws, their execution and control. 2.3. The Ministry of Finance and the Ministry of Economic Affairs and Development jointly supervise the preparation of the Government budget. The 2004 and 2006 Public Expenditure Reviews had highlighted the problems posed by the existence of two budget documents drawn up and executed by distinct units. The Government budget comprises two documents: (i) the General Government Budget (GGB) drawn up by the Directorate General of the Budget (DGB) of the Ministry of Finance and (ii) the Consolidated Investment Budget (CIB) drawn up jointly by the Ministry of Finance and the Ministry of Economic Affairs and Development. Only the General Government Budget is voted by Parliament. The CIB is attached as informational annex to the GGB. 2.4. Parliament intervenes downstream in the process with the adoption of the Draft Budget Law after a favourable opinion of the Council of Ministers. It further checks the execution of the budget with the adoption of the Budget Review Act. The Audit Court exercises external control of the draft Budget Review Act under its assistance to Parliament. Budget adjudication is carried out by the Ministry of Finance and the Ministry of Economic Affairs and Development. 37 2.5. In light of current public financial management practices, some aspects of the framework described above are outdated. In particular, the above-mentioned Public Finance Organic Law provides for outdated programme laws that have never been implemented, and does not provide for multi-year budgeting and program based budgeting. Consequently, it should be revised considering the current challenges of good governance and transparency in public financial management. To avoid implementation bottlenecks, the future Public Finance Organic Law should not contain details so as to allow for subsequent amendments through regulatory procedures. 38 Table 2.1 : Difficulties in coordinating the implementation of PFM reforms Since December 2004, the Islamic Republic of Mauritania has been involved in a program reforming its public finance management system, with the support of its international partners. An action plan of the priority reforms was elaborated (2005) in order to implement the recommendations from the evaluations of the public finance management system27 and from comparisons to the standards on the fiscal transparency. 28 Several domains were covered including taxation, payment authorization stage of public expenditure, financial control, programming and execution of the budget, public accounting, and development of tools for macroeconomics framework and analysis, and recently public finance management. After the return to Constitutional order in July 2009, authorities reaffirmed their will to continue the reforms in with the support of the technical and financial partners in the framework of the government’s three-year Program (2010-2012) concluded with the IMF in February 2010, the PRECASP financed by the World Bank and support by the European Commission. This drive for reform translated into a number of initiatives to elaborate action plans for the reform of the public finance management system since end 2008 but especially since end 2009. These action plans were essentially action plans elaborated by the four large general directions of the ministry of the finance (MoF). 29 A lack of coordination characterizes the elaboration of the various action plans within the ministry of the finance, between the MoF and the MAED, and between the ministries and the Audit Court. This lack of synergy makes for an inefficient progress in the reform process. This situation led the authorities to subscribe to an October 2010 European Union technical assistance mission to develop an integrated guiding framework that would result in an operational implementation plan. The axes and components of the integrated guiding framework focus on the following priorities: 30 ¾ Multiyear budgeting able to connect the objectives of public policy with the financial capacities of the State; ¾ A budgeting by programs for a better clarification of public policies and improved budget performance; ¾ An effective resource mobilization process that contributes to obtaining the necessary resources to finance the expenditures; ¾ A classification of the expenditures that responds to the need for management and for analysis; ¾ Budget execution procedures that allow risk management and give reasonable assurance to reach the objectives of the administration; ¾ A process of decentralization reinforced by a greater autonomy of the local communities; ¾ A better supervision of the public establishments in accordance with the principles of transparence, governance and responsibility; ¾ A reliable and secure public financial information system; ¾ An accounting system that produces regular financial reporting and is representative of the activity of the administration; ¾ Effective supervision and strengthened external control. The integrated guiding framework envisions piloting the reforms at two levels: (I) an inter-ministerial committee that includes the policy making organs to define the main policy orientations; (ii) and coordination structures with technical characters to support reform implementation. 27 World Bank Report on “Country Financial Accountability Assessment� (CFAA), November 2002. PEFA – PFM performance Report – March 2008. 28 IMF “Report on the observation of standards and fiscal visibility codes in Mauritania� (ROSC), December 2002. 29 The Directorate General of the Budget (DGB), the Directorate General of the Treasury and Public Accounting (DGTPC), the Directorate General of Taxation (DGT), and the Directorate General of Customs (DGC) 30 Under the Budget Support of the EU, the DUE concluded three conditions of eligibility with the Government of Mauritania: (i) signing of a programme with the IMF; (ii) completion of CSLP, and (iii) preparation of a master plan for Public Finance Management reforms. 39 II. BUDGET PLANNING, EXECUTION AND ACCOUNTING A. Budget Planning a) A non policy-based budgeting 2.6. There is a disconnect between PRSP, MTEF, PIP and budget law. The budget document submitted to the National Assembly for approval and adoption does not reflect government objectives and policies. The budget process is therefore not focused on results. Thus, it can hardly address the priorities laid down in the PRSP in terms of optimal allocation of expenditure and efficiency. Yet, coherence and alignment between the Poverty Reduction Strategic Framework (PRSP), the Public Investment Programme (PIP), and the MTEF constitute a guarantee for the credibility of the budget process. 2.7. The establishment of a Program Budget should be seen as a medium-to long-term objective to improve the efficiency of the budget process. Indeed, program budgets make alignment between PRSP, PIP and MTEF possible, and were therefore set up to link up budgetary allocations to public policy objectives and thus, focus the budget on results. 2.8. The establishment of a Program Budget requires an operational MTEF. Understanding the MTEF approach is an important step in the transition to a results-based budget programming. All preconditions necessary for the effective implementation of this approach have not yet been met. Yet, they are necessary in the budget preparation that may no longer be made on the basis of voted services or new measures, but rather on the basis of (i) personal and pecuniary accountability of the public accountant; (ii) clear definition of programmes, objectives and concepts; (iii) strengthening the responsibility of the programme manager; (iv) transparency in budget information; (v) reorganization of the administrative structure and clear definition of decision-making bodies, and (vi) revision of the accounting system with the introduction of management accounting and accrual based accounting. Some of these measures are being implemented; for example, accrual accounting, but they should be implemented under a clearly defined strategic reform based on a properly described sequencing. b) Medium-Term Expenditure Framework (MTEF) 2.9. The shortcomings highlighted in Chapter I on the disconnect between budget planning and budget execution require to move gradually toward effective implementation of the Medium-Term Expenditure Framework (MTEF). Multi-year budgeting is an indispensible framework for the implementation of structural reforms that will enable the allocation of the necessary funds to finance government priority actions. It enables spending ministries to have more visibility in the implementation of their projects. Multi-year budgeting further enables the ranking of priorities on the basis of the financial means of the Government. 2.10. However, recognizing the limited capacity at the national level and the challenge inherent in developing and making an MTEF functional, this analysis will focus on practical recommendations to help the country move gradually to the adoption of the MTEF approach in the medium term. Program budgeting is only recommended in the 40 medium to long run, once Mauritania has addressed fundamental PFM reforms and has a functional MTEF. 2.11. Failed efforts to implement global and sectoral MTEFS since 2002 point to lack of coordination and capacity. Between 2002 and 2004, eight sectoral MTEFs were drawn up for the following sectors: education, health, rural development, transport infrastructure, urban development, energy, fishing, as well as water and sanitation. This sectoral budget programming exercise was useless without a global MTEF to determine the macro-budgetary framework and define the sectoral allocations. In order to address this situation, the first global MTEF was drawn up by the Mauritania Policy Analysis Center (CEMAP)31 in 2003 covering the period 2004-2006, and the second was drawn up in 2007 and covered the period 2008-2010. The 2008- 2010 global MTEF was finalized in October 2007; too late to be used to prepare of the 2008 budget, one of the main goals of the MTEF. The Ministry of Economic Affairs and Development had planned to draw up a global MTEF covering the period 2010-2012 that would have underpinned the preparation of the 2011 and 2012 annual budgets. However, this did not materialize. 2.12. The latest global MTEF was well-developed and of good quality, and was validated technically but not implemented. The MTEF process in Mauritania can be conceptually placed at the stage 1 (MTEF 1) of the coding tree for MTEFs.32 2.13. The setting up of an efficient MTEF steering mechanism is an essential requirement for its success. This mechanism should be based on a system that will make it possible to (i) produce a summary of key MTEF variables: present value of the main macroeconomic and sectoral indicators, pace of execution of programmes for all ministries, status of mobilization of external resources , etc.; (ii) facilitate a deeper understanding of performances broken down into axes, sectors, and major programmes; and (iii) highlight the obstacles, translated into recommendations, and monitor their effective implementation. This process requires the training of a critical mass of actors directly and indirectly involved in the drawing up and implementation of the MTEF. The PRSP steering mechanism may further help monitor the MTEF. 2.14. The MTEF monitoring system will thus help fill the gap between development policies and their budgetary allocations, and will enable several converging processes. These processes are: public finance reforms (taking the macroeconomic framework into account, budget classification reforms), capacity building in budget adjudication based on performance, the monitoring of results of the different policies, and impact on future policy development process. The system should actively involve technical Ministries in the monitoring of the MTEF by proposing a reporting tool – trend chart. In fact, the core of this system is based on a 31Mauritania Policy Analysis Center (CEMAP – Centre Mauritanien d’analyse des politiques ) 32 The coding tree is a methodology used to categorize the MTEF process. The methodological description includes several stages. Mauritania has a Stage MTEF(1). Indicators for this stage include budget or other reports that contain the government’s medium-term macro-fiscal strategy, macroeconomic and fiscal forecasts, debt sustainability analysis, and/or details of IMF program, PRSP, or EU stability or convergence program target. Also, some spending agency or other reports explaining aggregate and sector expenditure objectives and strategies. But, Mauritania does not have a functional Rolling Multi-year Spending and Revenue Forecasts. 41 comprehensive trend chart that makes it possible In fact, to steer all MTEF programmes through more detailed sectoral trend charts. 2.15. This comprehensive trend chart: (i) facilitates essential sectoral data collection and selection (development of key indicators, strengths and weaknesses of the sector, implementation of priority actions, financial execution and the like) ; (ii) sets up routine data collection from the different technical departments and the projects under their supervision, in ministries where there are none; (iii) provides information on reporting lapses (delay in or failure to produce data) ; and (iv) enables departments officials to have a global vision of the strengths and weaknesses in the sector’s contribution to the objectives of the MTEF. Sectoral trend charts must include relevant and measurable indicators to make it possible to appreciate the coherence between expenditures, activities, and results obtained, and to propose possible measures to improve on performance. Indicators contained in the PRSP should be used as basic reference. 2.16. The effectiveness of the MTEF lies in its full integration in the budget law. The preparation of the annual budget based on an MTEF is of vital importance since the latter is an instrument founded on a medium-term macroeconomic framework and on the Government’s development priorities in connection with the PRSP to ensure multi-year allocation of resources to the different ministerial sectors and institutions. a) The adoption of a rolling three-year global MTEF makes it possible to supervise and strengthen the preparation and execution of the annual budget with visibility over the years and integrates all expenditure be it financed by own resources or external resources. b) In order for the MTEF to be useful, it must not only explain budgetary choices and set priorities; it must also be seen as an element in the preparation of the budget and serve as a basis for annual budgetary conferences. c) Although MTEF allocations should only be indicative, the annual budget preparation exercise should be based on amounts set by the MTEF that must, in the first year, correspond to the amounts established by the budget law. 2.17. To serve as base of the budget preparation process and thus constitute a dynamic instrument, the MTEF must be updated annually. Budget preparation must start well before the preparation of the budget circular. 42 Table 2.2 : Indicative calendar for MTEF and Budget Law preparation (MF/MAED) (Sectoral Ministries) Date Level Level Completion of the budget execution review for ƒ Execution report of action March plan for year n-1 the year n-1 ƒ Revision of the macroeconomic framework ƒ Update of action plan for ƒ Update of annual action plans (objectives, Year n+1, Yn+2 and Y n+3 April and May costs, new actions, …) ƒ Update of sectoral MTEF ƒ Update launch of MTEF and consolidation (costing) of sectoral MTEFs ƒ First arbitration and communication of ƒ Revision of sectoral MTEFs June sectoral multi-year allocations (Beginning) (preliminary) ƒ Submission of updated MTEF to Government June (End) ƒ Pre-arbitration on overall allocations of the n+1 Budget Law July ƒ Completion of intra- Distribution of budget circular sectoral breakdown (Beginning) ƒ Completion of MTEF September ƒ Budgetary Conferences ƒ Completion of draft budget law October ƒ Adoption of draft budget law and MTEF by Government November Submission of the draft budget law to the (Beginning) Parliament with MTEF in annex Start drawing up budget execution report of n January n+1 year Sources: CBMT – October 2007 2.18. Contrary to the practice in the other countries of the sub region33, the decree organizing the Ministry of Economic Affairs and Development34 confers to the Public Investment Planning Department, the duty to draw up the global MTEF in close collaboration with the Ministry of Finance (Directorate General of the Budget, Directorate General of the Treasury and Public Accounting). As a financial planning tool, the global MTEF must be the responsibility of the Ministry of Finance especially to synchronize the above-mentioned process. c) Budget preparation 2.19. The budget of Mauritania35 (Part II of the Budget Law) is a resource budget. It presents all budget resources (tax and non-tax income, debt, etc.) and all budget expenditures by 33 In the countries in the sub-region, the comprehensive MTEF is the responsibility of the ministry of finance, which prepares it in close collaboration with the ministry in charge of economic affairs and the central bank. 34 Article 2 of Decree n° 126-2010/PM defining the responsibilities of the Ministry of Economic Affairs and Development and the organization of the central government and its department. 