Report No. 36385-MR Mauritania Improving Budget Management to Promote Sustainable Development and Reduce Poverty Public Expenditure Review Update December 2007 PREM 4 Africa Region Document of the World Bank HSDP Health Sector Development Program ICT Information and communication technologies IFMS Integrated Financial Management System IGF General Inspectorate o fFinance IGE General Inspectorate o f the State IMF InternationalMonetary Fund ITD Information Technology Directorate LdF State Budget Law LdR Budget Execution Law MAED MinistryofEconomic Affairs andDevelopment MDGs Millennium Development Goals MDRE Ministryo fRuralDevelopmentandEnvironment MEFS MinistryofBasic and SecondaryEducation MF MinistryofFinance MMI Ministryo fMiningandIndustry MPEM Ministry o f Fishingand Maritime Economy M O H S A Ministry of Health and Social Action MTEF Medium-term ExpenditureFramework OECD Organization for Economic Cooperationand Development ONS National Statistical Office PAP Priority Actions Plan PASEP Public Enterprise Adjustment Program PE Public Enterprise PER Public Expenditure Review PIP Public Investment Program PRECASP Public Sector Capacity BuildingProject PRGF Poverty Reduction and GrowthFacility PRSC Poverty Reduction Support Credit PRSP Poverty Reduction Strategy Paper PSIA Poverty and Social Impact Analysis REER Real effective exchange rate ROSC Report on Observance o f Standards and Codes RPRP Regional poverty reductionprograms SIP Sector Investment Program SNDE NationalWater Company SNIM National Industrial and MiningCompany SOMELEC National Electricity Company SONELEC National Electricity and Water Company TOFE Monthly table o f government financial operations Vice President: Obiageli K.Ezekwesili Country Director: James P. Bond Sector Director: Sudhir Shetty Sector Manager: Antonella Bassani 11 TABLEOFCONTENTS ACKNOWLEDGEMENTS ................................................................................................................ V PREFACE .......................................................................................................................................... V I EXECUTIVE SUMMARY .............................................................................................................. vi1 1. RECENT ECONOMIC DEVELOPMENTS AND THEIR IMPACT ................................... 1 A 1 B. . Context ................................................................................................................... 2 C 5 D.. Macroeconomic Slippages .................................................................................... Poverty and Social Indicators .............................................................................. Conclusions and Recommendations .................................................................... 7 2. PUBLIC EXPENDITURE ANALYSIS ..................................................................................... 8 A The Impact of Off-Budget Expenditures on Total Government Spending .....8 B.. C Functional Classification of Expenditures 16 D.. Breakdownof Expenditures According to Economic Classification .............11 Conclusions and Recommendations .................................................................. ........................................................ 18 3. DISCONNECT BETWEENPUBLIC INVESTMENTPROGRAMAND THE PRSP PRIORITIES ............................................................................................................................. 20 A Costing of the PRSPPriority Actions Plan(PAP) ........................................... 21 B Consolidated Investment Budget (BCI) and Priority Actions Plan (PAP) ....22 C BCIExecution andthe Achievement of PRSP Objectives .............................. 24 D Consolidated Investment Budget (BCI) FundingSources .............................. 28 E HIPC Resourcesand PRSP Priorities ............................................................... 30 F... ... Conclusions and Recommendations .................................................................. 31 4. BUDGET PROGRAMMING. EXECUTION. AND PUBLICACCOUNTS ....................... 33 A. Previous Assessments of the PublicFinancial Management System .............33 B ImprovingBudget Presentation ......................................................................... 36 C.. Optimizingthe MTEFApproach inthe Budget Preparation Cycle D ImprovingBudget Execution Procedures ......................................................... ..............37 40 E F ImprovingPublicAccounting and Reporting 45 G.. .. Computerizing the Expenditure Chain in Support of Deconcentration ........44 Conclusions and Recommendations .................................................................. .................................................. 46 5. BUDGET CONTROL. AUDITS AND PUBLIC EXPENDITURE REVIEWS ...................49 A ....................................................................... .......................................................... 49 B Updatingthe Procurement Code 50 C Combating Corruption ....................................................................................... 51 D... . ImprovingInternal and Ex Post Control Reinforcing Judicial And Parliamentary Control ........................................... 51 ... 111 E. Main Problems Relatedto Audits 52 F. ..................................................................... 52 G . Monitoring and Evaluation of Public Expenditures Conclusions and Recommendations .................................................................. ........................................ 53 LIST OF TABLES: Table 1.1: Tax Revenues (inpercentage of GDP) ...................................................................... 4 Table 2.1: Military Expenditures in Selected Countries (inpercentage of GDP) Table 2.2: Economic Classification of Government Expenditures ......................................... ................13 Table 2.3: Operating Subsidies to the Main Public Enterprises (PES).1998-2002 ...............14 Table 2.4: Breakdown of Expenditures According to the FunctionalClassification 17 Table 3.1: BCIExecution by PRSP Pillar, 2001-2005 ............................................................. ................15 25 Table 3.2: Breakdown of ExecutedHIPC Resources (in millions UM) ................................. 30 LIST OF FIGURES: Figure 1.1: Evolution of Poverty Incidence. 2000.2004 5 Figure 2.1: Off-Budget Expenditures (YOof GDP) .................................................................... ............................................................ 9 Figure2.2: Public Revenues and Expenditures(YOof GDP) 9 Figure 3.1 :Breakdown of BCIAllocations to Pillar 2 ............................................................. .................................................... 23 Figure 3.2: Breakdown of BCIAllocations to Pillar 3 ............................................................. 23 Figure 3.3: PRSP Priority Activities and BCI Execution. 2002-2004 .................................... 24 Figure 3.4: Gross Enrolment Ratio (GER) in Basic Education 1991-2004 28 Figure 3.5: BCIFinancing Sources. 1992-2000 and 2001-2005 .............................................. ........................... 29 LIST OFBOXES: Box 1.1: Crisis of Confidence betweenthe Donor Community and Mauritania 3 Box 1.2: Current Monetary and Exchange Rate Policy Measures ........................................... .................... 4 Box 2.1: Method of Disbursing Off-Budget Expenditures 10 Box 2.2: Expenditure Funds ...................................................................................................... ...................................................... 11 Box 2.3: Low Civil Service Salaries: A Disincentive 12 Box 2.4: Subsidies to PES ........................................................................................................... ................................................................ 15 Box 3.1: Public InvestmentProgram (PIP) and ConsolidatedInvestment Budget (BCI) Box 4.1: HIPC AAP Indicators .................................................................................................. ...21 34 Box 4.2: Tunisia's Experience with Performance-Based Budgeting and Decentralizationof Budget Execution ........................................................................................................ 40 Box 4.3: FiscalManagement Measures Introduced in2004 ................................................... 41 Box 4.4: Recent Measures Taken After the Coup d'Etat of August 3, 2005 ......................... 43 LISTOFANNEXES: Annex 1: SelectedMacro-Economic Indicators (2001-2006) 56 Annex 2: Fiscal Revenues inMauritania .................................................................................. .................................................. 58 Annex 3: Evolution of Poverty and Social Indicators ............................................................................. .............................................................. 60 Annex 4: Evolution of PublicExpenditures 62 Annex 5: Cost Planningfor PRSP-I PAP ................................................................................. 65 Annex 6: ImplementationStatus for the Recommendations of the 2004 PER ..................... 66 1v ACKNOWLEDGEMENTS This Public Expenditure Review (PER) update was preparedby a team o f World Bank staff and their counterparts from the Government o f the Islamic Republic o f Mauritania (GIRM). Representatives from the French Cooperation Agency and German Agency for Technical Cooperation (GTZ) contributed to the analysis. The World Bank team was composed o f Hawa Cisse Wagut (Economist, AFTP4), Nicola Pontara (Senior Economist, AFTP4), Moustapha Ould El Bechir (Procurement Specialist, AFTPC), Amadou Oumar Ba, (Senior Agriculture Specialist, AFTS4), Ismael Ouedraogo (Senior Agricultural Economist, AFTS4), Alain Jean Catalan (Senior Financial Management Specialist, AFMS), Lucien Gaillard (Consultanflublic Finance Specialist), Sow Abdoulaye (Consultanflublic Finance Specialist ) and Ba Moussa Bathily (Consultant). Mr.Diagana MouhamedYoussouf, Director of Budget and Accounts inthe Ministry of Finance (MF), ledthe Mauritanianteam, composed o fThiam Diombar (Deputy Director o fBudget and Accounts, MF), Sid'Ahmed Ould Chekhna (Head o f Research, MF), Fall Khayar (Head o f Research, MAED), Christophe Reilharc (Public Finance Expert and Adviser to the Director o f Budget and Accounts) and Emma Kellner (Public Finance Expert and Adviser to the Court o fAuditors). This report has benefited from the overall guidance o f Robert Blake and Antonella Bassani, Sector Managers (AFTP4), and from the comments o f peer reviewers Adnan Fozzard (Country Manager, LCCHN), David Nielson (Senior Economist, AFTS2), Catherine Laurent (Senior Public Sector Management Specialist, MNSED), and Waly Wane (Economist, DECRG). The PER team thanks Frangois Rantrua, Country Manager (AFMMR),and the IMF team for Mauritania for their comments and support. The PER update team i s especially grateful to Pierre Lenaud (AFTP4) and Fatima Cherif (AFMMR)for logisticalassistance. V PREFACE This Public Expenditure Review (PER) update i s a follow-up to a PER that was carried out in 2004 but remainedunpublishedbecause the official statistical data on which it had been based was inaccurate. The new PER draws upon corrected macroeconomic figures published by the authorities for the period 1992- 2004, and was drafted once the Government made a commitment to regularize public financial management and to lay the foundations for achievingmacro-budgetary balance. The review addressesthe following main questions: What were the effects o f macro-budgetary slippages on priority sectors and poverty reduction expenditures? Have investment expenditures, designed to help to achieve the key Poverty Reduction Strategy Paper (PRSP) objectives, been planned and executed in line with the Priority Action Plan (PAP)? What were the results achieved? 0 Are the conclusions and recommendations o f the 2004 PER still relevant in light o f the new statistical data? The first part o f the report, "context and public expenditure analysis," presents an analysis o f the new economic data, particularly o f the statistics on budgetary execution for 1995-2004. This part provides answers to the first two questions. The second part, "evaluation o fpublic finance management," addresses the third question. The PERhas provided a basis for analysis and action under the Public Sector Capacity BuildingProgram (PRECASP), one o f whose components i s related to the improvement o f public finance management and i s expected to support the implementationo f the measures proposed inthis document. The present review takes into consideration three analytical studies being undertaken inparallel with the PER update: the Financial Sector Assessment Program (FSAP); the Country Economic Memorandum (CEM) presenting an analysis of the challenges relatedto oil revenue management; and an exercise aimed at costing the achievement o f the Millennium Development Goals (MDGs). These analyses deal with issues closely related to this review, including (i) problems faced by the financial sector and their the impact on public finances; (ii) the issue o f transparency with regards to the management o f oil revenues, and (iii)medium-term and long-term macroeconomic projections that determine the prospects for growth and for achievingthe MDGs. Regarding the analysis o f public expenditures within each priority sector, a review o f rural sector expenditures is being completed under the guidance o f the national authorities. Its objectives are to identify ways to improve public financial management in the sector and to alleviate the constraints on apcultural development, with a view to increasing productivity and reducing poverty in rural areas. Moreover, the PRECASP will finance expenditure reviews inpriority sectors. These studies are complementary and shall provide critical insights into the respective areas, as well as recommendations that will assist the Government in making optimal use o f the available natural and human resources, lay the foundations for sustainable economic growth, and reduce poverty and inequality. vi EXECUTIVESUMMARY 1. While it had earlier appeared as though Mauritania's economic performance had been rather favorable, the data now reveals significant shortcomings since the early 1990s. An expansionist budget policy, supported by unrestrained monetary policies, turned out to be the main instrument o f economic management. The economic authorities gave priority to satisfying short-term political demands outside o f the regulatory framework, without linkages to macroeconomic planning, and without assessing the impact o f the measures on the economy as a whole and on the development objectives specified in the Poverty Reduction Strategy Paper (PRSP). The results achieved in terms o f growth and poverty reduction fell short o fprogramtargets. 