35 Analysis based on the initial budget law for the 2010 government budget (January 2010) 43 economic classification (including special accounts of the Treasury) in comprehensive tables. In the detailed tables, the budget presents expenditure per heading (ministries), per chapter and sub- chapter (structure and subdivisions of administrative structures or project), and by economic nature of expenditures, even to the smallest nomenclature (part, article, paragraph). 2.20. The budget does not abide by the one-budget principle and this dual system makes it more difficult to understand public policies and the efforts made in the various sectors. The government budget comprises two documents: (i) the general government budget drawn up by the General Directorate of the Budget (DGB) of the Ministry of Finance, and (ii) the Consolidated Investment Budget (CIB), including external funding, jointly drawn up by the DGB and the Ministry of Economic Affairs and Development. The existence of two budget documents drawn up and executed in accordance with different procedures increases the risk of inadequate macroeconomic control, contributes to inefficient allocation of resources, and significantly reduces the coherence of sectoral policies over several consecutive years. Parts of CIB revenues and expenditures operations are not in line the rules of public accounting, which aim to ensure the proper use of public funds. The Ministry of Economic Affairs and Development is unable to efficiently monitor the execution of investment expenditure financed by external resources that are executed through established mechanisms under bilateral agreements with donors. In fact, these resources are deposited in accounts opened in commercial banks and do not transit through the Government Treasury Accounts. The financial execution procedures applied are those established by donors who can control expenditure by ensuring that available funds are allocated properly and used within the prescribed time-limit. 2.21. The unification of responsibilities for the government’s general budget and the investment budget will contribute to greater efficiency in budget execution. Similarly, a data collection mechanism on the execution of investment expenditure on external resources must be established. B. Budget Presentation – Budget Classification 2.22. As shown in the previous chapter, the monitoring of the progress of budgetary allocations based on the administrative classification has been made difficult by the chronic changes and systematic reorganization of Ministries. Thus, it is necessary to adopt and effectively implement a functional classification in accordance with the 2001 IMF statistics manual. Pursuant to Order No. 560/MF/98 of 30 August 1998 that lay down the detailed nomenclature of income and expenditure of the State budget, annex budgets, loan accounts, advance accounts, participation accounts, and earmarked accounts, a budget classification in line with international standards was adopted since 1998. The economic classification was used to present current budget expenditure since 1999. A new nomenclature was drawn up in 2008 that links up economic classification and the Accounting Plan of the State. Although a functional classification was adopted, it was not used in full.36 2.23. The current classification system used for the central administration budget preparation and execution is based on economic and administrative classifications. Income 36 IMF -Fiscal ROSC (2002), page 10. 44 and expenditure budgetary charges are defined by a code made up of figures (chapters and articles for income; heads, chapters, parts, articles and paragraphs for expenditures). This enables the classification of expenditure by nature and economic objective under the administrative classification for each ministry and budgetary entity of the Central Administration. Both the Budget Law and the Annexes of the draft Budget Review Law are presented under economic and administrative classification. The economic classification as used for the 2009 Budget Law and other budget execution reports is, for the most part, in accordance with the GFS international standards. 2.24. The Consolidated Investment Budget (CIB) falls under a sectoral classification (Rural Development, Industrial Development, Territorial Development, Human Resources, Institutional Development, and Multi-sectoral Projects) and sub-sectoral classification. Each sub-sector contains baseline projects, their executing agency, their source and nature of funding as well as the total cost of the project. Besides, the CIB contains a presentation of investment expenditure that conforms to the budget classification provided for by law. Budget annexes that cover Public Enterprise and Administration entities use a very general economic presentation. 2.25. The absence of a functional classification in the preparation and monitoring of the budget also constitute a serious constraint. This functional codification allows better allocation of public resources and the evaluation of the impact of the financing of some functions, especially social sector and poverty reduction programmes. The introduction of the functional classification was initiated in 2001 but has never been used to present the budget law. 2.26. The government software RACHAD is structured to introduce the functional budgetary classification. For the purposes of this review, a functional classification was done manually by gathering information from several data sources. Thus, technical functionalities offered by the RACHAD system must be used. Technically, the codification system still needs to be defined. These issues should be taken into consideration in the evaluation of the RACHAD intended by the Government. C. Expenditure Execution Circuit 2.27. Since 2005, Mauritania has undertaken reforms to improve the public expenditure chain. Nevertheless, major efforts are needed to make the budget execution process more effective. At the beginning of 2005, Mauritania undertook a budget execution reform to gradually transfer expenditure commitment to pilot ministries. Currently, the reform applies to expenditures on goods, a large part of which is included in the common costs managed directly by the Ministry of Finance. The reform focuses on redistributing duties under the budget execution framework with the main authorizing officials who are spending Ministers, strengthening financial control, as well as better monitoring by the General Directorate of Budget (DGB). 2.28. The Annual Initial Budget Planning Document (DAPBI), a intra-annual expenditures forecasting instrument, was not made fully operational and is yet to serve as an instrument for spending under the Government’ cash management plan. In fact, under the reform, each credit manager prepares a DAPBI to anticipate the conditions under which the budget law will be effectively executed. This DAPBI is a prerequisite to the actual allocation of 45 credits to the steering ministry. The Financial Controller approves the DAPBI after (i) checking the coherence of the entire document and the reliability of expenditures forecast and their uses, and (ii) analysing the consequences of the estimated costs on public finance and on the covering of mandatory expenditure. However, this document is only remotely related to the cash management plan of the Government, which is regularly updated at the Treasury. Also, the government still needs to adopt a single treasury account. The Technical assistance provided by the IMF on the management system of the treasury is expected to lead to the adoption of a single treasury account and improved efficiency in the management procedures of the treasury. 2.29. One of the challenges of the budget execution process remains the difficulty of managing the annual budgeting process based on the principle established in the organic budget law. In fact, it has been observed that inadequate expenditure planning results in an increase in payment authorizations at the end of the year. Consequently, there is a rush to consume unused credits and, very often, unexpected budget cuts as discussed in the previous chapter. This phenomenon has been worsened by the impossibility to carry over credits, as well as the use of overriding procedures and of “common expenditures�. 2.30. In Mauritania, the management of multi-year commitments and the impossibility to carry over credit allocations constitute one of the shortcomings of the budget execution process. Although the organic budget law provides for programme authorizations in terms of “higher ceiling of capital expenditures, the commitment of which is authorized by the budget law�, budget presentation in Mauritania does not distinguish between commitment credits (CC) and payment credits (PC). During the voting of the budget law, the CC and PC are merged into a single amount, thus disregarding the multi-year nature of government operations, notably those relating to multi-year investment projects. D. Integrated Public Financial Management System 2.31. The absence of a genuine integrated public financial information system is a handicap to the production of reliable financial information and budgetary data. The financial information system of Mauritania is made up essentially of un-integrated information systems deployed at the Budget, Customs, and tax administration, reducing their scope and the visibility of the budget process. As noted in the previous chapter, data varies from source to source, sometimes in the same Ministry37 and often between Ministries38. At the Customs, the SYDONIA++ monitors the collection of duties in the customs system. At the tax administration, Al Mouhassil monitors the collection of taxes but has no link with the Treasury system Beit El Mal. At the Budget Directorate, the RACHAD system handles budget allocation and commitment without dealing with payment related issues which are being monitored through Beit El Mal, the treasury and accounting system at the Treasury. The Automated Spending Chain Network (RACHAD) and its extension to the Treasury, Beit el Mal, function relatively well; thus, a file for the execution of expenditures may be processed in real time from commitment to payment, including the different levels at the Treasury, thereby enabling the 37 The files of the Budget Directorate not consistent with the files of the Treasury Directorate. 38 The files of the Ministry of Economic Affairs and Development on the CIB entirely different from the files of the Ministry of Finance. 46 compilation of the table of the four phases of budget execution. However, some expenditures are only integrated through a gateway, notably salaries and public procurement, while others, such as investment, are not integrated at all or if so, only partially. 2.32. Efforts to interconnect different systems are underway, but should be completed. The Government of Mauritania, through its PRECASP project financed by the World Bank, is involved in a process to set up an integrated public financial management system (IFMIS) by creating interconnections between the main computer systems for public financial management (RACHAD, Beit el Mal, Central Bank, Customs, Debt, Taxation, and Human Resources). The establishment of the IFMIS, while an ambitious plan, will help improve public financial management through proper communication of data, better integration of actual income receipts to facilitate cash management and modernization of the payment system between the Treasury and the Central Bank of Mauritania (BCM). For example, a service provider would be able to get his payment directly through a transfer from the Central bank; which contributes to improve transparency and efficiency in payment processing. 2.33. The creation of a coherent database on human resources is essential to control and manage budgetary drifts of the wage bill. In this regard, the census of civil servants begun since 2007 should be completed as soon as possible and be accompanied by the establishment of an integrated human resource management (HRM) system that will merge the salary database with the public service database. This HRM system should fit harmoniously in the planned Integrated Public Financial Management System (IFMIS). The setting up of a coherent database should be accompanied by an efficient wage bill management policy, about which suggestions where provided in chapter I. E. Accounting and Reporting 2.34. The monitoring of budget execution requires that regular and prompt information on budget execution be available to enable the Ministry of Finance (and the Government) to monitor the progress and, where necessary, determine the measures to be taken to correct the situation. The country has a unified chart of account and a status of budget execution is prepared monthly by the General Directorate of Treasury and Public Accounting (DGTCP). This is the monthly Government financial statistic (TOFE) available on the Treasury website.39 Budget forecast items are mentioned in the TOFE and include revenue and the main economic expenditure items covered during the payment order phase. The economic classification of data used is in conformity with international standards and allows a comparison at the level of the most significant aggregates. 2.35. The monthly report on the execution of the budget (TOFE) prepared by the DGTCP is available at most four weeks after the month covered. The last TOFE covering the 2010 fiscal year was available in January 2011 in accordance with the provisions in force and with recognized international standards (PEFA). 39 www.tresor.mr 47 2.36. The Budget Review Law, which is composed of financial statements consolidated at the end of the fiscal year, is essential for the transparency of the public financial management system. However, there are major delays in the production of draft budget review laws. In the Islamic Republic of Mauritania, the preparation of the budget review law is the responsibility of the General Directorate of Treasury and Public Accounting (DGTCP) of the Ministry of Finance, whose capacities are limited especially in this area. Law No. 78-011 of 19 January 1978 on the organic budget law governs the preparation and contents of the Budget Review Law. 2.37. The production of the delayed annual draft Budget Review Law should be a priority of the government. The Budget Review Law for 2003 2004, and 2005 was completed in August 2006 and was adopted by Parliament in August 2007.40 The DGTCP also prepared the draft Budget Review Law for 2006. However, the draft Budget Review Laws for 2007, 2008, and 2009 are yet to be drawn up. This situation does not make it possible to ensure that expenditure are carried out as planned and does not guarantee transparency in the budget execution process and Government accountability to Parliament. Consequently, there is a need to take the requisite measures to make up for the delays and submit the draft Budget Review Laws for 2007, 2008, and 2009 to the Audit Court for jurisdictional control. Among other procedures, this is done by establishing a calendar to produce the draft Budget Review Laws and by auditing Outstanding Payments Suspense Accounts, the amount of which stands which stood at 7.257 billion Ouguiyas at end-2009. III. PUBLIC EXPENDITURE EXECUTION CONTROLS A. Internal Control of public expenditure Prior control – Role of the Financial Controller 2.38. In Mauritania, prior internal control is still limited and inefficient. In fact, internal control is carried out mainly by Financial Controllers from Ministries, who do it based on documents only, and cannot make any statement on the relevance of the expenditure or the service delivered. This situation reduces the efficiency of the prior control of expenditures insofar as supporting documents are a necessary, but insufficient condition for the execution of expenditures. Besides, Financial Controllers have no control over the execution of transfers made to public sector entities. Consequently, and to improve efficiency in the prior control of spending, it would be important that (i) credit managers be given full responsibility to determine the relevance of expenditure, (ii) Financial controllers be given more powers to control services delivered, (iii) transfers to public entities be controlled based on the execution of preceding transfers through a reporting mechanism for public entities; and (iv) a performance contract be established between central administration and public entities to connect the execution of capital transfers to the attainment of objectives that are in line with the missions of public sector entities in Mauritania. 40 Between 1967 and 2000, no draft Budget Review Law was drawn up. The draft Budget Review Law of 2001 was drawn up in part. In 2002, the draft Budget Review Law was more complete. 48 Ex-post control 2.39. Lack of efficient coordination of controllers’ interventions affects the performance of institutions that undertake ex-post control. In Mauritania, most internal verification duties are simultaneously carried out by the State General Inspectorate (IGE) under the authority of the Prime Minister’s Office, the General Inspectorate of Finance (IGF) under the supervision of the Ministry of Finance, and Internal Inspectorates attached to different Ministries. Besides, in its capacity as the Supreme Audit Institution (SAI), the Audit Court exercises administrative and management control of public entities. A number of regulatory provisions determine the duties of these bodies and the rules governing audits. In practice, there is very little coordination between these bodies, which are endowed with very limited resources and with overlapping mandates. The establishment of an efficient coordination framework of the missions of the different control bodies will obviously help increase the performance of public expenditure control with optimal coverage of the resources committed by the government. 2.40. There is no systematic monitoring and implementation of recommendations made by control missions. Recommendations drawn by control missions are often not implemented since control bodies merely forward their reports to supervisory structures as provided for by the regulation in force (Ministry of Finance, Prime Minister’s Office, and Presidency of the Republic). Furthermore, criminal law on financial offences needs adaptation. Economic and technological development has not only contributed to the complexity of such offences, but has also caused the emergence of some categories of offences that have so far been ignored by traditional criminal law. These offences include over invoicing, favouritism hidden behind complex procurement schemes, misappropriation of corporate property, and illicit arrangements between beneficiaries of public contracts. Besides, in its annual general report of 2006 published in April 2009, the Audit Court observed that the current form of summons conceived by the law as a precondition for prosecution in matters of misappropriation of public funds hinders referral to the court if it is not issued by the auditors who noticed the offence instead of serving the objective of safeguarding public funds. Therefore, the amendment of the Criminal Code should not maintain summons as a precondition for triggering prosecution. Thus, it is essential to set up an administrative structure in charge of follow-up and coordination, as well as the application of the findings of control missions to guarantee the efficiency and effectiveness of public spending controls. B. Reform of the public procurement system 2.41. The quality of the government procurement system is essential both for economic and budget efficiency reasons, and for ethical reasons, as this aspect of public management is decisive in the perception of the quality of governance. 2.42. Since 2002, the Government of Mauritania has undertaken procurement reforms that resulted in the adoption of the law on the new Procurement Code in 2010. In July 2010, the law establishing the new Procurement Code was promulgated and the Ministerial Council adopted related implementation texts by decree in February 2011. The law is in accordance with internationally recognized principles and requirements of public contracts, and corrects the 49 weaknesses noted in the 2002 Country Procurement Assessment Review (CPAR), including restrictions on the eligibility for public contracts. It also provides for segregation of duties between incompatible procurement functions (regulation, control, award and remedies), including the setting up of an autonomous tripartite authority to regulate public contracts – ARMP – (government, private sector, and civil society) with an appeals body. The next steps for the implementation of the new law are to establish the new structures (ARMP, procurement commissions at the central and ministerial levels) and to implement a training program for new procurement bodies at the level of sectoral Ministries. 2.43. With the implementation of this reform, the Government of Mauritania will effectively embark on the modernization of the public procurement system. As this constitutes significant progress in good governance and the reduction of the often long procurement timeframes thanks to the elimination of the requirements of the Procurement Commission (CCM) endorsement and Prime Minister Approval. The government can further these reforms by addressing other very important aspects that have an impact on public procurement transparency and efficiency and require quick action. These aspects include contract planning, publication of results on evaluation and past contracts, monitoring and evaluation, and general access to contract information. Box 2.1 : Historical developments in procurement reforms in Mauritania Since the year 2000, there have been several procurement reforms in Mauritania. The drafting and adoption of the 2002 Procurement Code have set the foundation of a new regulatory framework that better meets recognized principles and requirements in this area, raising the capacity of sectoral ministries’ procurement commissions at the departmental level (districts). This has enabled them to participate more fully in awarding of public contracts and to improve the capacity to absorb funds by speeding up the procurement process. However, there were several shortcomings in the enforcement of the 2002 Code. The enforcement and prior control structures were often congested because the Procurement Commission (Commission Centrale des Marchés—CCM) was highly-centralized and continued to award most contracts while playing its role in prior control, dispute settlement, and regulation, and functioning with the lack of qualified and motivated staff. Also, some regulatory provisions of this Code were still inconsistent with internationally recognized, fundamental procurement principle. Also, the shortcomings of the 2002 Code included clarity of applicable rules, conditions of eligibility for public contracts, and incompatibility of some functions, especially those regarding regulation, awarding and attribution, control and remedy, which are all handled by the CCM. A second revision of the code was started in 2005 to correct the shortcomings identified in the 2002 updated review (CPAR). The new government guidelines aimed to make the country’s public contracts more attractive to national and international private operators and to reduce the timeframes and costs of acquiring goods and services for the Mauritanian government. The programme of this second reform sought to cover three main aspects: Setting up an institutional framework to handle the separation of the different procurement functions, i.e. regulation, awarding, control and remedy, as well as preparing and adopting enforcement regulations, preparing and publishing standard documents for bidding and for the execution of public contracts; Aligning tax provisions for public contracts with internationally recognized trade and import practices; Capacity and institutional building through the establishment of a permanent system for training and professional development in public contracting, supporting the procurement regulation function– including the monitoring and evaluation of system performance. On the whole, the implementation of this reform suffered significantly from the political situation in recent years and it took until July 2010 for the new law to be promulgated and its regulatory text adopted in February 2010, Nonetheless, authorities tried to move the reforms forward, including by training trainers at the National School of 50 Administration (ENA). IV. CONCLUSION 2.44. The PFM analysis undertaken in this Chapter has highlighted some important weaknesses and recommendations that the government of Mauritania will need to address to make its PFM system more efficient and transparent. However, considering the limited capacity of the government and the need to prioritize the reform agenda, the following actions are suggested as critical steps the government should consider to improve its PFM system. These reforms fit within the integrated guiding framework for PFM reforms the government is adopting. PFM reforms Responsible Time frame Set up the Procurement regulatory Government 1 to 2 years bodies (Central commission, departmental commissions and the procurement regulatory Authority- ARMA) and Make operational the new procurement code Unify the Budget preparation ME, MAED, Prime Minister’ office 1 to 2 Years Documents (General Government Budget and Consolidated Investment Budget) under the full responsibility of the Minister of Finance Draw up a MTEF covering the Minister of Finance/MAED 1 to 3 Years period 2012-2014 that should underpin the preparation of the 2012 and 2013 annual budgets Finalize the interconnection between Minister of Finance (DGB, DGCPT, 1 to 3 Years PFM information system to set up DGD, DGI, BCM) the Government IFMIS Submit the draft Budget Review Ministry of Finance 1 year Laws for 2007, 2008, and 2009 to (DGCPT) the Audit Court for jurisdictional control 51 CHAPTER 3:Evaluation of Public Investment Management in Mauritania Introduction 3.1. This chapter reviews public investment management (PIM), the efficiency of budget allocations and its results in Mauritania between 2004 and 2009. The importance of the role of public investment program cannot be underestimated as it is one of the government’s main tools for attaining PRSP-3 objectives. The PRSP-3 was adopted by the authorities in April 2011 and the authorities now face the challenge of implementing it. The PRSP-3 action plan requires substantial investments. 41 3.2. The chapter is organized as follows. Section 1 reviews the overall performance of the PIP during 2004-2009. Section 2 review changes in the national context and the government’s objectives (PRSP), including the link between the PRSP, the Medium-Term Expenditure Framework (MTEF), and the Public Investment Program (PIP). Section 3 analyses the public investment management system. Section 4 presents conclusions and recommendations to support reforms being implemented by the authorities to improve the public investment management (PIM) system and PIP expenditure efficiency. I. PERFORMANCE OF THE INVESTMENT BUDGET: 3.3. A note of caution is necessary about the data presented in this chapter. The information provided by the authorities is, by their own admission, incomplete, given that in the past some spending was off-budget, while other spending was not properly broken down. The authorities indicate that the available data still has inconsistencies. In addition the preparation, execution or reporting of the general budget (operating and investment) is not consolidated, making its management, execution and analysis difficult. 3.4. It is important to note the following for the study below: ¾ The Consolidated Investment Budget (CIB), prepared by MAED, includes all resources allocated for investment, including all domestic and foreign resources. It therefore includes national resources used for investment; counterpart funds; projects implemented by the central government but financed by partners; investments by public enterprises and entities, as well as the projects of partners implemented by non-government project implementation units. ¾ The finance law only retraces national counterpart funds in the investment budget. The CIB is included as an appendix in its finance law. 41 These issues are discussed in the 2006 Public Expenditure Review as well as in a number of government analyses done in 2009-2010 within the framework of preparations of the Donor Round-table of June 2010 in Belgium as well as within the framework of preparatory work for CSLP-3. 52 ¾ The results reported in the government financial operations table (TOFE) only relate to investments on internal or external funds administered by the central government. ¾ In other words, the government cannot produce complete CIB execution data. There are no systematic budget execution data for investments by businesses and public entities or for projects directly implemented by partners. The data reported for the CIB is payment authorization data. 3.5. The authorities cannot produce a systematic database of the physical results of the annual implementation of their CIB. This reduces the visibility of CIB implementation, preventing even a preliminary analysis of the allocative and execution efficiency of this budget. The problem is partially that of development partners who provide information on their project implementation late or not at all. 3.6. This problem has been exacerbated by the multiple institutional changes that have taken place in recent years in the organization of the ministries in charge of this budget. The change of teams has resulted in a significant loss of information on the effective mobilization, disbursement and monitoring of external resources. 3.7. The authorities are making concerted efforts to enhance the coherence and comprehensiveness of budget implementation by improving the RACHAD budget information system as well as reforming the laws and procedures governing the investment budget and government procurement. Furthermore, the MoF is capable of presenting an expanded TOFE that would include investment execution data if these were available from the MAED (which in charge of monitoring this portfolio). The MAED can and has presented programming and payment authorizations for overall investment expenditures, but not on the actual execution of these budgets. Finally, the MAED has just launched the development of comprehensive software for capturing, monitoring, analyzing and managing all information on the lifecycle of investment projects.42 Additionally, an action plan to address data consistency and reliability should be developed and implemented. C. Trends in Investment Expenditures 3.8. The programming of the overall investment budget more than tripled during the 2004-2009 period. This increase seems to have coincided with the political events of 2005-2006 and 2008-2009. It is underpinned by a five-fold increase in programmed foreign resources (Table 3) to support domestic investments, while investments programmed on domestic resources only doubled (Table 1). Given the political realities of the 2005-2009 period, the realism of this programming based on foreign resources is questionable. In fact, CIB programming of foreign financing is based on mobilization potential and on the status of negotiations with partners (donors). However, there is an element of uncertainty to this programming because it is characterized by (i) unpredictability in the disbursement of these resources, resulting in a low mobilization rate; (ii) limited absorptive capacity; and (iii) often poorly prepared sectoral project portfolios (weak preliminary appraisal, incomplete portfolios, etc.). 42 The software is provided by Synergy International Systems. 