2. Oil revenues, which began flowing in 2006, will progressively ease revenue constraints for approximately the next 20 years.' This temporary alleviation o f the financing constraints will not be beneficial unless Mauritania establishes adequate mechanisms for effective, efficient and transparent management o f public resources, including oil revenue. On the verge o f becoming an oil exporting country, Mauritania faces the following major tasks: (i)establishing transparent macroeconomic management, budgetary and accounting procedures to ensure long-term macroeconomic stability and transparency; (ii) diversifying its productive base to reduce its economic vulnerability and develop its non-oil sectors; and (iii) implementing public policies conducive to the achievement o f the MDGs through accelerated poverty reduction. 3. Inlight ofthe situation, the new government that took office inAugust 2005 startedto putinplace the basis for sound and transparent macroeconomic management. The new authorities have already ended off-budget spending by eliminating direct financing by the Central Bank o f Mauritania and stopping the use of treasury resources without preliminary budget credit clearance and necessary justification for the associated spending. It i s recommended to further strengthen these measures to limit the negative impact o fpast practices as the country enters into the oil era. Inmany countries that benefit fkom natural resource revenues a part o f total revenues and expenditures remain outside o f the regular budget cycle. One o f the consequences o f such practices is that the growth inrevenues does not translate into diversification o f the non-oil sectors o f the economy, and that it does not contribute to the alleviation o fpoverty and inequality. Preventinga similar situation fkom developing inMauritania requires government efforts to improve the efficiency and transparency inthe use o f these resources. 4. As for the analysis o f past expenditures inMauritania, the lack o f information makes it difficult to measure the impact o f off-budget expenditures on spending in priority sectors between 1992 and 2004. Indeed, while the new data2presents trends inthe overall volume o f expenditures, it does not allow for a breakdown by economic and functional classification. While keeping in mind these limitations, one notices the growth in unproductive expenditures related to military and security at the expense o f productive sectors such as agriculture and transport. The latter two received a lower amount o f resources while they represent important sectors for the creation o f sustainable economic growth and poverty reduction due to their positive externalities. It also becomes clear that the social sectors have benefited from larger resource allocations, while the absorptive capacity remains limited. Without an in-depth review o f intra-sectoral spending allocations, the expenditure analysis does not permit a detailed 1See Annex for a briefsurvey o f the sector and CEM2006 for more information. The expenditure analysis is related to the economic situation over the period 1992-2004 as detailed inthe data presented by the Mauritanian authorities in M a y 2006. vii evaluation o f the impact o f public spending and of the constraints to the achievements o f the expected results. 5. Military spending must be controlled to channel potential savings toward productive sectors. As a complement to this analysis, it i s recommended to conduct reviews o f public expenditures in some priority sectors such as education and health, as well as productive sectors such as rural development and transport. The review by sectors3would provide an understanding o f internal allocations and help identify the constraints that hold back the achievement o f the expected results. In the context of an increase in revenues thanks to oil production, such analyses could help find ways to improve the alignment o f budgetary allocations with sectoral policies. Moreover, they could identify solutions that strengthen the absorptive capacity o f the relevant sectors. Such an approach would also assist in the creation o f sectoral Medium-Term Expenditure Framework (MTEF) in line with the overall MTEF. Finally, the alignment would improve financing mechanisms and partnerships4for the sustainable development o f these sectors. 6. The Priority Action Plan (PAP) o f the Poverty Reduction Strategy for 2002-2004 did not serve as the basis for the planning and execution o f public investments. On the other hand, many actions that were not included in the PAP and did not contribute directly to the achievement o f the PRSP objectives were implemented. This deviation in the implementation o f the PAP was due to (i) various weaknesses that marred the preparation o f the PAP, which i s in fact a compilation o f needs identified by the sectors without a direct link to sectoral objectives; (ii) the fact that a share o f investment expenditures were executed o f fbudget; and (iii) the weak planning capacity o f the administration. 7. To ensure that the PAP serves as the basis for budgetary planning and execution, and in order for strategic objectives o f economic and social development to be achieved in a coherent manner, it i s important that (i) PAP identifies and integrates all the actions needed to deal with the key constraints the o f economic and social development in line with the sectoral strategies and MTEFs; (ii) the estimation o f the costs o f implementing the PAP calculates investments and recurrent costs based on an evaluation o f executed programs and unit costs; (iii) flexibility i s provided in the preparation and implementation process in order to take into account lessons learned and permit the addition o f new priority activities along the way; (iv) the exercise distributes the priority actions among regions in a way that addresses the specific critical constraints o f each wiluyu; (v) the PRSPS and the PAP are used as the principal source o f budget planning o fthe B C I and o fthe recurrent budget. 8. The evaluation o f Mauritania's public financial management system shows that various improvements have recently been implemented. However, major constraints remain in terms o f capacity buildingand modernization, inorder to sustain the progress made duringa short period. 9. Since the last trimester o f 2005, efforts have been made to strengthen the comprehensiveness o f the budget, thanks to (i) the adoption o f a revised budget that has made it possible to regularize off-budget spending for the year 2005; (ii)the preparation o f a budget for 2006 that i s more comprehensive and realistic; and (iii) the reclassification o f the 2006 budget on a functional basis inM a y 2006 to ensure the continued monitoring o f expenditure execution in priority sectors and other poverty-reducing activities. Despite these improvements, the existence o f separate budgets has still not been addressed, making it impossible to get a global view o f public expenditure. A 2007-2009 MTEF has been completed and is to be used inthe 2007 budget preparation. However, there still i s no clear calendar for the different stages o f the budget process, and the level o f involvement o f sectoral departments remains limited. Consideringthe recent improvements, the main recommendations o f this review are to consolidate the recent A review ofthe expenditure ofthe rural sector is under development. Itis expectedthat the PRECASP finances the realization o f the reviews of expenditure for certain priorities. Public-private, beneficiary implications, etc. V l l l ... achievements by continuing the formulation o f a comprehensive budget based on the MTEF, ensuring a better involvement o f the sectors in the budget preparation process, continuing enriching the presentation o f the budget with a breakdown according to the functional classification, and moving toward the integration o fboth budgets (Finance L a w and BCI). It i s necessary to put inplace a clear calendar for the budget planningprocess that identifies the different stages and responsibilities. 10. When it comes to the execution o f expenditures, the main recommendation o f the 2004 PER was related to a gradual deconcentration o f budget execution from the Ministry o f Finance (MF) to line ministries and finally to the regions to ensure the acceleration o f the expenditure cycle. This is the only recommendation that was implemented in 2005, but with considerable difficulties due to (i)a delayed deployment o f the necessary RACHAD software; (ii) the existence o f two types o f financial controllers (contrdleurs financiers); and (iii)the weak capacity at the line ministry level. This reform o f the assignment o fbudget execution shouldbe consolidated at the level o f the pilot ministries by strengthening sectoral capacities, determining where the line ministries' Financial Controllers (FC) should be anchored, and generalizing the use o f the RACHAD system. Gradually, the system should be extended to other ministries before being rolled out to the regional levels. The PRECASP will provide technical and financial support for this roll out. 11. Since January 2006, treasury balances have been prepared on a monthly basis. In addition, accounting information from the treasury is reconciled with that from the Central Bank (BCM) on a weekly basis. These two actions represent important improvements and indicate that better public accounting practices are now inplace. It i s important to institutionalize these new initiatives in order to avoid discontinuity in the event o f leadership changes in the MF and BCM. The improvement o f public accounting capacities also represents an important element. 12. The preparation o f comprehensive management accounts (comptes de gestion) has not been implementedyet due to the limitations posed by current accounting methods and practices not inline with regulations. The lack o f a dedicated unit exclusively responsible for.the production of management accounts and o f evaluation reports means that these documents are not prepared on a regular basis. The progress made, including the preparation o f the treasury balance and the closure o f account within the mandatedtimeline, should contribute to the timely preparation o fmanagement accounts inthe future. 13. A reorganization o fthe DGTCPto break up the different functions o f the treasury as recommended by the 2004 PER should be implemented. In the short term, it is recommended to create a main accounting entity incharge o f expenditures. A complete reorganization could be envisaged to establish (i) a department o f the treasury and public accounting responsible for managing the state treasury and for producing accounts and summary statements; and (ii)for external Treasury offices, a network o f government accountants subordinated to the central department. 14. Regarding budgetary control, while the rules and procedures are quite clear, the PER update revealed deficiencies inthe internal system o f control and verification due to the multiple parties involved inthe process, andthe lack o fregular supervision andappropriate sanctions. The PERupdate also showed that the capacity o f the control entities i s limited. Strengthening capacity o f certain control entities has been recommended and implemented, partly with the assistance of the German and French development cooperation agencies. With the recent creation o f the General Inspectorate o f the State (IGE), there now i s a third control entity. It i s highly desirable that the different control units consult before finalizing their annual programs so as to avoid overlap. Inaddition, better consultation would allow a wider coverage and a better use o f the limited resources. In many cases, simple revision of underlying texts and procedures could reduce considerably the possibilities for misuse and corruption. ix 15. The government has initiated a reform of the procurement system to break up the functions o f the Central Procurement Commission (CCM) (regulation, selection, control and appeal) in line with the 2004 PER and the Country Procurement Assessment Report (CPAR). In order to succeed, it i s important to involve all relevant parties in the different stages o f the revision o f the procurement code in order to ensure ownership on the part o f those in charge o f applying the code. The rules to be defined must be such that they can be executed impartially, in order to contribute to a stable and well regulated environment, to foster competition, to prevent corruption, and to promote transparency and accountability inthe managemento fpublicresources. 