53 Table 3.1 : CIB - National and External Funding, in billions of Ouguiya (Central Government). Year Planned (Revised Budget Law) Execution % General National Funding (Government Budget) External Total Funding (NF+EF) Debt Investment Total Amortization 2004 24.97 13.55 38.52 26.28 64.80 43.10 66.51 2005 26.76 16.60 43.36 26.80 70.17 36.60 52.16 2006 33.70 40.39 74.09 40.99 115.08 44.60 38.76 2007 36.56 9.40 45.96 56.17 102.12 58.50 57.28 2008 54.48 8.90 63.38 110.31 173.69 55.60 32.01 2009 69.55 7.62 77.17 130.13 207.30 51.80 24.99 TOTAL 246.02 96.46 342.48 390.68 733.16 290.20 39.58 Source: MF/MAED/IMF (TOFE) and World Bank Calculations. Execution by TOFE. 3.9. The execution rate of the overall investment budget is low (39.6 %) (Table 1). This budget was severely affected by the changing political situation and the recurrence of institutional crises from August 2005 to July 18, 2009, when multiparty elections ended one of the longest crises Mauritania has ever experienced. Other bottlenecks include the changing international economic environment, governance, absorptive capacity, investment budget management (including the capacity to prepare, implement, and monitor and evaluate financed projects), difficulties stemming from disbursement challenges and from the role of development partners’ procedures. Partners’ procedures and timelines may have two un-intended consequences. They impact project implementation and therefore their costs and benefits. In addition, payment delays push authorities to use counterpart funds for items that were not originally planned for, thus hampering the authorities’ efforts to improve budget transparency and ensure accurate execution of the programmed budget. Bottlenecks relating to management of the investment budget and the role of international partners will be discussed later in this chapter. 3.10. An average of 4% of the investment budget (data on payment authorization basis) was diverted to operating expenses over the period (from 2 % in 2008 to 6.8 % in 2009). The share earmarked for operating expenses should be reported in the operational budget to ensure transparency in public spending and to protect the investment budget for real investments. Debt repayment accounts for 8.57 % on average over the period, the country having benefited from the multilateral debt relief in 2006. 54 Figure 3.1 : CIB 2004-2009: Different expenditure categories included in investment spending (Authorized payment basis – in % of total) Source: MAED, 2010. Operating expenses categories include (i) Salaries, Wages and Accessories; (ii) Spending on goods and services and (iii) Subsidies and current transfers. 3.11. Mauritania is dependent on international support in the long run, which also makes it vulnerable to disruptions in this assistance. This is especially true as prospects for national oil revenues have clearly been disappointing. Figure 3.2 shows an increase in international budgetary assistance over the period, after a slight decline in 2005-2006 owing to the political situation. The execution of this support is weak, exceeding only 40 % of the investment budget in 2007, and its sharp decline in 2008-2009 highlights the weakness of this instrument that is central to the country’s development. The Mauritania donor round-table that took place in June 2010 pledged US$3.2 billion, whereas the government aimed to obtain only US$ 2.9 billion. 3.12. Grants, which constituted a significant share between 2004 and 2007, declined in favour of loans during the 2008-2009 periods, for both programmed foreign aid and authorized payments in the wake of the 2008-2009 political events (figure 3.3). In addition, the average authorized payment rate for programmed amounts is much lower for grants (46 %) than for loans (75 %). 55 3.13. Figure 3.2 : CIB 2004-2009: Share of foreign financing in the investment plan (% of total). Source: MAED Programming; Execution by MF and FMI (TOFE) and World Bank calculations Figure 3.3 : Planned foreign aid and authorized payments: 2004-2010 by type of financing (In M of MRO) Figure 3.3a - Programmed foreign aid: Figure 3.3b – Foreign aid authorized payments: 2004-2010 by type of financing (in M of MRO) 2004-2010 by type of financing (in M of MRO) Sources: MAED 56 3.14. The CIB financed through foreign funds and executed by the central government (and thus captured in the TOFE) had a low average disbursement rate (41.8 %) over the period (Table 3.2). The mobilization of foreign financing for the 2004-2009 periods was beset by political crises and absorptive capacity issues. While programming rose steadily, execution dropped steadily (from 70 % in 2004 to 11.5 % in 2009). The low CIB mobilization rate puts constraints on development implementation and compromises efforts to reach its MDG commitments. Table 3.2 : 2004-2009 CIB: Foreign financing (executed by the central government) In billions of Ouguiya Year Planned Executed % 2004 26.28 18.50 70.4 2005 26.80 15.20 56.7 2006 40.99 20.70 50.5 2007 56.17 28.60 50.9 2008 110.31 12.00 10.9 2009 130.13 14.90 11.5 TOTAL 390.67 109.90 41.8 Source: MAED programming data, IMF implementation data and World Bank calculations. 3.15. The programming of the CIB financed domestically has also increased steadily over the period (Table 3.3). This budget mainly represents counterpart funds to foreign financing. The increases in programming in 2006 and again 2008 and 2009 are driven by political crisis and the consequences decline in external funding. The 2006, political crisis and the beginning of petroleum production led to a higher CIB. In the case of 2008-2009, the Mauritanian authorities increased internal resource mobilization for investment to compensate for the impact of the freeze in foreign finance and programs. Thus, several infrastructure projects (roads, health centers) were financed and implemented. These two episodes of boosting domestic financing for investment achieved results because the government had an unique combination of domestic resources at hand: (i) the exploitation of petroleum resources had started from February 2006, (ii) Mauritania’s eligibility for the Multilateral Debt Relief Initiative in October 2006, which created new fiscal space, (iii) the good performance of the mining sector, which is driven by iron ore exploitation by SNIM43, but has also engaged in gold and copper production by Tasiast Mauritania limited and MCA, respectively, and finally (iv) a 20% cut of the recurring budgetary estimates of 2009, which was transferred to the government’s priority investment projects. 3.16. On the other hand, the amounts executed are more variable overtime compared to programming. Execution reached 54.5 % on average of the programmed budget during the 2004-2009 periods, with peaks in 2007 and 2008. Aside from the political issues facing the country, the lack of national absorptive capacity and weaknesses in are obvious constraint. 43 SNIM is a mixed ownership company, with 78.25 % of its capital is controlled by the government 57 Table 3.3 : 2004-2009 CIB, Execution - Domestic Financing (In billions of Ouguiya) Execution Year Programmed (Revised Budget Law) Total As % of Investment Debt Amortization Total Programmed Budget 2004 24.97 13.55 38.52 24.60 63.86 2005 26.76 16.60 43.36 21.40 49.35 2006 33.70 40.39 74.09 23.90 32.26 2007 36.56 9.40 45.96 29.90 65.06 2008 54.48 8.90 63.38 43.60 68.79 2009 69.55 7.62 77.17 36.90 47.82 TOTAL 246.02 96.46 342.48 180.30 54.52 Source: Finance Ministry. 3.17. Aid granted to Mauritania between 2005 and the first quarter of 2010 was mostly used to finance projects in the territorial development (63.61 % of total aid over the period), human resources, and rural development sectors. This is justified by the need to develop basic infrastructure that cannot be financed through private funding, improve the level of human resources, and help strengthen the rural development sector, one of the sectors with growth and job-creation potential. D. Sectoral analysis of the implementation of the 2004-2009 consolidated investment budget 3.18. Table 3.4 summarizes the sectoral distribution within CIB between 2004 and 2009 (at the payment order stage, all types of funding included, in percent of total). The sectors are presented in five main categories, namely, rural development, industrial development, territorial development, human resources, institutional development and multi-sector projects. This last category is not broken down and therefore does not lend itself to any analysis, although it accounts for an average of 16.4 % of the total investment budget allocated for the 2004 - 2009 period. To a large extent, the sometimes erratic changes in figures reflect political instability. 58 Table 3.4 : 2004-2009 Consolidated Investment Budget Sectoral Distribution (% of total scheduled budget, all types of funding included) S EC TEUR/S O US -S EC TEUR A 1. DEVELO PPEMENT RURAL AGRICULT URE GENIE RURAL HYDRAULIQUE VILLAGEOISE ET P AST ORALE ELEVAGE ENVIRONNEMENT RECHERCHE, FORMAT ION ET VULGARISAT ION 2. DEVELO PPEMENT INDUS TRIEL P ECHE MINES (hors SNIM) INDUST RIE, ART ISANAT ET T OURISME & COMMERCE ENERGIE 3. AMENAGEMENT DU TERRITO IRE HABIT AT , URBANISME, T RANSP ORT HYDRAULIQUE URBAINE INFRAST RUCT URES ROUT IERES INFRAST RUCT URES P ORT UAIRES INFRAST RUCT URES AERIENNES INFORMAT ION T ELECOMMUNICAT IONS 4. RES S O URC ES HUMAINES CULT URE, JEUNESSE ET SP ORT S JUST ICE EDUCAT ION SANT E ET AFFAIRES SOCIALES ALP HABET ISAT ION EMP LOI 5. DEVELO PPEMENT INS TITUTIO NNEL AP P UI A L’ADMINIST RAT ION AP P UIAU SECT EUR P RIVE 6. PRO JETS MULTIS EC TO RIELS TO TAL (h ors S NIM) 6. S NIM TO TAL G ENERAL TO TAL INVS T B UDDG ET /PIB (%) EDUC ATIO N EXPENDURE / PIB (%) S ANTE ET AFFAIRES S O C IALES /PIB (%) AMENAG EMENT DU TERRITO IRE/PIB (%) Source: World Bank calculations based on figures provided by MAED. 3.19. Territorial development, which includes major infrastructure projects (roads, air, ports and airports), receives the largest share of resources with an average of 44.1 % of the total investment budget for the period. Road infrastructure accounts for the lion’s share with 15.2 % of the total CIB amount. Overall, this share remains remarkably constant throughout the period despite the political instability mentioned above. This sectoral choice is in line with the political decision of the past 3-4 years to focus on infrastructure in order to facilitate growth and job-creation potential. Human resources come a distant second, with 14.1 %, followed by rural development 12.7 %. The housing, urban planning, and transport, as well as urban water supply 59 sectors account for a substantial part of the investment budget (12.8 % and 12.7 %, respectively, of the budget); in line with the importance the new government places on the housing needs of the population. Investments in urban water supply rose from 3.2 % in 2006 to 16 % in 2007 and 40 % in 2008. This is attributable to the “lumpy� nature of infrastructure investments in general, where large amounts are spent within a relatively short time. Also, the launching of the US$470 million multilateral Aftout Es Saheli project had an impact on the results. The significant decline in the sector in 2009 is due to the suspension of external financing after the August 2008 coup d’état. 3.20. The results on the Human development (HD) category are disappointing and require government’s attention. HD accounts for 14.1 % on average of the authorized investment budget; however, spending has declined consistently since 2004. The group accounted for 16.6 % of the budget in 2004 and only 9.2 % in 2009, with a short-lived recovery in 2006 (20.8 %). In fact, a review of the sub-groups with HD category reveals variability in annual expenditure. The education sector accounts for a modest share of the total amount (6.9 % of the budget), but a small percent of GDP (0.7 %). Although it benefits from multi-donor sectoral support, the annual execution rate ranged from 13.1 % (2006) to 1.9 % (2009). Health accounted for 3.3 % of the authorized budget and 0.3 % of GDP. These weak results underscore the lack of a strategic vision for the sector, the challenges of undertaking sectoral investments, and the consequences of political events. 3.21. The above analysis suggest three specific measures to minimize risks associated with donor support: i. Reclassify in the operational budget operating expenses that are still included in the national investment budget. This could free up an average 4% of the investment budget and contribute to budgetary transparency. This would require: • Consultation with MoF and MAED and sectoral ministries to reprogram the expenditures appropriately. • Adopt a more conservative approach to programming investment based on foreign finance so as not have to rely on use of counterparty funds to cover expenditures when foreign finance does not materialize or materializes late. ii. Implement the government’s integrated public finance reform program so that eventually the projects managed outside national processes will be incorporated into the national process (Paris Declaration). In the short term, this could focus on aspects of procurement (implementing the new law), quality control during project preparation and results reporting. iii. Given the importance of external funding in the investment budget, adopt multilateral budgetary reviews of the investment budget which could ease bottlenecks in the implementation of projects and inform the preparation of next year’s PIP. To avoid 60 parallel efforts, the review could be carried out under the direction of the government/donor committee responsible for implementing PRSP-3, which met for the first time in February 2011. This committee is supposed to meet on a quarterly basis and could therefore coordinate efforts to monitor the investment budget. It is recommended to conduct budgetary reviews of the investment budget during the two IMF annual visits to ensure coherence between the macroeconomic framework and the investment plan. II. CHALLENGES TO ACHIEVING PRSP-3 OBJECTIVES 3.22. This section briefly reviews challenges that affected the performance of PRSP-2 in view of the upcoming implementation of PRSP-3. 3.23. Although successive governments that led Mauritania since 2000 have all expressed their commitment to national development objectives through two PRSPs, the results achieved have been disappointing. Two PRSPs were designed during the decade: PRSP-1 in 2000 (covering 2001- 2004) and PRSP-2 in 2006 (covering 2006-2010). Although progress has been made in combating poverty and improving living conditions, overall, the results of the implementation of the two preceding PRSPs were not satisfactory as shown in the 2006 Public Expenditure Review (PER) as well as in a number of government analyses carried out in 2010 (Also see below). 3.24. The weak performance of PRSP-2 (2006-2010) is attributable to several factors. The poor performance of the oil sector, political events in the country, and external shocks—oil, food and then financial— from 2007 to 2009 partially explain why PRSP-2 outcomes were not achieved. Other factors also affected the strategy’s implementation. Although PRSP 2 tied in well with budgetary tools (budget, PIP, MTEF) which formed the basis of the documents presented to the advisory group in December 2007, its implementation was affected by the lack of practical coordination between different actors that had these budgetary tools. In addition, as section two shows, from 2004-2009, there were significant weaknesses in the implementation of the PIP. Also, the weakness of the statistical information and planning systems for the supervision, monitoring and evaluation of actions affected the strategy’s implementation. PRSP-2 had a broad and confusing matrix of indicators which complicated the monitoring process.44,45 3.25. The disconnect between the PRSP, the MTEF, the PIP, and the Finance Law (FL) is one of the critical factors responsible for this poor performance and should be addressed for PRSP-3 to succeed. Chapter two discusses the disconnect between these budgetary tools and the critical role of an MTEF, which can be used to translate the PRSP into a Public investment Program (PIP), Finance law, including result-based consolidated annual investment budgets. However it is important to briefly deepen the analysis of PRSP-2, PIP and MTEF and the disconnect among them. 44 The authorities recognize, in several analyses, the disconnect between the country’s development objectives (PRSP 1 and 2, MTEF, sectoral strategies) and investment budgets. They also note that the drafting of past PRSPs and MTEF was an academic exercise unrelated to items on annual general budgets and investment budgets. See PRSP 3- Volume 1: an assessment of PRSP2. 45 Ibid. 18. 61 Box 3.1 : Definitions and links: PRSP, MTEF, PIP, budgeted CIB and executed CIB. 1) The PRSP: Poverty Reduction Strategy Paper. The country has adopted PRSP-3 (2011-2015) in April 2011. It includes the same objectives as PRSP-2, as well as a more elaborate monitoring and evaluation component. 