16. Inview ofthe transitional nature' ofthe current Government, the authorities didnot seemready to formulate a medium-term strategy for reforms in the public financial management system. Defining the different stages and responsibilities o f such reform program would be strategically important. A Public Expenditure and Financial Accountability (PEFA) assessment should be scheduled after the transition period with a view to complement this public expenditure review to assess progress and lay the foundations for a reform strategy. The transition period runs through March 2007 and comprises the following main stages: Constitutional referendumin June 2006, municipal and legislative elections inNovember 2006, senate elections in January 2007, andpresidential electionsat the endofMarch2007. X 4 3 E: Lr, 2 a w W v) - n c C .3 x I . I . PART I CONTEXT AND PUBLIC EXPENDITUREANALYSIS xii 1. RECENTECONOMICDEVELOPMENTSAND THEIR IMPACT 1.1. This chapter provides an overview o f Mauritania's public financial management. After a general description o f the context, the chapter briefly discusses the country's economic performance for the past few years after the discovery o f the macro-budgetary slippages. The chapter ends with an analysis o f the evolution o fthe mainpoverty indicators following the implementation o f the PRSP action plan. A. CONTEXT 1.2. Mauritania i s a West Afncan country covering6 1,030,700 km2and home to 2.8 millions people. Over 65 percent o f the population lives inurban settings, o f which 600,000 inthe capital Nouakchott. The relative political stability that had prevailed inMauritania since 1990 deteriorated progressively, resulting in attempted coups d'Ctats between 2003 and 2005. After a coup d'Ctat in August 2005, a Military Council for Justice and Democracy (CMJD) was set up to ensure a 19-month transition towards the re- establishment o f democratic institutions.' 1.3. Inthe 1990s, the Mauritanianeconomy changed significantly as aresult ofreforms inthe context o f adjustment programs. The period was characterized by the privatization o f most public enterprises, the beginning o f the diversification o f irrigation-based agriculture, an improvement in certain aspects o f competitiveness, and a boost in the service sector. However, some structural weaknesses which already existed inthat period still persist today, on the eve o f the country's becoming an oil exporter. The relative progress achieved in the 1990s was undermined by considerable slippages as a result o f opaque macroeconomic management in the period 1992-2005. As a result, the country's economic performance has been very irregular and`fallen short o f the targets. The implementation o f the PRSP, prepared in2001, took place in the context o f an expansionist spending policy which was partly carried out off-budget and which led to an increase in money supply and inflation, chronic budget deficits, and deterioration in the balance o fpayments. 1.4. Annual growth in the period 2000-2004 was on average 3.3 percent, below the initial target o f 7 percent but higher than the average for Sub-Saharan Afnca. Growth reached its lowest point with 1.1 percent in 2002. The services sector contributed significantly to growth, especially thanks to the boom in telecommunications. Apart from the services sector, the construction sector also contributed to growth, despite the limited execution capacity o f the sector. The jigsaw pattern o f the primary sector's contribution to growth was due to the vulnerability o f Mauritania's agriculture to exogenous shocks, weather conditions, and scourges, several o f which (inparticular, cold rains that decimated the livestock, drought, and migratory locusts) ravaged the agricultural areas between 2001 and 2004, adversely affecting agricultural output. The contribution o f the mining sector strongly depends on international demand and prices for iron ore, which are determined by the international market. Despite favorable market conditions, the contribution o f the mining sector has been weak because o f capacity problems affecting the production o f the SNIM'. The other strong contributor to economic growth, the fishing sector, i s facing significant problems due to overexploitation, lack o f proper infrastructure for unloading, treatment and processing o f fish. The private sector, which i s considered a crucial factor for growth, faces problems related to regulation and operates ina generally unfavorable environment. 6More than 80% o f the country area i s occupied by the desert: only the southern end has sufficient precipitation to allow the growth o f vegetation. 7 'AThe new President was elected on March2007 through fair and open elections. contribution of the miningsector to the growth has experienced some declines in2001 & 2002. 1 1.5. The scarcity o f resources, which has affected the economy since its independence and made it dependent on external financing i s expected to abate progressively, thanks to the windfall revenues from oil production. Infact, inthe Chinguetti off-shore oil field extractionbegan inFebruary 2006; however, production forecasts are less than the expected 75,000 barreldday. Oil-related research, promotion and exploration activities are being intensified and are expected to lead to the exploitation o f additional fields inthe short term. Exploitation of other off-shore reserves is expected to start inthe next few years. The largest oil field, Thof, should be operational by 2010. As a result o f oil revenues, per capita GNP i s expected to double. However, the country hasjoined the group o f oil producing countries unprepared and inparticular without a strategic vision regardingthe oil sector. Buildingnational capacity inthe relevant areas (exploration, negotiation and monitoring o f production-sharing contracts, etc.), preparing national enterprises for the market opportunities that are opening up, and formulating a strategy for using the expected additional state revenue are areas that have not received adequate attention. The oil sector and its impact on the Mauritanian economy are reviewed ina country economic memorandum (CEM) that has recently been completed. B. MACROECONOMIC SLIPPAGES 1.6. Mauritania's economic authorities implemented expansionist budgetary policies without taking into account the sustainability o f expenditures and their impact on the economy as a whole and living conditions. The main question i s whether they had comprehensive information that would have allowed them to assess the effects o f their decisions. The likely answer i s no, as on the one hand, the practices used certainly precluded the availability o f timely and comprehensive information that could have informed the decision makers. On the other hand, decision makers changed frequently in the period 1992- 2004. Furthermore, there were deliberate attempts to shelter public financial management from all public scrutiny and to respond to short-term political priorities, without transparency and an assessment o f the effects on the macroeconomic balance. Finally, it is obvious that this management was not based on any short- and medium-term macroeconomic planning. Evaluation of the main macroeconomicaggregates 1.7. The latest economic statisticsg show that the Mauritanian authorities followed procedures that ledto major deviations fkom policies aimed at macro-budgetary stability, which the government was - intheory -committed to through the PRSPand its earlier public policies. Several macroeconomic aggregates began to gradually deteriorate in 1992, culminating-in 2003 and 2004-in the significant acceleration of inflation, successive budget deficits, and a gradual deterioration o f the external position. 9Publicly disclosed by the Mauritanian authorities inMay 2006. 2 Box 1.1: Crisisof Confidencebetweenthe Donor Community andMauritania In2003-2005 acrisis ofconfidence affected relationsbetweenMauritania andthe BretonWoods Institutions inview o f macroeconomic slippages and misreporting. In November 2004, the Mauritanian authorities requested the cancellation o f the Poverty Reduction and Growth Facility (PRGF)" concluded with the IMF. In2004, the World Bank suspended the preparation o f the first Poverty Reduction Support Credit (PRSC)." That situation prevented other donors, in particular the African Development Bank (AfDB) and the European Commission (EC), from providing budget support for Mauritania. The new authorities agreed to revise the economic statistics for the period 1992-2004, to have the BCM's financial statement examined by independent auditors as o f the end o f 2003, and to make available the information required for determining actual official reserves. These measures allowed a staff monitored program (SMP) to be concluded between Mauritania and the IMF for the first semester o f 2006. The implementation o f the SMP was successfully reviewed for the fust time inApril 2006. A three-year arrangement under the PRGF has been agreed inDecember 2007 between Mauritanian and the IMF.These contribute to a revival o f donor confidence inthe new Government's economic management. 1.8. As a result of heavy off-budget spending, the budget (excluding grants) ran successive annual deficits averaging 11percent o f GDP between2000 and 2004, reaching a peak in 200312with a deficit o f 16.4 percent o f GDP. These deficits were financed mainly by the BCM through monetary emissions. Official reserves fell to the equivalent (according to present estimations) to half a month's imports at the end o f 2004. The budget deficit inthe period 2000-2004 was financed by up to 4 percent o f GDP through access to exceptional external financing. Traditional budget support gradually declined, while reducing the pressure on the balance o f payments and financing a significant part o f imports. Internal financing amounted to an annual average o f 4 percent o f GDP during the same period. One result o f the slippages was the accumulation o fpayment arrears, which the new authorities are inthe process o f absorbing. 1.9. Despite the macroeconomic slippages, Mauritania has continuedto ensure the sustainability of its foreign debt13.The ratio of external public debt to GDP was estimated at 181.4percent in 2001, falling to 109.1 percent in 2005. Mauritania was one o f the countries benefiting from High Indebted Poor Country (HPC) debt relief initiative, as well as the Multilateral Debt Relief Initiative (MDRI). Various issues14 continue to affect external debt management in Mauritania such as a lack o f coordination between the parties involved, the lack o f a computerized and unified management system and poor technical capacity. 1-10. Extensive spending, financed through money creation, had a negative impact on prices. They rose significantly, notwithstanding projections in the PRSP" which aimed to contain inflation in order to reduce poverty and inequality. Infact, inflation during the period accelerated from 3.3 percent in 2000 to 10.4 percent in 2004 and 12.2 percent in 2005. Apart from monetary expansion and the depreciation o f the ouguiya (UM) against the Euro,16 the above trend was also due to the poor harvest in 2001 and the worldwide increase in oil prices. The external position (excluding imports) also experienced a shock and was in deficit throughout the period. The current account (excluding imports related to oil exploitation) was incontinuous deficit, peaking at 34.6 percent o f GDP in2004. lo The PRGF was concluded inJuly 2003. 11 The Coui-~tryAssistance Strategy (CAS) provided for a series of three PRSCs but the World Bank extended no budget support to the country inthe period 2002-2005. 12 Itwas the year ofthe emergency implementation plan. 13 Mauritania reached the HIPC Completion Point inJune 2002, which allowed Mauritania to reduce its debt service from 23 percent o f the exports in2000 to 9 percent in2004. 14.Installation o f the software "Systbme de gestion automatiste de la dette exttrieure" (SYGADE) at the B C M and MFlevelis envisagedinthe short term. 15 Based on the assumption o f a 2.4 percent inflationrate in2004. l6 Halfo fMauritania's imports originate inthe euro area. 3 1.11. Between2000 and2004, the money supply increasedby 23.3 percent. Internal credit also increased: government borrowing soared from UM 31.4 billion to UM 129.9 billion during the same period. The measures taken, namely, the establishment o f a monetary committee at the BCM and the use o f indirect monetary instruments", failed to lead to prudent monetary management. The instruments designed to strenglhen exchange rate policy have not come into effect. In fact, the planned creation o f the Extended Exchange Market (MCE) proved unable to reduce the premium in the parallel exchange market, thereby reflecting dwindling confidence in the national currency. The latter steadily depreciated against the Euro starting at mid-2001, following the movements in the Euro/dollar exchange rate. The ouguiya, on the other hand, remained relatively stable vis-&vis the dollar between 2003 and 2004, moving between UM 255 and 270 for US$1. Box 1.2: Current Monetary and ExchangeRate Policy Measures The current government is committed to redressing the macroeconomic situation through corrective measures aimed at sustainable growth and the optimized use o f oil revenue. Tight monetary policy measures have been adopted and the BCM-with a view to containing inflation and boosting the accumulation o f official exchange reserves-has pledged to abstain from direct financing o f the budget deficit. Inline with these objectives, the BCM continues to rely o n the obligatory reserve rate, to intervene inthe foreign exchange market inorder to contain the real effective exchange rate (REER), and to make use o fthe reference rate, which has been raised twice (inFebruary and October 2005). Evolution of publicfinances 1.12. A significant increase in revenue has partly offset the negative effects of expansionist budget policies. In the period 2000-2004, as a percentage o f GDP, government revenue as a whole (excluding grants) soared from 21.8 to 29.7 percent. This increase was partially due to an increase in tax revenues, which rose from 13.7 to 14.9 percent o f GDP. As the table below shows, Mauritania's tax revenues largely exceeded the average for low-income countries, although they lagged behind the tax revenues o f more developed economies, such as Tunisia and South Africa. Lower tax rates and simpler procedures were among the factors responsible for the increase intax revenue. The tax measures taken are described inAnnex 2. Inparticular, the tax base was broadened and collection clearly improved, particularly from the domestic value-added tax (VAT), income from the National Industrial and MiningCompany (SNIM), custom duties, and the real estate tax. Table 1.1 :Tax Revenues (inpercentage of GDP) 2003 2004 Mauritania 13 15 South Africa 24 26 Tunisia 21 21 L o w income countries 10 10 Source : WDI 17Interest rate management (determination by the BCM o f a scale o f rates depending o n loan duration and use) was gradually abandoned. Currently rates are fully liberalized and subject only to a ceiling equal to the reference rate, which is established by the BCM as a function of the spread betweenpurchase-and-resale contracts and government bonds. Set at 11 percent in October 2001, it i s currently at 14 percent, subsequent to recent raises decided by the BCM and approved by the Monetary and Exchange Policy Committee (CPMC), plus ten percentage points (10 percent) and an 8 percent minimumrate on savings deposits. 4 1.13. Despite increased revenue, tax administration still faces significant challenges. The tax base remains limitedbecause o f tax evasion by many firms and especially with regard to the informal sector. This situation is mainly a result o f human and material inadequacies in the General Tax Directorate (DGI), the chronic weakness of the taxpayer information system, the inadequate monitoring of firms not subject to real profit taxation, and the weak yield on real estate taxation. Customs performance i s constrainedby the system o f exemptions, the complexity o f procedures, and the inadequacy o f anti-fraud resources. Taxation o f oil companies is expected to be an additional challenge. 1.14. In the period 2000-2004, non-tax revenue doubled, exceeding 14 percent of GDP in 2004 particularly as a result of payments ( 86 million per year) received under a fishmg agreement between Mauritania and the EuropeanUnion (EU).However, part o fthis revenue (on average, 25 percent per year inthe period2002-2004) is off-budget. Off-budget non-tax revenuereached2.7 percent of GDPperyear inthe same period and was used partly to finance off-budget expenditures. However, the 2005 budget comprises the full amount receivedunder the above fishingagreement. 1.15. As a percentage of GDP, off-budget expenditures increased from 24.2 percent in 1995 to approximately 30.7 percent in 2000 and 47.2 percent in 2003, before falling to 37.7 percent in 2004. Recurrent expenditures were mainly boosted by spending on goods and services, while investment expenditures increased less than expected. Public expenditures are analyzed ingreater detail inChapters 2 and 3. c. POVERTYAND SOCIAL INDICATORS 1.16. In the period 2000-2004, rising public expenditures-in a context o f opaque budget execution practices characterized by overspending-failed to contribute significantly to poverty reduction and improvement o f the living conditions of the vulnerable sections o f the population. As the analysis below shows, save for the education sector, government policies inthe periodunder consideration didnot leadto the achievement o f the development objectives presented inthe PRSP. Figure 1.1: Evolution of Poverty Incidence, 2000-2004 1.17. The revised economic statistics reveal inflation rates higher than those previously published. The poverty line I , criteria used in household surveys were I 1 2000 5 1 0 i I I '*Povertyincidence recededby 3.3 percentage points inNouakchott but increased inthe other towns. 5 1.18. Poverty reduction was weaker than expected mainly because economic growth fell short o f the ambitious objectives set and was not accompanied by sufficient redistribution, despite the implementation o f some targeted poverty-reducing projects addressing the multidimensional character o f poverty. Rising inflation, caused by government overspending, as well as inclement weather and the rise in oil prices undermined gains inhousehold purchasingpower, particularly among the more vulnerable groups. Inthe rural areas, poverty reduction was slowed down by the population's nearly full dependence on farming and livestock, which are sensitive to weather conditions. In the urban areas, poverty reduction was impeded by the deterioration o f the generally substandard living conditions in the h n g e districts as a result o f the uncontrolledinflow o fnew settlers. 1.19. As shown in the table o f poverty and welfare indicators in Annex 3, access to basic education improved significantly, reaching 95 percent in 2004, while girls spectacularly caught up with boys inthat respe~t'~.The gross enrollment rate (GER) in secondary education also increased substantially (from 19.4 percent in2000 to 29.6 percent in2004), while remaining slightly higher among boys. On the other hand, children from better-off families remain at an advantage2'. Furthermore, the quality o f education has not improved as much as access, there i s a high repetition rate in basic education, and the retention and howledge acquisition ratesremain weak. This situation i s partly due to the following factors: . . Highdrop out rates inareas with schools offering an incomplete curriculum; The inefficient introduction o f bilingual education, as part o f the 1999 reforms, due to inadequate . capacities; Teachers who do not stay where they are appointed, as a result o f low remuneration, poor organization and a lack o f control. 1.20. On the whole, health indicators stagnated despite the implementation o f a national health policy and the mobilization o f substantial financing for the health secto?. The causes, linked directly to the management o fhealth services, includedthe following factors: Inadequate qualification and distribution o f health staff as a result of insufficient training, ... remuneration, and incentives; Poor health care equipment; Lack o f quality control o fhealth services; = Inadequateprevention and treatment o f infectious diseases; Limited availability o f essential medicines. 1.21. Access to dnnlung water improved noticeably between 2000 and 2004. The percentage o f households with domestic water taps (as opposed to those supplied from wells or by street vendors) increased. Disparities persist between urban and rural regions and between rich and poor. For instance, the price o f water charged by street vendors, the main source o f supply inpoor neighborhoods, i s tenfold the price paid by consumers connected to the National Water Company (SNDE) network. Inrural areas, only 12percent o fhouseholds have access to safe water sources. l9Contrary to the situation in2000, in2004 the GER was higher among girls (by 4.5 percentage points). 2oThe gap between the GERs of the first and the fifth income quintiles was 40 percentage points (50 percentage points among girls). *'Iti s estimated that realpublic expenditures o n health stagnated. 6 D. CONCLUSIONSAND RECOMMENDATIONS 1.22. On the eve of becoming an oil exporting country, Mauritania i s still confronted with major challenges that are mainly related to its capacity to ensure an adequate management o f public finances. The constraints exist at various levels and require the following measures: The establishment o f transparent mechanisms for macroeconomic, budgetary and accounting policies inorder to ensure sustainable macroeconomic stability; The diversification of production in order to reduce the vulnerability o f the economy and to . develop the non-oil sectors; and The implementation o f government policies aimed at accelerating poverty reduction and, by the same token, the achievement o fthe MDGs. 1.23. Oil revenue will progressively ease one o f the constraints, namely, fund availability, for approximately 20 years." Mauritania must rise to the challenge posed by that windfall by establishing adequate mechanisms for the rational management o f public resources, including oil revenue. The mobilization o f tax revenue should be intensified and an adequate tax base ensured, compatible with private sector development, so that public spending relies more on stable revenue sources, not volatile ones like oil. Monetary and budget policies should aim at controlling inflation and ensuring a real effective exchange rate (REER) favorable for Mauritania's competitiveness. Lastly, economic management should be based on comprehensive information and on assessments of the effect o f any decisions on short- and medium-term macroeconomic balances. 22Projected duration o foil exploitation. For more information, refer to the CEMupdate. 7 2. PUBLICEXPENDITUREANALYSIS 2.1 This chapter uses an economic and functional classification to analyze the development and composition o f government spending in Mauritania. This analysis i s based on the new expenditure data provided inMay 2006 by the authorities following the revision o f the national economic statistics for the period 1992-2004. As government expenditures are largely executed under the responsibility o f the central authorities, examination o f expenditures at the local level (regions, communes, or villages) is not necessary for the purposes o f this report. In fact, total transfers to local government units, called Local Development Funds (FDLs), accounted for 1.2 percent o f the state budget. Most o f the expenditures were committedby the GIRM and its regional offices. 2.2 The analysis o f public expenditures was implemented under a number o f constraints, involving in particular the need to reconstruct the national budget statistics, essential to reviewing the development and composition o f expenditures. Such challenges, encountered at various levels, exist in other developing countries as well but are more pronounced in Mauritania compared to the countries o f the Maghreb or the CFA Franc (CFAF) Zone, whose statistical information systems are considered to be more reliable. 2.3 The main challenge was the long and systematic practice o f executing public expenditures outside the budget and the complete lack o f transparency. The lack o f supporting documents and the involvement o f various bodies inthe process prevented full tracking o f the expenditures inquestion. Consequently, the volume o f some o f these expenditures and their composition according to international classifications could only be approximately determined. Moreover, some departments, such as the Ministry o f Fishing and Maritime Economy (MPEM) and the Ministry of Mining and Industry (MMI) spend directly part of the revenue derived from the exploitation o f the respective resources. The volume o f spending executed at that level was probably not high, but the practice affected the estimation o f total expenditures to be analyzed. Furthermore, the presentation and monitoring o f the budget were not based on a functional clas~ification?~ A functional classification was formally introducedin2001, but has not been usedinthe preparation o fthe state budget. A bridge table was prepared for the 2004 PER with a view to reprocessing part o f the government expenditures and classifying them according to the functional classification. Nevertheless, this exercise has not made it possible to process a significant volume o f expenditures, especially those outside o f the budget. A. (THEIMPACTOFOFF-BUDGETEXPENDITURES TOTALGOVERNMENT ON SPENDING 2.4 Between 1995 and 2004, total government expenditures increased by approximately 236 percent innominalterms, from UM44 billionto UM 149.6 billion. As apercentage ofGDP, they increasedfrom 24.2 percent in 1995 to 47.2 percent in 2003 and 37.7 percent in 2004. Government expenditures in Mauritania are higher than their average in Sub Saharan countries, where such spending amounts to approximately26 percent of GDP. 232006budgetwas re-elaborated infunctional classification (type COFOG) 8 Figure 2.1: Off-Budget Expenditures (%of GDP) 37 5 1995-99 2000-04 Source: MF 2.5 Extra-budgetary expenditures during the period under consideration were a significant factor in the increase in overall government spending. In 2003, the year in which the Emergency Plan was implemented, they accounted for 44 percent o f total expenditures, compared to 20 percent in 1995. As shown inthe figure, off-budget expenditures accounted on average for approximately 6 percent o f GDP during the period 1995-1999 and for more than 13.5 percent o f GDP in the period 2000 to 2004. As a consequence o f this high volume o f off-budget expenditures, the overall balance has experienced a continuous deficit with an average o f 11percent o f GDP for the period 2000-2004 and 5 percent between 1995 and 1999. Figure2.2: PublicRevenuesandExpenditures (Yo of GDP) I 60.0 j 1 ~ Revenuesexiudinggrants 50.0 A I ? 40.0 30.0 20.0 10.0 0.0 zoos] 2ooH z o o e ] 120ou -10.0 -8.9 -1.2 -8.0 -14.4 U -16.4 Source: MF 2.6 The breakdown o f off-budget expenditures provided by the authorities has not been sufficiently detailed. Since they are executed without any transparency, they may well have been an important source o f corruption especially in public procurement. It appears that off-budget spending began in the early 1990s but it has been difficult to estimate its volume between 1992 and 199424.Inthe period 1995-2003, off-budget expenditures increased from 21.5 to 44.9 percent o f total government spending and, as a 24 Itappears as though off-budget spending accounted for 6 percent of GDP in 1993 and 1994. 9 percentage o f GDP, from 4.9 to 20.6 percent. Inthe period 1995-2004, they accounted on average for 9.7 percent o f annual GDP. As shown in the table below, approximately 75 percent o f extra-budgetary expenditures were carried out through direct debit authorizations (LDAs) and financed directly by the BCM. Box2.1:Method ofDisbursingOff-Budget'Expenditures LDAs are one method of disbursing off-budget expenditures. These expenditureauthorizations are addressedby the Directorate of Budget and Accounts (DBC) to the Treasurer Generalfor immediatepayment. On that basis, the Treasury makes disbursementswithout requiringany prior budget allocation.According to regulations, LDAs should subsequentlybe settledby budget warrants. Approximately 37 percent of expenditures made under the 2003 and 2004 budgets were disbursed by means of LDAs. A breakdown according to the functional classification of the expendituresexecutedthrough LDAsbetween2000and2004 is attachedinthe annex. As the table shows, the BCM was the main source of financing for extra-budgetaryexpenditures.It carried out orders that it received from the highauthorities and didnotdirectlycooperatewith the DBC or the Treasury.About 47 percent of the off-budgetexpenditureshavebeenfinanced by the BCM. The BCM funding includes, among others, expenditures for the `fonds de contreparties' managed by the Ministry of Economic Affairs and Developmentbut locatedat the BCM. Analysis of extra-budgetary expenditures, 1995-2004 1995 1996 1991 1998 1999 2000 2001 2002 2003 2004 Offbudget expenditures (in % GDP) 4.9 6.8 5.4 5.5 1.3 9.0 13.8 11.5 20.6 12.7 Offbudget expenditures (in % totalofGov expend.) 