2) MTEF: Medium Term Expenditure Framework: a document which summarizes overall expenditure, including both operating and investment for a rolling three-year period (comprehensive MTEF). They are then broken down by spending by sector (sectoral MTEF) which show how various ministries intend to implement government policies and achieve their goals. 3) The PIP comprises all projects that Mauritania would like implemented. The projects originate from sectoral departments, finance ministries, or the presidency. 4) Budgeted CIB: in theory, the first year of the PIP is included in the CIB, which includes all investments. In practice, the PIP is prepared by the MAED, but the CIB remains with the Directorate of Programming (DP) in the Ministry of Finance (the DP being the relic of efforts to merge the two ministries in 2008). Also some projects have been included in the CIB whereas they are not in the PIP. 5) Executed CIB: Projects that are eventually implemented. They may be different from the CIB included in the budget due to natural emergencies or political requests raised during the year. 3.26. The objectives of the 2006-2010 PRSP included: 1. Reduce the incidence of poverty from 46.7 % in 2004 to less than 35 % in 2010; 2. Reduce the incidence of poverty in rural areas from 59 % in 2004 to less than 51 % in 2010; 3. Reduce the incidence of poverty urban areas from 28.9 % in 2004 to less than 15 % in 2010; 4. Ensure an annual average growth rate of 9.4 % during the period; 5. Provide universal access to quality basic education for at least 9 years; 6. Reduce the illiteracy rate among adults aged over fifteen years to less than 20 %; 7. Increase primary health care coverage within a 5km radius to 90 %; 8. Reduce infant mortality rate (IMR) to less than 60 %, the juvenile and infant mortality rate to less than 70 % ,and the maternal mortality rate to less than 400 per 100,000; 9. Increase the rate of water supply in urban areas to 48 %; 10. Increase the rate of water supply in rural and semi-urban areas to 62 %; 11. Increase food self-sufficiency. 3.27. Most of these ambitious objectives were not achieved. The country witnessed progress in the social sectors and there are expectations for it to attain several of the Millennium Development Goals (MDGs), but other MDGs are far from being attained (See Chapter 1). Also, a recent evaluation of PRSP-2 by the government, conducted within the framework of the Donor Round-table of June 2010 in Brussels, shows that growth projections between 2006 and 2010 were very optimistic, most likely due to the promise of petroleum production. In fact, between 2006 and 2010, the economy grew at an annual rate of 3.3 % per year (excluding oil) and 3.8 % 62 per year (including oil) whereas an average growth rate of 9.4 % per year had been forecasted. The actual growth was mainly driven by the tertiary sector comprising transport and telecommunication, trade, restaurants and hotels and other private services, whose contribution to growth during this period was 1.6 %, or close to half of overall average growth (3.7 %). (See annex for breakdown of proposed annual allocations by sector). 3.28. There are questions about the methodologies and assumptions used for the development of both PRSP-2 and PRSP-3. For instance, what is the underlying basis of the objectives proposed in the PRSP, the total amounts programmed and their sectoral distribution? In what way is this breakdown “optimal�? What is their link with the growth rate proposed in the objectives? How will this distribution boost growth to a level sustainable level? If there is no analytical link between the figures proposed and the objectives sought, as seems to be the case, of what use are the amounts proposed, given that their financing remains hypothetical and their possible mobilization could depend on donors? Also, how realistic are the objectives of PRSP-2 which look more like a “wish-list�? These questions underscore the weaknesses of the connection between PRSP-2 with the economic forecast, the comprehensive MTEF and sectoral MTEF.46 3.29. PRSP-2 is misaligned with the PIP. PRSP-2 (2006-2010) attempts to foster an operational structure that includes the PRSP, the MTEF and the budget and therefore proposes one distribution among axes, which are less likely to fluctuate. In the PRSP logic, these axes are defined broadly, with sectors included under the various axes. Therefore, the design and development of CLSP-2 components does not provide details about sectors (PIP). The financing details for the various sectors should have been determined within the framework of the budgetary process.47 The table below attempts to establish a link ex-post. The table presents sectoral programming for PRSP-2 and authorized payments for the CIB over the 2006-2009. It clearly underscores a disconnect between these two instruments. Table 3.5 : PRSP-2 sectoral programming and CIB (authorized payments) (In millions of MRO) - 2006-2009* PRSP CIB PRSP CIB PRSP CIB PRSP CIB Sector 2006 2006 2007 2007 2008 2008 2009 2009 Rural Development 7423 5634 12513 12401 17769 12538 19179 1503 Industrial Development 9165 3636 13426 3941 14979 6801 15397 5472 Territorial Development 31136 21354 41356 37403 90885 76570 105020 16057 Human Resources 9252 10098 12409 9388 16132 13864 16318 3885 Institutional Development 1466 803 2571 2753 6947 4018 7339 1894 Multi-sectoral Projects 7258 7136 8724 2743 9700 8659 28494 13505 Sources: MAED CIB; PRSP data on the February 2011 PRSP implementation report. 46 Precisely, how can allocating a total of 126 billion UM to the “Growth Acceleration� result in an annual growth projected rate of 9.4%, a level Mauritania might never have seen? Or how can allocating 1.261 billion to investment and operation during the 2006-2010 period lead to a reduction in poverty from 46.7% in 2004 (which is the base year) to less than 35% in 2010? 47page 59 of the PRSP 2006-2010 63 3.30. The PRSP-2 is also mis-aligned with the MTEF. The authorities recognize the fact that the 2008-2010 MTEF is not linked to PRSP-2.48 In PRSP-2, it was indicated that the second phase would be accompanied by an MTEF for the 2006-2010 period, which would determine the overall cost of the action plan both at the operating and investment levels, and define the source of financing necessary for its implementation. This MTEF was only finalized in October 2007 to cover the 2008-2010 periods. But even this MTEF was not used by the authorities in their 2008- 2010 programming.49 A second effort to update the MTEF was undertaken in mid-2010 with a view to integrating it into the 2011 budgetary process. However, time and human capacity constraints forced the authorities to focus on the 2011 budget and to defer the updating of the 2011-2015 MTEF to later. 3.31. The MTEF methodology to determine the total and sectoral budgets must be reviewed. In fact, while the MTEF states that it has taken into account government constraints and priorities, it is not obvious that the approach is different from that of the PRSP. The MTEF explains that for operating expenditures, the method used consisted in (i) evaluating current essential/mandatory spending (ii) determining margins from projected total expenditure volumes; and (iii) distributing these margins by ministry based on departments’ priorities as expressed in their three-year action plans. The investment expenditures only took into account the funding acquired at the time of preparing the MTEF for projects implemented by the central government. For investment spending from domestic resources, essential spending on ongoing operations or conventional counterpart funds was first determined. Resources available after essential spending were distributed according to the relevance of the departments’ programs and their alignment with PRSP objectives. The document underscores that MTEF budgets should only be indicative. The annual budget preparation exercise is based on budgets set by the MTEF which must in the first year correspond to those set by the Finance law. 3.32. The table below compares PRSP and MTEF allocations with the budgeted CIB and authorized CIB by PRSP component. It underscores the major differences between the development vision and the programming and execution of the budget. However, our interpretations can only be cautious and indicative as this table is a compilation of data extracted from different documents, presented by axes for the first two columns, and by sector for the two last. It was possible to compare them only by designating the axes as sectors in the PRSP document. Also, given that there is no breakdown for multi-sector projects (2.1% of the programmed CIB); it was not possible to classify them under the axes. Finally, the political upheavals and 2007-2009 global crisis certainly had an impact on public expenditure decisions and therefore on the alignment efforts among the various instruments. 48 The 2008-2010 MTEF is a remarkable document both for its analyses and the orientations it proposes to improve the budgetary process, placed rightly at the heart of public finance reforms. The MTEF underscores the need to set up mechanisms aimed at seeking expenditure efficiency (whose need is underscored above). For example, the MTEF puts forward measures to rationalize expenditure in goods and services which contributes to the improvement of procurement processes, including systematic recourse to competition and the establishment of stock accounting. 49 A first MTEF had been prepared by the Centre Mauritanien d’Analyse de Politiques in 2003 and comprised the first effort at comprehensive medium term expenditure management 64 3.33. What is highlighted here is magnitude rather than exact numbers. The following points are worth noting: 1. Programming: A brief comparison of PRSP and MTEF programming reveals a large difference on axe 1 (acceleration of growth and stabilization of macroeconomic framework) and axe 2 (anchoring growth in the economic sphere of the poor) although both were developed along the same axes. 2. Execution: Programmed and executed CIBs are relatively aligned, with a 14 % to 17 % difference. These differences are once again on components 1 and 2 of the PRSP. 3. Programming vs. Execution: There is a striking difference. Achievements under component 3 are not close to PRSP or MTEF programming, whereas they largely exceed the PRSP on component 1. Table 3.6 : Comparison of PRSP, MTEF, and CIB (budgeted) and CIB (executed) (In % average compared to the total over the indicated period) Component/domain PRSP (2006- MTEF CIB CIB 2010) (2008-2010) (budget) (executed) (2004-09) (2004/09) 1. Acceleration of growth and stabilization of the 23.5 33.9 39.2 44.1 macroeconomic framework 2. Anchoring growth within the economic sphere of the 21.2 14.4 22.9 19.8 poor 3. Development of human resources and expansion of 45.3 43.3 14.7 14.1 basic services 4. Improving governance and strengthening capacity 5.8 7.8 4.5 4.0 5. Strengthening management, monitoring, evaluation and 1.1 0.1 - - coordination 6. Territorial development 1.1 0.6 2.3 1.7 7. Environment 2.0 - 14.3 16.4 8. Multi-sector Projects - - 2.1 0.1 9. SNIM - - Total 100.0 100.0 100.0 100.0 Source: World Bank Mission Calculations. 3.34. Despite ongoing efforts, there is still a lot to be done: PIP and PRSP figures are not completely aligned and deviate significantly from the macroeconomic framework, raising questions about the realism of the PRSP document (see table below). The difference between the figures in the macroeconomic framework and those in the PIP/PRSP underscores the need to finalize the updating of the 2012-2015 MTEF and, in the medium term, to adopt the MTEF methodology to ensure realistic investment budget programming by aligning the “top-down� approach of the macroeconomic framework with the “bottom-up� approach of project development in sectoral ministries. In fact, officials responsible for developing PRSP-3 have attempted to balance the macroeconomic constraints with the objectives of sectoral strategies. The MAED made significant efforts to faithfully align the 2011-2015 PIP with PRSP-3 65 objectives. However, as the table below shows, there are major differences, especially in 2011. A possible explanation is that the PRSP-3 investment plan is based on the 94 projects presented at the June 2010 Roundtable held in Brussels and not on the current PIP. Furthermore, the decline noted in the amount planned for investment in the 2011 PRSP is attributable to the fact that the cycle of a large number of projects ends in 2011 and a new cycle starts in 2012. Regular update of the PIP will help ensure the realignment of budgets and their contents. Table 3.7: Comparison of future programming (in billions of MRO) 2011 2012 2013 PIP (2011-2015) 206.8 219.4 257.5 Capital expenditure (according to IMF 88.4 92.8 97.9 TOFE) PRSP (2011 – 2015) 178.3 154.5 223.7 Source: PRSP 2011-2015 and 1st review of the program with the IMF 3.35. Experience in coordinating the implementation of PRSP-2 can inform measures to be taken in the implementation of PRSP-3. While recognizing the challenges involved in achieving a perfect alignment between the PRSP, PIP, MTEF, CIB and FL. This chapter suggests the following approach in order to improve concordance among these instruments: • Define the macroeconomic framework (years N, N+1, N+2) with IMF services; • Use this framework to finalize the overall MTEF; • Budgets for sectoral ministries are proposed by technical services and are ratified by the government; • These budgets constitute the ceilings for sectoral MTEFs (top-down) and the ministries align their annual allocations with the strategic framework included in the PRSP; • These allocations are then broken down into a CIB and a multi-year operating budget; • The sectoral distributions and breakdown between investment expenditure and current expenditure are confirmed during budget conferences according to an iterative process (combination of top-down and bottom-up). 3.36. The following table summarizes these steps: 66 Table 3.8 : Alignment between PRSP, PIP, MTEF, CIB, and FL Measures Actors Comments Finalize the macro-economic framework PD, DGB, PIPD (MAED) The MAED’s PIPD should be closely (N, N+1, N+2) with the IMF involved in finalizing the framework Use the framework as a basis for PIPD (MAED), PD (Finance) The interaction between the MAED completing the overall MTEF and the Finance Ministry is essential to improving alignment among the instruments. Short of merging the two ministries, it is proposed that PIPD and PD work closely with each other On the basis of the framework, finalize PIPD, PD finalize (technical An iterative process confirms the first sectoral budgets (according to work) and cabinet ratifies sectoral allocations (top-down and the top-down principle) (political ownership, since the bottom-up) budget is a political instrument) Send budgets to sectoral ministry for Sectoral ministries, PIPD, PD discussion. These budgets constitute the (finance). ceiling for sectoral MTEFs Breakdown sectoral MTEFs into current Sectoral ministries , PIPD, PD and investment budgets (finance) Finalize sectoral MTEFs following Sectoral ministries, PIPD, PD budget discussions. (finance). III. DIAGNOSTIC ANALYSIS OF THE PUBLIC INVESTMENT MANAGEMENT FRAMEWORK 3.37. The diagnostic analysis of Mauritania’s public investment management system is based on the analytical framework designed by Rajaram et al (2010)50. The analysis is based on “must- have features� for an efficient public investment system: • Investment budget programming: o Investment guidance, project development and preliminary screening o Formal project appraisal o Independent review of appraisal o Project selection and budgeting • Project implementation: o Project adjustment • Completion review and evaluation: o Facility operation 3.38. This approach highlights the basic processes and controls (linked to the appropriate stages of the broader budget procedure) that are likely to yield the greatest assurance of efficiency in public investment decisions. It asks specific questions to identify the “must-have features�. It seeks to identify the essential institutional features that would minimize major risks and to provide an effective country-specific systemic process that is likely to enhance public 50 Anand Rajaram, T. Le, N. Biletska et J. Brumby, “A Diagnostic Framework Assessing Public Investment Management�, Policy Research Working Paper 5397, August 2010, World Bank. 67 investment efficiency. It does not seek to identify best practices that can be found in sophisticated systems in countries such as Chile or in OECD countries. 3.39. This analysis is complemented by a recent index of public investment efficiency.51 This index captures the institutional environment underpinning public investment management, including (i) strategic guidance and appraisal; (ii) project selection; (iii) project implementation; and (iv) evaluation / audit. This analysis provides “benchmarking� of the general index and the four sub-categories across 71 countries, including 40 low-income countries. The index is based on the arithmetic sum of 17 variables and goes from zero to 4, where a greater value indicates better performance in public investment management. Mauritania scores an overall value of 1.72, slightly below the mean achieved by Ethiopia (1.65), but lower than Ghana (1.87), Mali (2.16) and Burkina Faso (2.09). 3.40. The “benchmarking� of 4 indices for the 4 sub-categories for Mauritania is as follows: (i) a value of 1.67 for strategic guidance and appraisal; (ii) a value of 2.0 for project selection; (iii) a value of 1.20 for project implementation; and (iv) a value of 2.0 for evaluation/audit. The sub-indices show management weaknesses in the four domains, but highlight strong weaknesses in categories (i) and (ii). Era Dabla-Norris, J. Brumby, A. Kyobe, Z. Mills and C. Papgeorgiou. “Investing in Public Investment: An Index of Public Investment Efficiency�, February 2011, International Monetary Fund (WP/11/37). 68 Table 3.9 : Institutional framework for a successful investment program: Good practices and needed reforms in Mauritania. Desirable institutional Indicators of good practices proposal for reforms arrangements 1. Strategic orientation • Existence of a realistic national poverty reduction and • Refine and reinforce PRSP 3 as the main instrument for guiding focus preliminary development strategy (PRSP) and detailed strategies the allocation of resources and their use screening for major sectors with costing • Update and refine sectoral strategies for main sectors such as • All projects are aligned with the PRSP and the strategy transport, education and health and agricultural of the sector concerned • Adopt systematic screening for major projects (consistent with • All the projects are real investments and not operational strategic and clear objectives) fit for country level capacity. budget items Consolidate the operational and investment budgets while clearly distinguishing between investment and operational expenditures in the budget 2. Formal standardized • Formal criteria available for an economic, social and project appraisal environmental study and a cost/benefit analysis • Adopt clear appraisal criteria and disseminate them, in the form • All PIP projects have been appraised (format and of a handbook, to the sectors and other public entities. scope of the appraisal adapted to the size and significance of the project) • Create a project design and appraisal training fund Skills available for the design and appraisal of the project • Train a large number of staff on project design and appraisal 3. Clear roles and independent review of • Central service for the appraisal of major complex • Clarify the institutional mechanism for project design, appraisal appraisal projects and selection (role and creation of a central service and decentralized evaluation services in sectoral ministries) • Independent verification of all major complex projects to ensure the objectivity of the evaluation and the • Create inter-ministerial committees to examine proposals at the clarity of implementing mechanisms preliminary appraisal and screening stages, with the possible participation of NGOs. Establish independent mechanisms to review proposals chosen before introducing them into the CIB 4. Project selection • A structured credible medium term budgetary aligned on budgetary framework sets the sectoral budgets for public • Avoid including in the investment budget projects that were not process investments screened by the appraisal process • Project selection process (phases 1-3 above) integrated into the annual budgetary process • Formalize the rolling PIP to ensure multi-year project 69 • operational and maintenance expenses of existing and programming new investments fully budgeted for • Complete the MTEF reform to complete the integration of the investment and operating budgets and to give the budget a more strategic and medium-term orientation • Systematically evaluate operational and maintenance expenses in sectoral MTEFs and include in the budget 5. Transparent and • Directives for material and financial execution of efficient project published projects. • Adopt and disseminate a project implementation manual implementation • Financial execution is in line with GFP procedures • Cost-effective, competitive and transparent • Implement new budget execution procedures, comprising a procurement computerized and simplified expenditure cycle • Implement the new procurement code • Deviate from procurement procedures established by law only in clearly defined cases 6. Closely monitored, but flexible project • Implementation report of all major projects regularly • Formalize reporting requirement (quarterly and annual) of implementation published executing agencies to the MAED and MoF for all projects. • Flexible financial review procedure allowing changes • Formalize responsibilities (material and financial) for in disbursements (or cancellations) to adapt to evaluation/ audit of sectoral ministries and MAED changing circumstances • Produce material and financial implementation reports for all • Regular updating of cost-benefit analyses. major projects • Accurate updating of valorized assets registers • Adopt a comprehensive database of projects (based on the existing project dossiers) to feed the PIP. Establish an electronic link between this database and implementation agencies for major projects to facilitate immediate reporting. • Adopt accounting of public assets (physical and values) • Introduce a more targeted risk-based monitoring system, especially for major projects. 70 • Develop a transparent and flexible methodology to cancel non- performing projects or similar projects targeting the same recipients. 7. Post-closing • Evaluation criteria integrated from design phase of evaluation major projects ¾ Adopt an evaluation system, initially covering only the critical major projects, then gradually extend it to others • Institutional mechanism in place to evaluate closing projects and publish the results of major projects 71 3.41. To facilitate the diagnostic exercise, Graph 3.1 provides an overview of the institutional framework of the PIM system, including the stakeholders, the beneficiaries, their roles and interactions. This information is detailed in the discussion below. Graph 3.1 : Institutional framework of the PIM: stakeholders, actors, beneficiaries and roles I PIM P S N R BCI PIP MAED S PIU S MAED PIM D C D K P BCI PIM P MF P PIP I PIP CBI PRSP PIP CBI 72 Investment budget programming: • Investment guidance, project development and preliminary screening 3.42. As discussed above, despite government efforts between 2004 and 2009, there was no clear and systematic guidance for Public Investment Management (PIM). The Public Investment Program (PIP), which constitutes the investment planning and implementation framework, was neither linked to the PRSP nor to the MTEF. The PIP includes some guidance, but the effort is inadequate both in terms of strategic orientation and instructions for detailed project design. Moreover, this guidance is relatively unknown in many sectors. In addition, many sectors have not updated their strategies. These problems led to non-alignment of priorities and failure to meet sectoral needs. 3.43. The Ministry of Economic Affairs and Development (MAED) and the Ministry of Finances (MoF) act as integrators of various sectoral requests. The MAED sets up the system and launches the process, requesting contributions from sectoral ministries. The MoF negotiates intra- and inter-sectoral PIPs. The MoF also provides sectoral budget allocations, which, in the absence of a suitable budgetary calendar, arrive late and do not allow enough time for sectors to prepare a budget. Moreover, the MoF has rudimentary software for updating the CIB and monitoring the financial situation of projects (see box below). Box 3.2 : Investment Programming and Monitoring System in Mauritania (SYPSIM) Management of the Public Investments Program (PIP) currently depends primarily on the Investment Programming and Monitoring System in Mauritania (SYPSIM), an application developed at the end of the 90s. The software is used in (i) updating the Public Investment Program (PIP); (ii) developing the “CIB Investment� component of the finance law and; (iii) monitoring project financial implementation. SYPSIM thus provides a range of information on investment projects/plans: project costs, donors and their contributions, executing agencies, annual programming, etc. Currently hosted in the Directorate General of the Budget in the Ministry of Finance, SYPSIM, however, poses two problems. First, it is a separate application, not linked to the central expenditure management application (RACHAD). In addition, the module envisaged for the Directorate General of Investment Projects and Programs in MAED is still not operational and the “decentralization� of the system to sectoral ministries has not achieved the desired results. Finally, the system has a relatively limited capacity to feed the database due to weak or non- existent sectoral data systems or the lack of well established and respected protocols for data collection. 3.44. The practice of programming investment expenditures separately from operating expenditures contributes to the lack of efficiency in programming.52 Furthermore, investment budgets are divided into investments financed by the national budget and investments financed by donors. The lack of integration between these two investment budgets contributes to the disconnect between the development objectives of the authorities, on the one hand, and investment planning and implementation, on the other hand. 52 This inefficiency is in addition to the inefficiency related to the disconnect among budgetary instruments discussed above and weaknesses in budgetary control and off-budget expenditures discussed in the preceding chapter 73 3.45. Until recently, selection of project proposals was not necessarily based on their fundamental alignment with government policy and guidance. The MAED has a pipeline of priority projects waiting for funding from donors. The process for including a project in the PIP is based on a classification, from 1 to 5, which helps to identify the stages to obtain project financing. The box below explains these 5 stages. Often, however, the project selection process seems to work backwards: the government first seeks funding, and then adjusts the budget to include it. Such process results in a supply-driven system: where there is money, ministries and departments invent activities or investments to use it. It is a source of disconnect between the PIP and the PRSP. The grounds for rejecting proposals therefore seem to be the availability of external funding rather than an in-depth analysis of project proposals and their strategic importance. Box 3.3 : PIP: Project selection phases Stage 1 Project ideas (TORs). Feasibility study, detailed and summary draft projects prepared by the sector or donors active in the sector. Most projects conduct this study, but the methodology is not uniform and quality is low. Stage 2 A funding request is submitted to donors. Request for tenders launched in anticipation of delays often experienced with the process. Stage 3 A convention with a donor is signed or becomes effective (approval of Parliament may be needed). Projects included in the BCI are at least at this stage. Stage 4 Funding request is approved (there is a funding agreement). Approval by cabinet and Parliament of the budget. Stage 5 Administrative and technical kick-off of projects or ongoing project. 3.46. Beyond the availability of the funds, criteria for inclusion and the process of screening a proposal are at best rudimentary. Factors affecting this selection are (i) the project status (if a project is already “well prepared� or is already underway) (ii) its social importance (if its inclusion or its implementation would provide service to a large percentage of the population). 3.47. Even these basic criteria are not completely clear. For example, a new project can be included when the project is still an idea or when it is well designed. In addition, the quality of proposals prepared by the sectors varies. Some of the sectoral ministries, such as health, education and rural development have systems which produce projects and prioritize them. The education sector uses a “sectoral approach� supported by multiple donors. Sectoral ministries note that MAED does not always listen to their justifications for proposed projects developed on the basis of the country’s annual needs and sectoral strategies. Indeed, the sectoral justifications may or may not affect the decision to include or exclude a project; if it is approved, the budget amount allocated to the project can be much lower than what was requested for effective launching and appropriate implementation. 3.48. Finally, a large number of projects seem to have been included for political reasons and are therefore not necessarily appraised or designed with the rigor necessary to ensure their relevance, impact, or cost. These “automatically approved� projects replace other projects which have passed the screening process, however rudimentary, at sector or MAED levels. 3.49. The current government has added two screening criteria; however, these criteria do not bring greater transparency and efficiency to the annual budget exercise. These new criteria reflect government objectives: (i) the projects must be related to infrastructure or good 74 governance, and (ii) they must be physical/ hard investments (such as infrastructure) as opposed to investments in computer purchases, capacity building, etc. 3.50. Given the current government’s new efforts to improve investment efficiency and results, the MAED is striving to improve this disjointed approach by giving greater consideration to sectoral priorities. The practical approach suggested in the preceding section and the comprehensive software that is being developed for data entry, monitoring, and analysis of project and investment expenditure management are expected to ensure a much more systematic approach. In view of this, the authorities need to: ¾ Update and refine the strategies of high-spending sectors, such as transport, education and health, and agriculture, developing or updating sectoral MTEFs. ¾ Adopt systematic screening for major projects based on transparent criteria that are compatible with country specific needs and capacities. Build institutional and human capacities to implement the screening methodically. Formal project appraisal 3.51. The government does not have a culture of systematic and in-depth project appraisal. It uses only one rudimentary project document model and does not have an organizational chart of screened projects. Most projects are financed by development partners, with the state providing only counterpart funds. Donor-financed projects are subject to the same process of inclusion as those financed nationally. 53 Given the important role that donors play in the investment budget, the majority of the project appraisal work that is done and evaluated by 54 donors . 3.52. It is uncommon to have a methodologically rigorous appraisal and the different donors have different basic appraisal standards. The majority of these appraisals seem basic and specific to donors, and are aimed at ensuring that the proposed project fits into their area of intervention. In addition, project acceptance criteria are specific to donors. For example, Arab donors more readily agree to support projects. 3.53. A systematic appraisal methodology, applicable to various sectors, should therefore be developed and transmitted to the sectors. This implies: ¾ Taking into account appraisal and cost-benefit analyses needs that are specific to the various sectors. ¾ Informing the sectors of these new processes. 53 The only difference between national projects and those financed by donors is that the Senate and Parliament have a procedure of different approval if projects financed by donors are financed with loans 54 In 2008, the proportion was 68 % of the total PIP budget. The proportion has reduced following the 2008 coup d’état which interrupted relations with donors. But since the 2009 elections, donors have resumed their support to the country. 