21.5 28.6 23.9 24.1 30.2 30.6 40.1 31.5 44.9 34.8 Depenses hors budget en Milliards d ` U M Off budget expenditures 9.0 13.5 11.4 12.7 18.2 23.1 39.5 35.9 69.5 50.6 Direct debit authorization (LDA) 0.8 1.2 1.6 2.6 4.9 7.6 6.6 24.8 19.1 Emergency Plan 15.7 1.9 BCM 5.7 9.6 5.8 6.3 9.0 10.9 23.4 13.6 11.9 12.0 Pension fund deficit 0.3 0.4 0.4 0.5 0.6 0.1 0.8 0.9 1.1 1.2 Exchange loss 2.2 2.0 3.3 3.0 4.4 4.2 4.5 4.5 6.9 1.4 lntere~ton government bonds 0.5 0.5 0.5 1.1 1.1 1.1 1.1 1.1 1.1 1.1 lntcrcrl on OvcrdraRr 0.2 0.2 0.1 0.3 0.5 1.0 1.1 0.1 1.6 3.6 Instances of payment (PU+Ent. Publ.) 0.1 0.4 1.1 8.5 6.4 4.2 Ventilation des depenses hors budget selon le mode d'exbcutlon et type de depenses couvertes (en % des dhpenses hors budget) Direct debit authorization (LDA) 6.0 10.8 12.2 14.2 21.2 19.1 18.5 35.7 37.9 Emergency Plan 22.6 3.8 BCM 63.6 71.1 50.8 49.8 49.4 47.2 59.2 37.9 11.1 23.1 Pension fund deficit 3.9 2.6 3.3 3.1 3.1 2.9 1.9 2.6 1.6 2.4 Exchange loss 24.9 15.1 29.1 23.5 24.0 18.0 11.4 12.5 9.9 14.6 lntcrcrt on government bonds 6.0 4.0 4 . 1 8.4 5.9 4.6 2.7 3.0 1.5 2.1 lntercst on overdraflr 1.1 1.2 1.2 2.3 3.0 4.5 2.9 1.8 2.3 1.1 Instances of payment (PU+Ent. Publ.) 0.4 1.1 2.8 23.6 9.3 8.3 Source : MF A significant volume o f off-budget expenditures corresponds to foreign exchange losses related to overseas spendi@. The gap inquestion resulted fromthe calculation, until2005, ofthe costs ofoverseas expenditures on the basis ofanexchange rate, called the "chancellery rate," applied since 197326 without taking into account any subsequent depreciation o f the ouguiya. Consequently, budget appropriations for the expenditures in question were below actual costs. On average, such foreign exchange losses accounted for more than 18 percent o f off-budget spending. The emergency planlaunched in2003 further bloated government expenditures. The plan's objective was to provide assistance to population groups that were affected by drought and cold rains. In 2003, the plan accounted for approximately 22.6 percent o f off-budget expenditures. Successive deficits o f the retirement pension fund also did not appear in the LdF and were covered by off-budeet exoenditures. Interest on eovemment bonds and bank overdrafts were funded directlv as well. albeit to a lesser extent. 2.7 Inorder to break away from the badpractices, the adoption ofthe new Finance Law of 2006 - which i s more realistic than previous finance laws- includes all expenditures. In addition, chancellery rates corresponding to real exchange rates have been used for internationally executed expenditures. No 25 Embassies' operational costs and scholarships. 26 Inthe period 1973-2005, the "chancellery rate" remained constant at US$l =UM40. 10 off-budget issues have been raised by the authorities with regards to the financial statements for the 1st four months o f 2006. Box 2.2: ExpenditureFunds There i s a growing demand to create (( funds)) for ministries to provide more management flexibility. Considering the volume o f the amounts (often higher than the regulatory standards), the confusion o f responsibilities between the requestor and the allocating unit (it i s not uncommon to find a Secretary General nominated as the allocating responsible) and the lack o f preliminary authorization by the FC for payments, there is however the danger o f future deviations. The implementation o f RACHAD should limit such irregular activities, often blamed on excessively slow processing time. T o avoid any problems, it i s important to ensure that such procedures be used only for small expenditures (as a principle, no procurement should be financed by such funds), that expenses are regularly reconciled withm the complete file and that the separation o f responsibilities between the requesting and allocating entity is maintained. B. BREAKDOWN EXPENDITURES OF ACCORDING TO ECONOMICCLASSIFICATION 2.8 Incomparison with other developing countries, expenditures inMauritania are rather high.The ratio o f public spending to GDP i s higher than in the countries o f Sub-Saharan Africa as well as in Mauritania's northern neighbors. While the growing volume o f non-productive expenditures such as military expenditure i s specific to Mauritania, the other expenditure trends follow a similar path with those countries at similar levels o f development. Inother words, as i s common in developing countries, we observe a growth inthe expenditures for goods and services as far as recurrent spending i s concerned, and an increase ininvestment spending inthe social sectors at the expense o f the productive,sectors o f the economy. These sectors, which rely on private investment via the banks, thus have access to limited public resources. They are unable to take o f f and play their role as drivers o f economic growth and as a key element inpolicies aimed at reducing poverty. 2.9 According to the table below, government expenditures in almost all categories increased significantly. In the period 1995-2004, the budget increased by 16 percent per year. Like in other developing countries, recurrent expenditures continue to represent the largest part o f the state budget, equivalent to an average o f 68 percent for the period between 1995 and 2004. These recurrent expenditures have experienced a more rapid growth than overall spending. When it comes to investment expenditures, the average annual increase o f 22 percent peaked in 2000 and 2001 with record growth o f 59 percent and 50 percent, respectively. These peaks are due to the availability o f HIPC resources during those years. Goods and sewices 2.10 Government expenditures on goods and services are crucial for the effectiveness and productivity o f public administration. Amounting to more than 22 percent o f total expenditures, they constituted the largest recurrent expenditure category. In the period 1995-2004, however, 22 percent o f expenditures on goods and services were executed outside the budget, through LDAs, direct B C M financing, and on the retirement pension hnd deficit. Moreover, despite the relative size o fthis expenditure category - aimed at improving the functioning o f the administration -- the work environment o f civil servants remains substandard as a result o fthe inadequacy and inefficient use o f the available resources. 2.1 1 A significant share o f expenditures on goods and services was not allocated to the various ministries. Instead, the envelope of "common expenditures," which until2005 amounted to approximately 11 30 percent o f the annual general budge?', was executed by the MF.It generally served to provide budget extensions to some departments, meet some expenditures o f the Ministry o f Rural Development and Environment (MDRE) aimed at addressing exogenous shocks frequently affecting the agricultural sector, finance transportation and treatment o f poor patients abroad, etc. It has been difficult to obtain detailed information on the execution o f these common expenditures and to asses their impact. 2.12 A more detailed analysis o f the various subcategories o f expenditures on goods and services mightrevealthe main causes for their inefficient utilization. In2003, government expenditures on goods and services in Tunisia and South Africa accounted, respectively, for 8 and 12 percent o f total expenditures, compared to Mauritania's 22 percen?. Yet the quality o f public administration inthese two countries i s far superior. Treatment of salaries 2.13 Inthe period 1995-2004, the total wage and salary billremained a remarkably stable percentage o f GDP, averaging approximately 4.6 percent. But as a percentage o f total expenditures, it decreased from 18.9 percent in 1995 to 11.5 percent in 2004. It is difficult to determine whether Mauritania could have reigned in salary spending and to compare the situation inMauritania with that o f neighboring countries. This is because some salary expenditures are included elsewhere, namely under (i) military spending which finances directly the salaries o f military personnel; (ii) spending on goods and services, which covers the salaries o f contract workers; and (iii)off-budget spending, which financed supplementary salaries for embassy personnel. Inthe future, it would be useful to integrate all personnel spending under salary expenditures inorder to better analyze this type o f spending inits entirety. Box 2.3: Low Civil Service Salaries: A Disincentive Although the PER does not comprise a detailed analysis o f the public service situation, it i s quite obvious that the discouraging level o f civil service salaries is one o f the causes o f inefficiencies in public administration. In fact, the gradual deterioration o f the wage and salary scale (taking into account all categories) i s the main factor behmd poor performance and lack o f conscientiousness among state employees. While their income diminished inreal terms, the salary structure was compressed, resulting in remuneration differences between grades. Attracting and retaining qualified civil service staff i s problematic. Fringe benefits (temporary accommodation, car, etc.) have been reduced and tend to be offered inequitably, primarilybenefiting senior officers. An expert paid by external resources inthe context o f a project earns up to 8 times more than a manager inthe public service. Inaddition, career prospects are not based on objective criteria such as performance, but rather on client relationships. The weak motivation o f civil servants is also a factor inthe proliferation o f corruption. Successive wage and salary raises were partly offset by inflation. An increase in civil servant salaries is foreseen in order to make the civil service more attractive to prospective candidates, and to reduce the incentives for corruption. Military expenditures 2.14 Contrary to the picture obtained from earlier analyses, particularly the 2004 PER, military expenditures2' have not leveled off. Inthe period 1995-2004, they increased at an average annual rate o f 21 percent. Their considerable increase in 2003 (67 percent) was mainly funded through off-budget . expenditures. As a percentage o f GDP, Mauritania's military expenditures in 2003 largely exceeded the average for low income countries, but were slightly below those o f neighboring Morocco3'. The rise in 21 In2006, these expenditures account for approximately 12percent o fthe budget. 28 A rate comparableto Madagascar's (20 percent). 29 Military expenditures are the thirdcategory o f expenditures over the period 1995-2004. 30 Highmilitary expenditures inMorocco may be due to the Western Sahara conflict. 12 militaryexpenditures was partly due to measures taken against several attempted coups d'ktat, significant pay raises, and improvements in the working conditions of the military. In the period 1995-2004, on average 47 percent of annual military expenditures were executed outside the budget, mainly through LDAs (24 percent of military expenditures) and by direct BCM financing (18 percent). Although still high, the share of the general budget allocated to military expenditure, which was 12.8 percent in2004, decreased incomparison to 1992 (27.5 percent). Table 2.1: Military Expenditures in Selected Countries (inpercentageof GDP) 2003 Low income countries(average) 2 Mali 2 Mauritania 5 Chad 3 Senegal 4 South Africa 1 Morocco 6 Tunisia 2 Source: WDI 13 Table 2.2: Economic Classification of Government Expenditures 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 Inbillionof MUM Total expenditures 44.5 48.7 48.9 55.1 62.5 78.0 100.0 117.3 159.5 149.6 Total off-budget expenditures 9.0 13.5 11.4 12.7 18.2 23.1 39.5 35.9 69.5 50.6 Recurrent expenditures 29.2 31.8 35.9 37.5 42.0 52.9 60.0 69.8 105.0 96.7 Wages andsalaries 8.4 8.8 10.4 10.0 10.4 11.8 12.8 16.1 16.0 17.2 Expenditureon goods and services 8.4 9.2 10.5 12.5 13.6 17.9 20.2 27.8 35.3 36.9 Subsidies 2.7 3.0 3.4 3.6 4.3 6.0 5.9 7.8 26.3 9.6 Military expenditures 4.2 5.2 6.3 4.8 6.7 9.0 13.3 9.9 16.4 18.6 Govemment debt interm 3.9 4.3 4.5 5.8 6.4 7.8 7.7 7.7 9.3 11.9 External 3.2 3.6 3.8 4.4 4.8 5.7 5.5 5.9 6.6 7.3 Intmal 0.7 0.7 0.7 1.4 1.6 2.1 2.2 1.7 2.6 4.7 Special accounts 1.7 1.3 0.9 0.7 0.6 0.4 0.1 0.6 1.7 2.4 Investment expendituresand ner loans 10.0 8.5 9.5 12.3 13.7 20.5 30.0 41.4 45.3 43.2 Investmentexpenditures 10.0 8.5 9.5 12.3 13.7 20.5 29.9 41.2 44.0 43.0 Refinancingandnet loans 0.1 0.2 1.4 0.2 Otherexpenditures 5.3 8.4 3.5 5.3 6.8 4.6 10.0 6.1 9.2 9.6 Inpercentageof GDP Total expenditures 24.2 24.6 23.0 23.9 25.0 30.2 34.9 37.6 47.2 37.7 Total off-budget expenditures 4.9 6.8 5.4 5.5 7.3 9.0 13.8 11.5 20.6 12.7 Recurrent expenditures 15.9 16.1 16.9 16.3 16.8 20.5 20.9 22.4 31.1 24.4 Wees andsalaries 4.6 4.4 4.9 4.4 4.2 4.6 4.5 5.1 4.7 4.3 Expenditureon goods andservices 4.6 4.6 4.9 5.4 5.4 6.9 7.0 8.9 10.4 9.3 Subsidies 1.5 1.5 1.6 1.6 1.7 2.3 2.1 2.5 7.8 2.4 Military expenditures 2.3 2.6 3.0 2.1 2.7 3.5 4.6 3.2 4.9 4.7 Govemment debt interest 2.1 2.2 2.1 2.5 2.6 3.O 2.7 2.5 2.7 3.0 External 1.7 1.8 1.8 1.9 1.9 2.2 1.9 1.9 2.0 1.8 Internal 0.4 0.4 0.3 0.6 0.6 0.8 0.8 0.6 0.8 1.2 Special accounts 0.9 0.7 0.4 0.3 0.2 0.2 0.0 0.2 0.5 0.6 Investment expendituresand nn loons 5.5 4.3 4.5 5.3 5.5 7.9 10.5 13.2 13.4 10.9 Investmentexpenditures 5.5 4.3 4.5 5.3 5.5 7.9 10.4 13.2 13.0 10.9 Refinancingandnet loans 0.0 0.1 0.4 0.0 Other expenditures 2.9 4.2 1.6 2.3 2.7 1.8 3.5 2.0 2.7 2.4 In percentageof totalexpenditures Total off-budget expenditures 20.2 27.8 23.4 23.1 29.1 29.7 39.5 30.6 43.6 33.8 Recurrent expenditures 65.6 65.3 73.5 68.1 67.2 67.9 60.0 59.5 65.8 64.7 Wages andsalaries 18.9 18.0 21.2 18.2 16.7 15.1 12.8 13.7 10.1 11.5 Expenditureon goods and services 18.9 18.9 21.6 22.8 21.7 22.9 20.2 23.7 22.1 24.7 Subsidies 6.0 6.1 6.9 6.6 7.0 7.7 5.9 6.7 16.5 6.4 Military expendimes 9.3 10.7 12.9 8.8 10.7 11.6 13.3 8.4 10.3 12.5 Government debt interest 8.7 8.9 9.2 10.4 10.2 10.0 7.7 6.5 5.8 8.0 External 7.2 7.5 7.9 8.0 7.6 7.3 5.5 5.1 4.2 4.9 Internal 1.5 1.4 1.4 2.5 2.6 2.7 2.2 1.5 1.7 3.1 Special accounts 3.8 2.7 1.7 1.3 0.9 0.6 0.1 0.5 1.0 1.6 Investment expendituresand net loans 22.5 17.4 19.4 22.3 21.9 26.2 30.0 35.3 28.4 28.9 Investmentexpenditures 22.5 17.4 19.4 22.3 21.9 26.2 29.9 35.1 27.6 28.8 Refinancingandnet loans 0.1 0.1 0.8 0.1 Other expenditures 11.9 17.2 7.1 9.6 10.9 5.9 10.0 5.2 5.8 6.4 Source :MF 14 Subsidiesand transfers 2.15 Subsidies comprise transfers to administrative public entities (PES), industrial and commercial public enterprises (EPICS)such as the National Social Security Fund (CNSS), local government bodies, scholarships to study abroad, and international contributions. Contrary to countries such as South Africa, where transfers account for more than half o f total expenditures (56 percent in 2003), Mauritania's 2003 subsidies and transfers accounted for slightly more than 17 percent in 2003, compared to 12 percent in Madagascar in the same year. One third o f subsidies and transfers (34 percent on average in the period 1995-2004) was executed off-budget, through LDAs and foreign exchange losses. Box 2.4: Subsidiesto PES The impact o f remaining PESon public finances can not be determined with certainty. Infact, the very definition o f a subsidy to a PE i s unclear. The MF does not consider that ex ante funds extended to an EPA-for instance, to help it to fulfill obligations under a performance contract, settle an electricity invoice, or (in the case o f Air Mauritania) operate unprofitable flights-constitute a subsidy. T o be a subsidy, a payment should cover PE losses a posteriori. Accordingly, the authorities argue that there have been no PE subsidies since the structural adjustment program was launched inthe mid-l980s, save to Mauripost, which-since post and telecom activities were split-was not financially viable (OED 2004). 2.16 Total subsidies to the 7 largest PESfluctuated slightly from year to year and increased from UM 385.7 million in 1990 to UM449.5 million in 2001, reaching a peak (UM 1,757.7 million) in 1998 as a result o f the recapitalization of Air Mauritania in light o f its deteriorating financial situation. Subsidies to SONELEC31and, recently, to Mauripost should also be noted. S N I M receivedno state subsidies. Table 2.3: Operating Subsidiesto the Main PublicEnterprises(PES), 1998-2002 1998 1999 2000 2001 2002 Air Mauritania 1,639.0 201.5 201.5 NIA NIA SNIM 0.0 0.0 0.0 0.0 0 SONELEC 108.6 113.3 149.2 ... ... SOMELEC ... ... ... 0.0 NIA SNDE NIA NIA MAUFUPOST ... , 465.8 449.5 NIA MAUFUTEL ... 1 . . NIA NIA NIA Total 1,747.6 314.8 860.3 449.5 NIA Source :OED (2004) Interest onpublic debt 2.17 As a percentage o f GDP, interest payments on public debt increased gradually between 1995 and 2004. As a percentage o f total expenditures, they increased in the period 1995-1999 and decreased between 2002 and 2004. They constituted the third largest category o f recurrent expenditures. Their increase was due in part to interest paid on bank overdrafts and government bonds. Internal debt repayment took place outside the budget, particularly through LDAs. Generally spealung, LdFprovisions for internal debt servicing were based on underestimations and failed to cover bank overdraft charges. In 31InJuly 2001, SONELEC was split into two enterprises: the National Electricity Company (SOMELEC) and the National Water Company (SNDE). 15 the period 1995-2004, on an annual basis, internal debt interest accounted for approximately 25 percent o f total annual debt servicing (11percent on overdrafts and 14percent on government bonds). Investment expenditures 2.18 As a percentage o f GDP and o f total expenditures, investment3* spending increased considerably in the period 2000-2004 compared to the period 1995-1999. Inthe period 1995-2004, it accounted for approximately 22.6 percent o f total expenditures per year. The annual rate o f public investment increased steadily between 1995 and 2003 but followed a downward trend starting in200411percent. As inthe case o f recurrent spending, a part o f investment spending was also executed outside the budget. The next chapter contains a more detailed discussion o f investment expenditures, including linkages to the PRSP. A more thorough analysis would require the identification o f the percentage o f investment expenditures that i s actually used to meet operational needs and maintenance costs necessitated by the rapid deterioration o f infrastructure as a result o f adverse weather conditions (high temperature, drought, sandstorms, and erosion). Although maintenance expenditures are economically important, their impact i s politically less visible than the effect ofnew investments. c. FUNCTIONAL CLASSIFICATION OFEXPENDITURES 2.19 Upto the end o f 2005, the DBC didnot use the functional classificationinpreparingthe budget or assessing the impact o f expenditures. The lack o f proper classification handicaps the formulation o f consistent economic policies and their assessment to inform future actions. Its absence also makes it difficult to compare Mauritania's budget statistics with those o f other countries with similar income levels. For purposes o f preparing the 2004 PER, the DBC has drawn up a bridge table in order to reclassify expenditures into the functional classification. This same table was used to reclassify expenditures for the years 2000-2004. However, the classification does not include certain lunds o f off- budget expenditures. 2.20 . Expenditures as a whole can be classified under four main headings33: Social services, comprising public services provided directly to communities and households, .. such as education, health, and social protection; Generaladministration services, comprising public administration, law enforcement, andsecurity; Economic services, comprising regulation of, or support for, the transport, energy, apculture, and . industrysectorswith a view to ensuringtheir efficient operation; and Other uses, includingdebt servicing andtransfers to various government bodies. 2.2 1 Functional classification o f government expenditures in the period 2000-2004 revealed that allocations to the social sectors increased considerably, and faster than other expenditures. Table 2.4 shows changes inexpenditures, broken down by function. 32The capital expenditures inpercentage o f total expenditures have grown from 21 over 1995-1999 to 29 over 2000- 2004. 33Pradhan(1996). 16 Table 2.4: BreakdownofExpendituresAccording to the FunctionalClassification 2000 2001 2002 2003 2004 2000 2001 2002 2003 2004 Enmilliardsd'UM En%du PIB General public adminishation services 5.8 6.6 5.0 11.2 13.0 2.2 2.3 1.6 3.3 3.3 Defense 5.4 7.5 7.4 16.5 14.3 2.1 2.6 2.4 4.9 3.6 Law enforcement and security 4.0 5.1 6.5 11.0 9.8 1.5 1.8 2.1 3.3 2.5 Education 6.3 8.8 9.9 11.2 12.2 2.4 3.1 3.2 3.3 3.1 Health 1.9 2.8 5.0 5.8 6.6 0.7 1.0 1.6 1.7 1.7 Social security and related activities 2.0 0.6 2.0 10.7 6.6 0.8 0.2 0.6 3.2 1.7 Housing andcollective development 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Recreation, culture, andreligious activities 0.2 1.0 1.1 0.4 1.8 0.1 0.3 0.4 0.1 0.5 Fuel and energy 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Agriculture, forestry, and fishing 3.1 4.0 4.3 8.5 7.7 1.2 1.4 1.4 2.5 1.9 Mining and manufacturing 0.1 0.2 0.1 0.2 0.4 0.0 0.1 0.0 0.0 0.1 Transport and communication 3.3 4.2 3.7 5.8 8.4 1.3 1.5 1.2 1.7 2.1 Other economic activities 0.1 0.1 0.1 0.2 0.4 0.0 0.0 0.0 0.1 0.1 Non-allocated expenditures 47.2 59.3 72.3 78.1 68.4 18.3 20.7 23.2 23.1 17.2 Total 79.3 100.4 117.6 159.5 149.6 30.7 35.0 37.6 47.2 37.7 Source: MF Socialservices 2.22 Inrecent years, the social sectors, notably education, health, and social affairs, were the main recipients o f Mauritania's public expenditures. The sectors o f education and health were also the main recipients o f HIPC resources. In the period 2000-2004, health and social security expenditures almost tripled, reaching 3.4 percent o f GDP at the end o f the period. However, these expenditures had a mixed impact on health indicators, which on the whole tended to stagnate. In fact, improvements in the state o f health inMauritania were so slight that, according to several assessments, attainment o f the health-related MDGs is hardly conceivable. Although the number o f health units increased as a result o f the expenditures in question, the number o f patients' visits remains weak. Infant, juvenile, and maternal mortality rates were much higher compared to other developing countries and to the average for Sub- Saharan Africa. 2.23 Expenditures on education doubled inthe same period, increasing from 2.5 percent o f GDP to 2.3 percent. These expenditures have had a positive impact, especially on access to primary education. According to data from the Ministry o f Basic and Secondary Education (MEFS), the GER in basic education reached 95 percent at the end o f 2005. On the other hand, indicators o f education quality deteriorated interms o frepetition andretentionrates and level o f knowledge acquisition. General government services 2.24 Expenditures on defense and law enforcement have increased substantially. As a percentage of GDP, they rose from 3.6 percent in 2000 to 8.2 percent in 2003, before being limited to 6.1 percent in 2004. However, as a percentage o f total expenditures, they decreased in comparison to the early 1990s, when they had accounted for 24.5 percent.34 2.25 General administration expenses increased from 2.2 percent o f GDP in 2000 to 3.3 percent in 2004; as a percentage o f total expenditures, they remained stable on the whole. As indicated in the discussion o f expenditures based on the economic classification, this expenditure category had little impact interms o f modernizingpublic administrationand enhancing the quality o fpublic services. 34 According to the sectoral classification used inthe 1994PER (World Bank). 17 Economic services 2.26 Although, in terms o f potential growth and poverty reduction, agriculture and livestock are crucial to Mauritania's economic and social development policies, allocations to those sectors increased only modestly and, inthe period 2000-2004, amounted to 2 percent of GDP35on average. With regard to agriculture, diversification targets were not attained and as result, the sector continued to be vulnerable to exogenous shocks, while poverty incidence was highinrural areas. 2.27 As a percentage of GDP, expenditures on transport and communications during the period increased fiom an average o f 1.6 percent to 2.1 percent at the end o f the period. However, they did not lead to any significant expansion o f the road network. Limited infrastructure development is a major obstacle to economic growth. For instance, fulfilling the considerable potential o f irrigated agriculture along the Senegal River in the southern part o f the country requires substantial investment, largely in infrastructure. Moreover, the vastness o f the country i s a challenge to the development o f an extensive road network, including much-needed rural feeder roads. With regard to telecommunications, access to services improved, particularly through enhancement o f G S M coverage, but the rates charged were higher inMauritaniathan inthe MaghrebandCFAFZone countries. 2.28 Expenditures on other economic activities showed no notable change during the period. Finally, spending on interest payments represented, on average, 20 percent o f GDP between2000 and 2004. D. CONCLUSIONS AND RECOMMENDATIONS 2.29 Analysis based on the economic and functional classifications has shown that the nominal value o f government expenditures increased steadily and significantly but the outcomes achieved were mixed. Execution o f expenditures outside the regulatory framework did not help to achieve the government objective o f reducing unproductive spending, such as military outlays. Unproductive military- and security-related expenditures tended to increase, to the detriment o f spending in favor o f productive sectors, such as agriculture and transport, which received limited resources even though, given their positive externalities, they are crucial to sustainable development and poverty reduction. Further increases inunproductive expenditures will eventually pose a problem interms o f development and sustainability . Despite rising unproductive expenses, allocations to the social sectors continued to be quite high. Furthermore, expenditures on goods and services failed to improve the efficiency o f public administration. As a result o f unsatisfactory salaries and operating conditions, the civil service remains unattractive. 2.30 Execution o f expenditures outside the budget leads to a lack o f information on government spending as a whole. Accordingly, the authorities decided to put an end to that practice by prohibiting Treasury debits without prior budget allocations and the necessary supporting documents. These measures should be strengthened in view o f the increased revenue expected from oil reserve exploitation. In a number of developing countries with substantial revenue from natural resources, part o f the government revenue and expenditures i s processed off budget and consequently, those resources fail to contribute to the development o f other sectors and to poverty reduction. Redoubling efforts in this area is a matter o f political resolve. 35This spendingcategory includes expendituresonthe development of fishing and forestry. 18 2.31 An analysis limited to total expenditures by sector and not comprising a detailed intra-sectoral review o f appropriations would not allow for an accurate assessment o f the impact o f budget allocations or the identification of obstacles to attaining the expected results. It is therefore recommended to undertake PERsin some priority sectors, such as education and health, and some productive sectors, such as rural development and transport. Inview o f the expected increase in public expenditures as a result o f oil revenue, such sectoral analyses would help to devise ways to better coordinate sectoral policies and budget allocations; buildthe absorptive capacity o f the sectors concerned; provide useful information for the formulation o f sectoral MTEFs in line with the global MTEF; and identify the financing and partnership mechanism^^^ best suited to sustainable development inthese sectors. 