75 ¾ Strengthening human capacity in the sectors as well as in the MoF and MAED to (i) in the short term, be able verify the quality of the appraisal done by consultants; (ii) in the longer term, undertake appraisal work as part of project development. o Independent Review of Appraisal 3.54. There is no independent review of project appraisals and no plans envisaged in the medium term to develop such capacity. ¾ It would be desirable to develop the capacity to conduct independent reviews of project appraisals in the long run. • Project selection and budgeting 3.55. Final project selection is done within the framework of the budget procedure. However, the procedure for adding new projects to the budgeted PIP lacks the transparent controls to avoid the inclusion of projects that have not gone through existing rudimentary reviews. For example, in the past, the selection of politically motivated projects did not include checks with the PRSP or preliminary appraisal procedures. A few major projects that are not listed in the current PIP and that are not aligned with the PRSP are currently being implemented. 3.56. The MAED is preparing a new investment project selection process which will be based on two principles. The first is to ensure that the PIP is the vehicle for actions relating to the matrix of priority actions under PRSP-3. Effectively establishing its relationship with the PIP will help with implementation of PRSP-3. The second principle is to make monitoring- evaluation possible and practical. This can be done by adopting a small number of existing indicators to capture the results of PRSP-3. This involves replacing the long inefficient lists which were integrated into the preceding PRSP with approximately 50 indicators that currently capture results and changes. In addition, a rolling non-binding matrix has been adopted and includes the new indicators (if they are available). 3.57. To improve project selection process, it would then be desirable to: ¾ Adopt a transparent project screening methodology, specifying the factors/criteria which determining project choice and prioritization. Also, clarify the criteria for accepting “political� projects. ¾ Distribute this information to the sectoral technicians as well as to those in the MAED and MoF. ¾ Ensure all participants receive sufficient training to adopt the methodology. • Project Implementation 3.58. The duality of responsibilities between the Ministry of Finance (MoF) and the Ministry of Economic Affairs and Development (MAED) adversely affects management of the PIP. In the past, the MAED signed agreements with donors, who provide approximately 85 % of the funds for investment projects, while the MoF provided counterpart funds (estimated at 76 15 % of total investment project costs). A chronic case of poor communication and coordination between the two ministries has resulted in poor information flow and has contributed to the weaknesses in the investment budget implementation and supervision. This dual responsibility even called into question the current MTEF (2011-2013) development process. The experience of Mauritius suggests that consolidating responsibilities for the operating and investment budgets will contribute to greater efficiency in budget execution. 3.59. Execution of projects by project implementation agencies (PIA) affects the PIP. Donor-financed projects are generally implemented by PIAs, which are created by donors according to the fiduciary and management requirements of their organization or their country. In 2009, the number of these PIAs was estimated at about 30, of which 9 belonged to the World Bank55. While these “independent� entities can facilitate the implementation of specific projects, they create institutional and political dilemmas for the authorities who have very little control over these entities and their actions. At the institutional level, PIAs are known to attract the best and brightest among civil servants through higher wages, without developing the technical or management capacity of the administration. Given that their lifespan is limited, these PIAs end up closing and if the projects which they implemented are not sustainably integrated into the national system; they bring no long-term benefits to the country. With regard to policy, because of the size of the various projects and the expertise of the managers of these PIAs, they often dominate the sectoral policy space. Finally, the dichotomy between national legal and administrative procedures and donors results in delays in the launching and implementation of certain projects. For example, the US$ 470 million Aftout Es Saheli project, which was designed in 2002 was launched in 2005, after a 3-year delay stemming from the complexity of coordinating the participation of multiple donors. 3.60. The authorities do not fully control project cycles financed through external funding. The authorities intervene at specific moments in the project cycle funded by donors, for example, in the early stages when the project is designed and included in the investment budget and in contributing the national counterpart funds. However, the feasibility or cost-benefit analysis is usually not controlled by the authorities. The MAED also exercises some financial monitoring by issuing payment authorization notifications for project implementation, but the actual implementation (by the PIAs) are conducted outside the national budget process. MAED’s interventions clearly not sufficient to properly control the projects. 3.61. In addition, the two monitoring agencies lack the means to monitor and the authority to impose sanctions. (i) The project monitoring and evaluation directorate is under the authority of the MAED. It is supposed to ensure financial, administrative and physical control, but it cannot do the latter as it lacks human and material resources and it does not have the authority to impose sanctions which makes its current supervisory powers is rather weak. Moreover, although its recently created “physical and financial monitoring division�, has been staffed and has a plan of action, it is yet to start operations due to lack of funding. (ii) The national agency in charge of project monitoring (for major projects). It is an independent agency with the same responsibilities as the project monitoring and evaluation directorate, but it only deals with major projects. It reports directly to the President of the Republic. 55 In 2009, there were close to 270 project units, equal to the number of projects in the CIB that year, of which 30 were project implementing agencies (project implementing agencies outside the government structure but under government supervision.) 77 3.62. There is no database on the financial and physical implementation of the entire CIB, a shortcoming that exists even for projects financed with national funds. This serious handicap hinders authorities’ efforts to monitor and supervise. 3.63. It would then be desirable to: ¾ Consider consolidating the two monitoring agencies; empower the new entity with the authority to impose sanctions and provide it with human and material capacity to make it functional. ¾ The development and implementation of the project management software mentioned above will be critical to improve transparency in the financial and physical implementation of the BCI. 3.64. The use of procurement procedures is binary in nature. Until now, projects supported by donor funds and implemented by PIAs have included fiduciary and procurement clauses that are aligned with donor country rules. As a result, international enterprises are authorized to bid. 3.65. At the national level (counterpart funds as well as projects financed with national funds), ministries use procurement plans. RACHAD is used by the ministries for the execution of credits received from the national budget. However, the ministries still have to face cash management issues. Once national appropriations are entered into RACHAD, the ministries sometimes find it difficult to access their appropriations into the second half of the financial year because of cash management issues at the Treasury. This affects the schedule for launching and implementing projects financed through the national budget as well that of counterpart funds to be disbursed for projects. 3.66. Contracts are awarded after competitive bidding processes, but it was noted that the national private sector lacked the capacity to properly implement the projects. There are many facets to this lack of national capacity. On the one hand, some companies make multiple bids and obtain multiple contracts but do not have enough technical and human resources to implement them and cannot therefore respect the deadlines or meet the required quality standards. In addition, only a small number of companies have the capacity to respond to calls for tenders on projects financed by donors. This is particularly true for small and medium sized enterprises (SME), which do not have the know-how to obtain international contracts and leads to possible collusion in the submission of tenders, notably for infrastructure projects.56 3.67. Furthermore, as discussed in Chapter two, up to end 2010 the procurement process still faced weaknesses identified in the updated CPAR review of 2002. These include a process that is highly centralized at the level of the Central Procurement Commission (CPC); lack of qualified and motivated staff; as well as measures that are not aligned with internationally recognized procurement principles.57 . 56 With respect to infrastructure, one option would be to use the cost of a kilometer of road as a basic measure for road costs. Another solution would be to have existing state enterprises compete with private companies involved in collusive actions. Competition in the public sector, although not optimal, would be an attempt to dissociate companies that collude. In addition, the government plans to take legal action against companies involved in collusion. 57 Aspects such as clarity in rules of application, eligibility conditions for procurement and incompatibility of some 78 3.68. It is critical that the new procurement code, adopted in July 2010, be implemented as soon as possible. As discussed in Chapter two, this involves creation of new institutions required by the new law as well as capacity building to facilitate implementation of the new code. The new code aims to improve the attractiveness of the country’s public projects to national and international private operators and to reduce the deadlines and costs of procuring goods and services for the Mauritanian government. Its successful implementation will be an additional positive factor that will ultimately consolidate all projects under national procedures (Paris Declaration). • Project adjustment 3.69. Management of existing projects apparently does not include systematic rehabilitation of non-performing projects. Until September 2010, no project had been cancelled. The authorities were developing an approach to deal with the problem projects. They had developed a simple methodology to identify the causes of delays in a project: depending on whether the implementer is a donor, the government, or a company. They were prioritizing projects to determine which to cancel or which to postpone. One of the methods used by the authorities is to cancel projects with overlapping goals and targeted beneficiary, who would not be affected by such a cancelation. 3.70. To improve the CIB’s visibility and efficiency, the authorities should streamline their portfolio. This streamlining will help cancel projects with no funding that are still in the priority project portfolio; and close projects that are non-performing, have been seriously delayed, or whose costs exceed the benefits. To this end, the authorities should consider adopting flexible financial review procedures to allow changes in the disbursement (or cancellations). This will give them the flexibility to adjust to changing circumstances and regularly update the cost-benefit analysis. ¾ It is clearly necessary to streamline some projects. This can mean unifying similar projects affecting the same beneficiaries. It also means cancelling non performing projects in order to use resources allocated to them on performing projects. The authorities should ensure appropriate consultation with the sectoral beneficiaries of the non-performing projects, ensuring that the rational for streamlining or canceling of the projects are clarified, and their allocated resources transferred to optimize the chances of reaching the development objectives of the PRSP-3, potentially to performing projects within the same sector. • Completion review and evaluation 3.71. Given the absence of a database and of project management software, the government does not fully monitor or evaluate projects managed by PIAs. Monitoring and functions. In particular, those functions related to regulations, contract award and attribution, supervision, and recourse which are handled by the CPC. 79 evaluation at the national level is weakened by the multitude of PIAs, the lack of rigor in donor reports and lack of institutional and human capacity at the government level. 3.72. The problem of supervision and evaluation is multidimensional. On the one hand, PIAs do not perform their job of reporting properly and impartially. PIAs are supposed to send quarterly progress reports, but not all do so. Even when reports are written, the majority are unrealistically optimistic regarding the project’s progress and do not reflect the reality of the project’s effective implementation. On the other hand, the authorities do not have the necessary tools for proper supervision and evaluation. There is no suitable model for evaluation of the performance of projects or that of their implementing agencies (PIA). In addition, the departments of sectoral ministry do not necessarily supervise the project’s implementation in their own sector/\. 3.73. Lastly, current project “success� indicators are often derived from bad indicators. For example, disbursement of a payment instalment is considered progress in implementation, of the project ultimate success or sustainability. However, this is often a very narrow indicator of progress. Implementation delays can stem from a design defect or technical or management problem. These delays affect monitoring and evaluation and, by extension, the rating of the project by donors. Managers of donor agencies and PIAs may seek to satisfy a donor who needs action and progress by disbursing funds on the basis of revised actions within the framework of the project, without resolving the technical or design defects or the project management issues. 3.74. The government does not conduct a final evaluation of projects and does not calculate the current net value of projects completed. ¾ A project database mentioned above could include a module that could be directly installed in PIAs and connected to the MAED to ensure regular information transfer. • Operation of facilities and fundamental review and evaluation of projects after completion 3.75. There are no procedures for transmitting the responsibility for using and maintaining created fixed assets to service delivery agencies. In practice, coordinated use and maintenance are not well programmed. Moreover, service delivery agencies do not necessarily have sufficient budget to exploit and maintain these fixed assets. The operation of these facilities is not monitored. This is due to the lack of good project design and lack of impact analysis on the population. 3.76. There is no assets registry and no inventory of public goods. Not all public goods have a legal title, especially land and buildings. They are clearly not assessed in accordance with good accounting principles, in compliance with the accounting definition of an asset, with depreciation in the asset’s value and, if possible, updating the value of the assets, taking into account price changes. 80 ¾ The government should launch a comprehensive assessment of public sector assets as a first step, followed by the creation of an inventory of public goods and an audit of all government assets. IV. POLITICAL REALITIES, HUMAN AND INSTITUTIONAL CAPACITY AND FEASIBILITY OF REFORMS. 3.77. This chapter presented an in-depth study of the public investment issue. It analyzed existing data and noted the significant presence of donors in this exercise, underscoring the problems associated with implementing projects under a non-consolidated investment budget. The analysis also noted an average of 4 % of operating expenses in the investment budget. Finally, the investment management analysis reveals serious weaknesses. 3.78. The chapter proposes a number of reforms that are technically necessary to improve the Public Investment Management system overall and overtime. Some of these reforms may not be feasible in the short run due to lack of human or institutional capacity. The reforms may also face lack of policy agreement among all interested parties or lack of political will to implement them as they affect special interest groups within and outside the government. 3.79. The current government has expressed strong interest in and will to get results from CIB expenditures that improve the living conditions of the population. The President of the Republic has emphasized the importance of improving governance and accountability on several occasion. However, institutional and human limitations make it advisable to select and sequence reforms. We propose several feasible short and medium term reforms that will impact the CIB management, building up momentum for the remaining reforms discussed in the chapter. In the short term: ¾ Ensure that operating expenses are not included in the investment budget. This requires the consultation between MoF, MAED and sectoral ministries, followed by clear instructions from the MoF and MAED to sectors to stop such practice. ¾ MAED should implement, as quickly as possible, the integrated software under development that will monitor the project cycle from inception to impact. This software should include information on appraisal, approval, accountabilities and supervisory responsibilities, work time horizon, costs, results and impact indicators, among other details. In a second stage, this software could include a module directly installed in PIAs and connected to the MAED and MoF to ensure regular information transfer. ¾ The government should lead a biannual multilateral review of the investment budget, which could help eliminate bottlenecks in the implementation of projects and inform the preparation of the next year’s PIP. These biannual reviews can be placed under the aegis of the PRSP-3 Supervisory Committee and combined with PRSP-3 reviews. They can further be timed to coincide with the two annual IMF missions to ensure macroeconomic consistency of PRSP-3 and CIB plans. In the medium term (3 year horizon): 81 ¾ Streamline the portfolio to improve the visibility and efficiency of the CIB. This can mean unifying similar projects affecting the same beneficiaries. It also means cancelling non performing projects in order to use resources allocated to them on performing projects. This action will require strong political commitment. While MoF and MAED would be the natural champions of this effort, engagement at the higher political levels will lend credibility to the importance of this task in improving the efficiency and results of the CIB. The authorities should ensure appropriate consultation with the sectoral beneficiaries of the non-performing projects, ensuring that the rational for streamlining or canceling of the projects are clarified, and their allocated resources transferred to optimize the chances of reaching the development objectives of the PRSP-3, potentially to performing projects within the same sector. ¾ MAED should adopt a much more rigorous and systematic methodology for the development and adoption of projects into the CIB portfolio. This includes a clear methodology for preliminary appraisal and cost-benefit analyses for any project/sector and criteria that are more rigorously linked to PRSP-3 and to sectoral strategies. This will help eliminate projects that have not undergone the required appraisals and analyses and not proven their possible contributions to PRSP-3 objectives. Capacity building at MoF, MAED and sectoral levels will help facilitate implementation of the new methodologies and selection criteria. Encouraging donors to unify their data requirements, methodologies and processes for appraisal and their selectivity criteria will also facilitate improving the process of developing and choosing a project into the CIB portfolio. In summary, PIM reforms Responsible Time frame Ensure that operating expenses are MoF, MAED and sectoral ministries 1 Year not included in the investment budget MAED should implement, the MAED, Prime Minister’ office. 1 Year integrated software under Sectoral ministries, MoF development that will monitor the project cycle from inception to impact The government should lead a Minister of Finance/MAED, donors, 1 Year biannual multilateral review of the PRSP supervisory committee, investment budget, to eliminate Project implementation units. bottlenecks in the implementation of projects and inform the preparation of the next year’s PIP Streamline the portfolio to improve MAED, Prime Minister’ office. 1 to 3 Years the visibility and efficiency of the Sectoral Ministries, CIB. MAED should adopt a much more MoF, MAED and sectoral 1 to 3 Years rigorous and systematic ministries. methodology for the development and adoption of projects into the 82 CIB portfolio 83 Annex 1: Millennium Development Goals, 1990-2015 CSLP MDG 1990 1996 2000 2002 2004 2008 2010 - 2015 2015 In percentage of total population 1. Reduce abject poverty and hunger Overall incidence of poverty 56.6 50.0 46.3 -- 46.7 42.0 35.0 25.0 28.3 Incidence of poverty in 36.1 25.4 29.2 -- 25.9 15.6 -- -- -- Nouakchott Prevalence of infant malnutrition 47.6 -- 32.0 -- 30.2 39.4 26.0 21.0 23.8 In percentage of age group corresponding to elementary level 2. Ensure primary education for all Gross primary school enrolment 49.2 75.9 89.1 89.8 76.7 90.9 98.0 100.0 100.0 rate In percentage of the total number of children enrolled in the first year of primary education Retention rate in 5th year of 75.3 -- 59.6 60.6 52.9 63.7 68.8 100.0 100.0 primary education In percentage of the total number of children enrolled 3. Promote gender equality Percentage of girls in the total 42.1 46.0 48.0 49.0 79.0 93.5 49.0 50.0 50.0 number of children enrolled in primary education For 1,000 live births 4. Reduce infant mortality Infant mortality 129.5 125.6 122.0 -- -- -- 128.0 55.0 45.7 (under 5 years) For 100,000 live births 5. Improve maternal health Reduce maternal -- -- 747.0 -- 820.0 820.0 400.0 300.0 -- mortality rate In percentage of 15-24 year olds 6. Control HIV/AIDS, malaria and other diseases Reduce HIV/AIDS -- -- 0.5 0.6 0.5 -- <1.1 <1 1.0 prevalence by half In percentage of the population 7. Ensure a sustainable environment Access to better water 37.0 40.0 50.0 -- 52.0 62.0 65.0 75.0 -- sources Access to electricity -- -- 18.0 -- 24.0 30.6 -- -- -- Source: Mauritanian authorities 84 Annex 2 : Country at a Glance Mauritania at a glance 3/23/11 S ub- K ey D ev elo pment Indi cat o r s S aharan L ow Mauritania Africa income Age distribution, 2008 (2 0 0 9 ) Male Female P opulation, mid-year (millions ) 3.3 81 9 828 75-79 S urface area (thous and s q. km) 1,031 24,242 17,838 60-64 P opulation growth (%) 2.4 2.5 2.2 Urban population (% of total population) 62 36 28 45-49 30-34 GNI (Atlas method, US $ billions ) 1.8 887 379 15-19 GNI per capita (Atlas method, US $ ) 600 1,082 457 GNI per capita (P P P , international $ ) 1,960 1,973 ,1 1 37 0-4 10 5 0 5 10 GDP growth (%) -1.2 5.2 6.2 percent of total population GDP per capita growth (%) -3.5 2.7 3.9 (mo s t r ecent es t i mat e, 2 0 0 3 – 2 0 0 9 ) P overty headcount ratio at $ 1 a day (P P P , %) .25 21 51 .. Under-5 mortality rate (per 1,000) P overty headcount ratio at $ 2.00 a day (P P P , %) 44 73 .. Life expectancy at birth (years ) 57 52 57 200 Infant mortality (per 1,000 live births ) 74 83 77 180 Child malnutrition (% of children under 5) 23 25 28 160 140 Adult literacy, male (% of ages 1 and older) 5 64 72 73 120 100 5 Adult literacy, female (% of ages 1 and older) 50 54 59 80 Gros s primary enrollment, male (% of age group) .. 105 107 60 Gros s primary enrollment, female (% of age group) 102 95 100 40 20 0 Acces s to an improved water s ource (% of population) 45 60 64 Acces s to improved s anitation facilities (% of population) 26 31 35 1990 1995 2000 2007 Mauritania Sub-Saharan Africa a N et A i d F lo ws 19 8 0 19 9 0 2000 2009 (US $ millions ) Net ODA and official aid 175 236 221 311 Growth of GDP and GDP per capita (%) T op 3 donors (in 2008): E uropean Commis s ion 7 22 64 40 25 S pain 0 0 5 34 20 F rance 20 53 23 29 15 10 Aid (% of GNI) 22.5 .. 20.2 9.4 5 Aid per capita (US $ ) 1 15 1 19 85 97 0 -5 L o ng-T er m E co no mi c T r ends -10 95 05 Cons umer prices (annual % change) .. 6.5 6.8 2.2 GDP implicit deflator (annual % change) 16.9 4.0 1.3 -5.9 GDP GDP per c apita E xchange rate (annual average, local per US $ ) 45.9 80.6 238.9 262.0 T erms of trade index (2000 = 100) .. 14 1 100 .. 2000– 09 19 8 0 – 9 0 19 9 0 – 2 0 0 0 (average annual growth %) P opulation, mid-year (millions ) 1.5 2.0 2.6 3.3 2.6 2.7 2.6 GDP (US $ millions ) ,01 1 7 1,075 1,081 3,031 1.8 2.9 4.7 (% of GDP ) Agriculture 35.7 34.4 27.6 20.2 1.7 -0.2 0.8 Indus try 23.5 24.7 29.7 35.1 4.9 3.4 5.0 Manufacturing 4.2 4.7 9.0 4.1 -2.1 5.8 -1.5 S ervices 40.8 40.9 42.6 44.7 0.6 4.9 5.5 Hous ehold final cons umption expenditure 61.6 65.9 82.8 73.9 1.4 1.8 7.0 General gov’t final cons umption expenditure 31.9 24.6 25.8 20.5 -3.8 5.4 4.8 Gros s capital formation 27.2 23.8 19.4 24.1 6.9 .. 10.3 E xports of goods and s ervices 25.7 43.3 46.2 50.2 2.1 -1.3 5.6 Imports of goods and s ervices 46.5 57.6 74.2 68.8 0.2 0.6 10.4 Gros s s avings .. 6.0 0.8 1 1 .6 Note: F igures in italics are for years other than thos e s pecified. 2009 data are preliminary. .. indicates data are not available. a. Aid data are for 2008. Development E conomics , Development Data Group (DE CDG). 85 Mauritania B alance o f P ayment s and T r ade 20 00 20 09 Governance indicators, 2000 and 2009 (US $ millions ) T otal merchandis e exports (fob) 345 1,364 T otal merchandis e imports (cif) 336 1,447 Voice and accountability Net trade in goods and s ervices -199 -562 Political stability Current account balance -98 -379 Regulatory quality as a % of GDP -9.0 -12.5 Rule of law Workers ’ remittances and compens ation of employees (receipts ) .. .. Control of corruption R es erves , including gold 49 238 0 25 50 75 100 Country’s percentile rank (0-100) C ent r al Go v er nment F inance 2009 2000 higher values imply better ratings (% of GDP ) Source: Kaufmann-Kraay-Mastruzzi, World Bank Current revenue (including grants ) 24.7 25.5 T ax revenue 13.7 15.2 Current expenditure 20.5 23.5 T echno l o gy and Inf r as t r uct ur e 2 00 0 20 08 Overall s urplus /deficit -6.0 -5.1 P aved roads (% of total) 1 .3 1 26.8 Highes t marginal tax rate (%) F ixed line and mobile phone Individual .. .. 00 s ubs cribers (per 1 people) 1 67 Corporate .. .. H igh technology exports (% of manufactured exports ) .. .. E xt er nal D ebt and R es o ur ce F l o ws E nv ir o nment (US $ millions ) T otal debt outs tanding and dis burs ed 1,963 1,767 Agricultural land (% of land area) 39 39 T otal debt s ervice 83 78 F ores t area (% of land area) .. .. Debt relief (HIP C, MDR I) 920 465 T erres trial protected areas (% of s urface area) .. 0.9 T otal debt (% of GDP ) 81 1 .5 49.2 F res hwater res ources per capita (cu. meters ) 145 124 T otal debt s ervice (% of exports ) 16.9 4.5 F res hwater withdrawal (billion cubic meters ) 1.7 .. F oreign direct inves tment (net inflows ) 40 -38 CO2 emis s ions per capita (mt) 0.46 0.62 P ortfolio equity (net inflows ) -6 .. GDP per unit of energy us e (2005 P P P $ per kg of oil equivalent) .. .. Composition of total external debt, 2008 E nergy us e per capita (kg of oil equivalent) .. .. IBRD, 0 IDA, 243 Short-term, 301 IMF, 16 Private, 0 Wo r l d B ank Gr o up po r t f o li o 2 00 0 20 09 (US $ millions ) Other multi- lateral, 473 IB R D T otal debt outs tanding and dis burs ed 0 – Dis burs ements 0 – Bilateral, 734 P rincipal repayments 2 – Interes t payments 0 – US$ millions IDA T otal debt outs tanding and dis burs ed 449 282 Dis burs ements 57 44 P r iv at e S ect o r D evelo pment 2000 2 0 09 T otal debt s ervice 4 4 T ime required to s tart a bus ines s (days ) – 19 IF C (fis cal year) Cos t to s tart a bus ines s (% of GNI per capita) – 34.7 T otal dis burs ed and outs tanding portfolio 22 14 T ime required to regis ter property (days ) – 49 of which IF C own account 22 14 Dis burs ements for IF C own account 7 0 R anked as a major cons traint to bus ines s 20 00 20 0 9 P ortfolio s ales , prepayments and (% of managers s urveyed who agreed) repayments for IF C own account 5 1 Acces s to/cos t of financing .. 21.6 Anticompetitive or informal practices .. 14.4 MIGA Gros s expos ure 0 5 S tock market capitalization (% of GDP ) .. .. New guarantees 0 0 B ank capital to as s et ratio (%) .. .. Note: F igures in italics are for years other than thos e s pecified. 2009 data are preliminary. 3/23/11 .. indicates data are not available. – indicates obs ervation is not applicable. Development E conomics , Development Data Group (DE CDG). 86 Millennium Development Goals Mauritania With selected targets to achieve b etween 1990 and 2015 (es timate clos es t to date s hown, +/- 2 years ) M aur it ania Go al 1: halv e t he r at es f o r ext r eme po v er t y and malnut r it io n 19 9 0 19 9 5 2000 2008 P overty headcount ratio at $ 1 a day (P P P , % of population) .25 .. 23.4 21.2 .. P overty headcount ratio at national poverty line (% of population) .. 50.0 46.3 .. S hare of income or cons umption to the poores t qunitile (%) .. 6.3 6.2 .. P revalence of malnutrition (% of children under 5) .. .. 30.4 23.2 Go al 2 : ens ur e t hat childr en ar e able t o co mplet e pr i mar y s cho o li ng P rimary s chool enrollment (net, %) .. 55 63 77 P rimary completion rate (% of relevant age group) 29 34 51 64 S econdary s chool enrollment (gros s , %) 14 16 18 24 Youth literacy rate (% of people ages 1 5-24) .. .. 57 .. Go al 3 : eliminat e gender di s par i t y in educat io n and empo wer wo men R atio of girls to boys in primary and s econdary education (%) 69 82 95 103 Women employed in the nonagricultural s ector (% of nonagricultural employment) .. .. 36 .. P roportion of s eats held by women in national parliament (%) .. .. .. .. Go al 4 : r educe under -5 mo r t alit y by t wo -t hir ds Under-5 mortality rate (per 1 ,000) 129 126 122 1 18 Infant mortality rate (per 1,000 live births ) 81 79 77 75 Meas les immunization (proportion of one-year olds immunized, %) 38 67 62 65 Go al 5 : r educe mat er nal mo r t ali t y by t hr ee-f o ur t hs Maternal mortality ratio (modeled es timate, per 1 00,000 live births ) .. 870 .. .. B irths attended by s killed health s taff (% of total) 40 58 57 .. Contraceptive prevalence (% of women ages 1 5-49) .. .. .. .. Go al 6 : hal t and begin t o r ev er s e t he s pr ead o f H IV/A ID S and o t her majo r di s eas es P revalence of H IV (% of population ages 1 5-49) 0.1 0.1 0.6 0.8 Incidence of tuberculos is (per 100,000 people) 230 250 280 320 T uberculos is cas e detection rate (%, all forms ) 120 68 43 26 Go al 7 : hal v e t he pr o po r t io n o f peo ple wit ho ut s us t ainabl e acces s t o bas ic needs Acces s to an improved water s ource (% of population) 37 35 35 45 Acces s to improved s anitation facilities (% of population) 1 6 1 8 21 26 F ores t area (% of total land area) .. .. .. .. T erres trial protected areas (% of s urface area) .. .. .. 0.9 CO2 emis s ions (metric tons per capita) 1.3 1.3 0.5 0.6 GDP per unit of energy us e (cons tant 2005 P P P $ per kg of oil equivalent) .. .. .. .. Go al 8 : dev elo p a glo bal par t ner s hip f o r dev elo pment 00 T elephone mainlines (per 1 people) 0.3 0.4 0.7 2.4 00 Mobile phone s ubs cribers (per 1 people) 0.0 0.0 0.6 65.1 Internet us ers (per 1 people) 00 0.0 0.0 0.2 1.9 P ers onal computers (per 1 people) 00 .. 0.6 1.0 4.5 Education indicators (%) Measles immunization (% of 1-year ICT indicators (per 100 people) olds) 125 100 80 70 100 75 60 75 50 50 50 40 25 30 25 20 0 10 2000 2002 2004 2006 2008 0 0 1990 1995 2000 2007 2000 2002 2004 2006 2008 Primary net enrollment ratio Mauritania Sub-Saharan Afric a Fixed + mobile subscribers Internet us ers Ratio of gir ls to boys in primary & secondary educ ation Note: F igures in italics are for years other than thos e s pecified. .. indicates data are not available. 3/23/11 Development E conomics , Development Data Group (DE CDG). 87