2.32 Insummary itisrecommended : Toreigninmilitary expenditure for the benefit o fpriority andproductive sectors To include all staff expenditures under the salary line item To undertake PERsinsome priority sectors, such as education andhealth, and some productive sectors, such as rural development and transport. Regardinggovernment expenditures on goods and services, which i s the largest recurrent spending category butrather ineffective interms of improving the quality o f public service3', some measures have been taken in2006 to rationalize the management o f the relevant appropriations. All recurrent costs incurredby the various departments shouldbe broken down by allocation with a view to savings. To improve the remuneration o f government employees, and encourage their conscientiousness through the implementation and dissemination o f the professional code o f ethics and the formulation o f performance-based career development policies in order to reinforce the motivation and commitment o f civil servants which i s crucial to the quality o fpublic administration. 36Public-private partnership, participation of beneficiaries, etc. 37 Inthat connection, some technical and financial support is provided under the PRECASP for the implementation o f public adrmnistration modernization measures recommended in the report drawn up by the Inter-Ministerial Committee for Good Governance. 19 3. DISCONNECTBETWEENPUBLICINVESTMENT PROGRAM AND THE PRSPPRIORITIES 3.1 This chapter analyzes investment expenditures inrelationto the priority action plan (PAP), which addresses public investment needs in the PRSP priority sectors. The PRSP should be the framework for the planning and execution o f government expenditures. In a macroeconomic environment characterized by government overspending, it was essential to determine whether public investment policy had been consistent with priority sector needs and executioq capacity in order to formulate recommendations for the second PAP and its budgetary implications. 3.2 The 2004 PER38showed that, in the period 1998-2002, the Government considerably increased investment allocations to the social sectors and to targeted poverty-reduction programs. In the same period, financing o f the Consolidated Investment Budget (BCI) with national resources increased, attaining approximately 40.7 percent o f total BCI financing in 2002. The overall rate o f BCI execution improved, despite weak absorption rates in the social sectors. This chapter analyzes the relation o f the investment budget to PRSP priority actions aimed at poverty reduction. The PAP comprised only investment expenditures estimated at the sector level. 3.3 Investment expenditures increased considerably inthe two first years o f the PRSP and decreased somewhat in 2003. They partly covered the PAP, which comprised five priority areas3' with a strong poverty- and inequality-reduction potential. Targeted and integrated poverty reductionprograms had been identified with a view to addressing the needs o f disadvantaged groups o f the population in the poorest and most vulnerable areas o f the country. For the initial period, 2001-2004, the PAP was built around the following four pillars o f the PRSP: . Pillar 1: "Stimulating broad-based growth," comprising activities necessary for improving the private sector environment, promoting competition, reducing factor costs, developing the . productive sectors, etc. Pillar 2: "Anchoring growth in the economic environment o f the poor," comprising activities aimed at improving the social and economic environment o f the vulnerable population groups in . the rural and poor urban areas. Pillar 3: "Expanding basic services," comprising activities related to education and health, and aimed at improving the dnnlungwater and power supply and promoting the new information and communicationtechnologies (ICT). Pillar 4: "Promoting good governance, and capacity building,'' focused on modernizing public administration, promoting economic and environmental governance, and strengthening the justice sector and civil society. 38This document is an update of the 2004 PER, which was not published. 39Education, health, water, rural development, and urban development. 20 Box3.1: PublicInvestmentProgram(PIP) and ConsolidatedInvestmentBudget(BCI) The PIP translates the priorities defined inthe Government's development policy into investment programs and projects, excluding only projects and programs implemented by some PESand fianced through arrangements negotiated directly by them with external partners. PIP projects and programs are at various stages o f development (ranging from early to advanced project design phases) to be implemented in the medium term by the public sector (broadly defined). The BCI provides an annual framework for budget planning and for follow-up o f all PIP investments, whether funded under the budget from the State's own resources or financed with external funds (not included inthe budget and not executed through government expenditures). The Public Investment Monitoring System (SYPSIM) is used for the completion o f PIP/BCI at the Ministry o f Economic Affairs and Development (MAED). Despite the improvement made inthe past few years, this planning tool presents various deficiencies such as: o Lack o f long-term data o f the BCI o The absence o f connection between the DBC (Finance Ministry) and the financing department at the MAED o Lack o f a monitoring and evaluation module for investment expenditures; lack o f reporting on non-used credit; and o The centralizedstructure o fthe MAED A. COSTINGOFTHEPRSPPRIORITYACTIONSPLAN(PAP) 3.4 The first costing o f the PAP, presented in the PRSP report, was followed by updated costing in the annual PRSP follow-up reports drawn up in connection with the revision o f strategic objectives. Financial planning for the implementation o f PRSP priority actions and objectives was not based on unit costs or on the absorptive capacity o f the given sectors and, as a result, there were considerable differences between successive annual cost estimates contained inthe various PRSP documents. 3.5 Early consolidated estimates concerning PAP activities formulated in the 2001 PRSP report for the period 2002-2004 were considerably lower than those o f subsequent years. Estimated costs peaked in the 2003 PRSP progress report. The 2002 program cost, underestimated in the PRSP approved in 2001, was subsequently revised upwards (from UM 18.6 billion to UM 34.6 billion). Moreover, in the period 2002-2004, the annual PRSP program cost practically doubled and PRSP expenditure projections increasedfrom UM32 billion to UM62 billion as a result o f higher expenditures on health and education. The significant increase o f PRSP program costs in2002,2003, and 2004 was mainly due to an increase in the expenditures executed (in2003) and to substantial 2004 appropriations for pillar 1, one thirdo f which were earmarked for the component entitled "Growth-supporting infrastructure." In all planning documents, cost estimates were higher for pillars 2 and 3 than for pillars 1and 4. 3.6 Differences between PAP cost estimates from year to year were mainly due to the administration's inadequate planning capacities (failure to take into account unit costs, activity objectives, and the sector's execution capacity). Inorder to compare PAP costs and BCI figures for the same period, investment data were reprocessed by summing up sectoral expenditures under the pillar concerned. Revised costs in the 2004 monitoring report, which was based on relatively realistic information and on sector needs, were quite close to the corresponding BCIprojections. The PAP cost estimates provided in that report have been usedinthe remaining discussion. 21 3.7 Poverty-reduction investments in the PAP and BCI were not broken down by region. Regional poverty reduction programs (RPRP), which express PRSP objectives in a regional framework, began to be developed in 2002. Three RPRPs have so far been drawn up for Mauritania, and concern the poorest regions. For these regional strategies to be functional guides for public action, the PAP and the BCI should be presentedby sectoral objectives andby regions. B. CONSOLIDATEDINVESTMENTBUDGET(BCI) AND PRIORITY ACTIONS PLAN (PAP) 3.8 A breakdown o f planned 2001-2004 investment expenditures by PRSP pillar shows that approximately 60 percent o f the BCI was allocated to pillars 2 and 3. Although this reflected some consistency between budget planning and poverty reduction objectives, no systematic method was used by the authorities to ensure such correlation. Since the costs o f attaining the various objectives was not estimated ina way that would permit the calculation o f unit costs and the preparation o f sectoral MTEFs to be taken into account in drawing up the LdF and the BCI, it has been difficult to establish any correspondence between budget allocations and actual financing requirements as described under the pillars and the sectors concerned. Pillar 1: "Stimulating broad-based growth" 3.9 Inthe period 2001-2004, an annual average of23 percent oftotal BCI, or 3 percent of GDP, was allocated to pillar 1, the only pillar for which the BCI and PAP amounts were almost identical. There were, however, differences regarding the level o f allocations for individual pillar components. In fact, investment allocations were much higher for growth-promoting infrastructure than for the productive sectors. 0 Inthe period of implementation of the PRSP, allocations for growth-promoting infrastructure, comprising transport, energy, and ITC,40 increased. The transport subcomponent was the main beneficiary. Energy and the ICT were provided for inthe BCIbut not inthe PAP and received, on average, annual allocations amounting to 17 percent o f the BCI and 2 percent o f GDP. Allocations to the transport sector doubled between 2001 and 2004 and were earmarked mainly for road construction and airport rehabilitation. Allocations to the productive sectors-fishing, mining, manufacturing, and services-were not significant, as the authorities gave priority to investments inthe social sectors. 0 Investment resources earmarked for private sector promotion were insignificant, since the measures to be financed-strengthening the reform o f the legal and judiciary framework for the development o f business-require political resolve to implement existing provisions rather than new laws or regulations. 40Inthispillar, energy andITCare examined interms of access to basic services. 22 Pillar 2: "Anchoring growth in the economicenvironment of thepoor" Figure3.1: BreakdownofBCIAllocationsto Pillar2 3.10 In the period 2001-2004, an annual average of 42 percent of total BCI, or 5 percent 7x of GDP, was allocated to pillar 2. Although no method linlung PRSP priorities and budgetary planning was used, the B C I allocations in question amply covered-and in fact exceeded 19% by more than UM 21 billion -the costs of activities under that pillar, which were aimed at Ca Rural development improving the social and economic environment S Environment of the vulnerable population groups in the rural 0 Habitat andwban development 0 Multi-sector projectsfor poverty reduction . and peri-urban areas. #lEmployment About 14 percent of the annual BCI were earmarked to the "rural development" component. About 88 percent were allocated to agriculture, rural infrastructure and water supply. Few investment resources were allocated to the "livestock development" subcomponent, but improvement of the rural water supply was beneficial to that area as well. However, insufficient resources were allocated to the "dissemination, research, and training" subcomponent, even . though it i s essentialto rural development. Allocations earmarked for urban development and housing accounted on average for 8 percent of the total annual BCI. These investments were aimed at improving housing conditions in the . poorer areas of the major towns, particularly by correcting living conditions in the newer neighborhoods through infrastructure and sectoral institutionbuilding. Multi-sector projects for poverty reduction (on average, 15 percent of the BCIper year) targeted large disadvantaged population groups inrural districts and in urban areas (shantytowns). These projects were fundedmainly from HIPC resources and were implementedby the Commission for HumanRights, PovertyReduction and Integration(CDHLPCI). Pillar 3: Ymprovingaccess to basic services" Figure3.2: BreakdownofBCIAllocationsto Pillar 3 3.11 In the period 2001-2004, an annual average o f 28 percent of total BCI, or 3 percent of Water and Culture GDP, was allocated to pillar 3. The allocations in Sanitation 1% question fell short of the cost identified in the PAP. In addition, the allocations varied significantly between sub-sectors and between years, and, as shown in the diagram 3.2. They were earmarked for the sectors of education (43 percent), health (35 percent), and water (21). 35% e As an annual average, allocations to the education sector amounted to 12 percent 23 o f total BCI. As a percentage o f GDP, they fluctuated between 1percent (2001 and 2005) and 2 percent (2.002-2004). The decrease was mainly due to the insufficient execution capacity which characterized the social sectors, particularly inrelationto external financing. As an annual average, allocations to the health sector amounted to 10percent of total B C I and 1 percent o f GDP. Assisted by a number o f development partners in recent years, the sector i s currently supported only by a few donors. Attainment o f the health-related MDGs i s highly unlikely. A s an annual average, allocations to the water and sanitation sector amounted to 6 percent o f total BCI and included funds for enhancing access to drinlungwater inrural and urban areas through digging wells, building urban networks, and installing hydrants. It has not been possible to analyze funding for access to electricity and ICT. Some o f the allocations financed the establishment o f the agency for universal access to basic services. Pillar 4: "Promotinggood governanceand capacity building" 3.12 BCI allocations to pillar 4, although ten times higher than the corresponding PAP provisions, were low compared to allocations to the other three pillars, probably because the attainment o f good governance objectives depends mainly on the resolve o f the political authorities to implement existing legal provisions and apply sanctions if needed. Limited resources were earmarked for institutional and public administration support. B C I and PAP amounts differed mainly because the PAP contained only activities for building civil society capacities and improving macro-budgetary planning and monitoring, while the B C Iincluded some allocations to the sector o fjustice. c. BCIEXECUTION AND THE ACHIEVEMENT OF PRSPOBJECTIVES 3.13 An assessmento fprogress towards the PRSP objectives reveals that PAPimplementationhas not been completely satisfactory. Only 61.5 percent o f the activities planned were actually executed, approximately 22 percent were in progress and 17 percent were not executed. Additional activities were carried out even though they had not been planned and did not necessarily contribute directly to the attainment o f PRSP objectives. This limited implementation o f strategic priorities slowed down the attainment o fPRSP objectives. Figure 3.3: PRSPPriorityActivities and BCIExecution, 2002-2004 140000 7 Axe 1 Growth Axc 2 Pro- Axe 3 Basic Axc 4 Total poor growth S C ~ V ~ C C S Gouvcmance Stratemc axes 24 3.14 The rate o f BCI execution varied between sectors and, in some cases, between periods. As indicated inthe 2004 PER, although investment allocations to the social sectors in recent years increased significantly, their rate o f execution remained low compared to the overall rate of execution o f the General Investment Budget (BGI), which clearly improved. The main causes o f the persisting low absorption capacity o fthe social sectors were the following: Planned investments were often overestimated because they were not based on actual data and needs; The quality o f any feasibility studies carriedout was not satisfactory; Infi-astructure work was often slowed down because o f inadequate local capabilities, particularly limited capacity o f the construction sector; There were too many procedures with regard to external financing by various donors; and The frequent fluctuations inthe price o fbuildingmaterials inthe local market, and the difficulties o f access to certain areas, especially during the rainy season, which often leads to a halt in construction work for several months. Table 3.1: BCIExecutionby PRSPPillar,2001-2005 2001 2002 2003 2004 2001 2002 2003 2004 PlannedBCIinbillion ofUM Execution(inpercentageofplannedamount Axe 1 :Stimulatingbroad-basedgrowth 7,011 9,222 12,081 13,019 Growth sectors 2,823 3,087 2,549 2,821 101.9 150.5 111.3 94.6 Infrastructures 4,154 6,106 9,480 10,147 75.9 110.4 136.4 128.9 Privatesector development 34 30 53 52 83.4 66.7 34.3 25.3 Axe 2 :Anchoringgrowth in theeconomicenvironment 11,048 15,380 14,220 17,618 Ruraldevelopment 4,909 4,969 5,754 7,155 98.8 101.1 97.4 113.6 Environment 323 1,179 791 114 45.9 113.6 122.9 17.7 Habitatandurbandevelopment 1,649 2,818 2,483 3,246 89.6 110.0 69.4 74.7 Multisectoralproject for povertyreduction 4,162 5,010 5,044 6,099 84.8 92.1 79.3 89.0 Employment 5 1,404 149 1,004 0.8 98.3 9.1 100.0 Axe 3 :Improving access to basic services 7,327 7,511 5,348 7,675 Education 2,154 2,261 1,532 2,216 111.7 38.6 23.9 33.7 Health andsocial affairs 1,589 2,372 2,586 3,072 65.4 53.8 52.8 61.4 Urban water 3,512 2,649 974 2,156 208.9 171.6 42.4 49.3 Culture 56 203 148 184 20.9 95.2 68.9 1,229.8 Youth andsports 16 26 108 47 33.0 100.0 473.6 41.6 Axe 4 :Promotinggood governanceand capacig buildin 2,175 2,371 3,634 3,877 Judicial system 28 187 159 182 161.2 100.0 79.6 82.9 Institutionnaldevelopment 1,091 1,107 1,764 1,873 78.6 69.2 80.5 62.7 Capaicitybuildinginthe administration 1,057 1,077 1,711 1,822 78.5 69.3 84.0 65.5 Total (hors SNIhl)** 27,562 34,484 35,283 42,190 91.1 90.7 76.9 81.8 SNIM 26 457 529 162 0.5 18.3 20.3 6.0 Total 27,588 34,941 35,812 42,351 76.8 86.1 73.7 77.9 Financing sources Nationalresources 11,846 16,494 17,068 22,525 99.5 103.1 79.4 89.4 Externalfinancing 14,651 17,349 16,987 17,965 127 147 131 130 Externalfinancing(credits) - . 10,346 12,487 12,829 14,175 83.2 83.4 87.2 97.7 Grants 4,304 4,862 4,158 3,790 44.1 63.3 43.7 32.2 Source:Mauritanianauthorities(MAEDIMF) * Rectificativebudgetwas takenintoaccount ** SNIM : Nationalindustrial andminingcompagny 25 Pillar 1: "Stimulating broad-based growth" 3.15 A highBCI execution rate was achieved for pillar 1. The execution exceeded PAP considerably due to weaknesses in multiyear planning, which failed to take into account the sectors' real needs and execution capacity. Allocations to the transport sector (with an average execution rate o f 112 percent) were limited. That sector's total budget amounted to less than 3 percent o f GDP during the past decade, out o f line with the respective objectives and strategies. Investment carried out during the PRSP period contributedto enlarging the road network4'. Transport sector projects consisted inbuildingapproximately 1,500 km o f inter-town roads, rehabilitating parts o f the "road o f hope" (Route de l'espoir), and improving the roads in Nouakchott; but failed to remedy inadequacies in that type o f infrastructure: approximately 60 percent o f the Mauritanian road network i s in a poor state and impassable during the rainy season, a situation impeding agricultural development and rural poverty reduction. Private enterprises and funding issues were responsible for considerable delays in road construction work. Funding for road maintenance and rehabilitation was limited, in the face o f rapid infrastructure deterioration as a result o f adverse climatic conditions. With regard to growth-supporting infrastructure, B C I allocations to the productive sectors such as fisheries and mining were small and satisfactorily executed. Those sectors performed favorably with small resource allocation. Pillar 2: "Anchoringgrowth in the economicenvironment of thepoor" 3.16 Pillar 2 absorbed approximately 40 percent o f total BCI allocations-which were considerably higher than PAP provisions-and featured an annual rate o f execution close to 100 percent (with some variation betweenpillar components and between years). 3.17 The "rural development" component was one of the areas with a 100 percent execution rate. Although significant investments led to incipient agricultural diversification and the creation o f agricultural infrastructure, Mauritanian agriculture was left vulnerable to weather conditions and such scourges as locusts. As a result, the agricultural GDP growth rate developed irregularly in the period considered. In 2000-2004, poverty incidence in the rural areas declined by 5.8 percent, compared to the 6.3 percent target (i.e., an annual reduction o f 1.5 percentage points). In terms o f arable land, the country's potential (500,000 ha, including 137,000 ha o f irrigable riverside areas) i s largely underutilized, since only 42,000 ha are developed, o f which only 20,000 ha are exploited in a single year. Inaddition to inadequate investment allocations, other factors that are exogenous to the sector inhibit its development and reduce its profitability. These include regional isolation, accentuatedduring the rainy season, and lack o f infrastructure (especially power). The agricultural sector expenditure review, which i s in progress, is expected to help to identify the main strategy, funding, and institutional management problems in that area. 3.18 The overall execution rate on investment allocations to the housing and urban development component was 90.5 percent but varied considerably from year to year. In2003 and 2004, the execution rate for the urban development sub-component was below 75 percent. That subcomponent's main ' objective was to increase income and improve living conditions inthe poorer neighborhoods in the cities and towns. According to the 2004 EPCV, conditions in those neighborhoods improved but considerable effort was still necessary in order to attain the PRSP objectives: 23.3 percent o f households lived in precarious housing; with regardto access to property, 43.2 percent o f the urbanpopulationwas tenants. 3.19 The execution rate on multi-sector projects for poverty reduction averaged 85 percent per year, reflecting the fact that they were executed directly by the CDHLCPIthrough management procedures that 41Despite this performance-which mainly consisted in the construction of the Nouakchott-Nouadhibou road and the rehabilitation of some airports-many areas inthe country remain isolated. 26 were more flexible than the State budget process. These projects were meant to be integrated activities targeting particularly poor rural and urban areas, where sector departments could not intervene. Budget execution by the CDHLCPIexhibited the following weaknesses: 0 A lack o f synergy and complementarities between CDHLCPI activities and the actions undertaken by line ministries. This i s a result o f inadequate coordination and non-inclusion o f CDHLCPIactivities inthe sectoral strategies. 0 A lack o f a global vision that would serve as a framework for CDHLCPI activities. The result is dispersion and an uncoordinated response to often unjustifiedrequests. 0 CDHLCPI spending in excess o f annual allocations through recourse to bank overdrafts. Eventually, new annual allocations were usedto reimburse liabilities incurred inearlier years. 3.20 In addition to their impact on economic growth, rural investment projects and targeted activities contributed to poverty reduction in the rural areas. In fact, according to the EPCV, the incidence o f poverty between 2000 and 2004 decreased inthe rural areas by 1.8 percent per year (namely, from 66.2 to 59 percent), faster than inthe country as a whole (less than 1.1percent per year). This poverty reduction, however, occurred mainly (to the extent o f 11 percentage points), in the Senegal River valley and was relatively modest inother rural areas. Pillar 3: "Improving access to basic services" 3.21 As indicated in the 2004 PER, the social sectors included in pillar 3 showed the lowest BCI executionrates. Although the underlying causes probably apply to all sectors, the rate o f execution varied between sectors and between periods. Particularly large disparities in the execution rates can be found in the priority areas of pillar 3: the execution rate was higher in the health sector than in the water and education sectors. 3.22 In the period 2000-2004, the BCIexecutionrate inthe education sector was low, except for 2001, when it exceeded 100 percent. In2002-2004, it was below 40 percent, with a 23 percent minimum in2003. That weak performance didnot affect progress towards the PRSP objective ofuniversal access to basic education, a target whose attainment by 2015 seems probable. Inthe period 1991- 2004, the gross enrolment ration (GER) increased from 48.7 to 98.3 percent.42However, the low execution rate i s probably correlated with teaching quality, whose low level gives cause for concern. In fact, in 1999-2005 the rate o f retention was low (48.8 percent43)and the knowledge acquisition level inadequate because, inter alia, some establishments offered an incomplete school curriculum and the implementation o f the 1999 education reformran into difficulties. In other words, the increased financial effort did not fully translate into better performance. Education expenditure equal to 1percent o f GDP translated into 1.6 years o f additional schooling, compared to 1.8 years in IDA member countries in Africa. That outcome was due in part to limited capacities for managing and supervising the education system as a whole. 42According to the EPCV carried out in2004-05, the GER was 76.3 percent in 2003-2004 and 81.4 percent in2004- 2005. 43Estimationbased on school administration figures. 27 Figure3.4: GrossEnrolmentRatio(GER)inBasicEducation 1991-2004 100% 80% 20% -"I" 0% ~1990- 1992- 1995- 1996- 1997- 1998- 1999- 2000- 2001- 2002- 2003- 1991 1993 1996 1997 1998 1999 2000 2001 2002 2003 2004 Source: Report on the National Education System (RESEN), 2006 3.23 In the period 2000-2004, the BCI execution rate in the health sector was approximately 56 percent and showed relative stability from year to year. The investment funds financed the construction and outfitting o f health units under the Health Sector Development Program (PDS). In view o f past trends, attainment o f Mauritania's health-related MDGs i s highly unlikely. In the above period, the malnutrition indicator (weight for age) declined from 32 to 30.2 percent but fell short o f the objective (20 percent by 2004). The rate o f assisted childbirths was 57 percent in 2004 (PRSP objective: 70 percent). Vaccination coverage increased from 32 percent in2000 to 79 percent in2004. 3.24 In the first two years of PRSP implementation, the BCI rate of execution in the water and drainage sector was particularly highbecause o f limited funding in the face o f immense needs. On the other hand, that execution rate fell to only 43.3 percent in2003 and 2004. It shouldbe stressed that access to drinking water i s still a problem in this desert country, where only 53.5 percent o f the population has such access, compared to 50.4 percent in 2000. Use o f domestic taps increased (from 15 percent in 2000 to 18.5 percent in 2004) to the detriment o f other water supply sources. The improvement, however, concealed the following pronounced disparities: (i) The rate o f domestic tap use was 13.9 percent inrural areas compared to 29.5 percent in urban areas; (ii) it was 7.1 percent among very poor households compared to 21.4 percent o f among households above the poverty line; and (iii) the price o f water charged by street vendors, the main source (44.4 percent) o f supply inpoor neighborhoods, was 11times the price invoiced by the National Water Company (SNDE). Pillar 4: "Promotinggood governanceand capacity building" 3.25 Allocations for pillar 4 were not high, but on the whole the rate o f execution was satisfactory in that area. The planned and executed B C I totals largely exceeded total PAP estimates, because the B C I covered a number o f areas, such as justice, that were not addressed in the PAP. In fact, B C I allocations for justice were low but increased every year. With regard to support for public administration and institution building, the rate o f execution was on average 71 percent per year during the PRSP period o f implementation but few satisfactory outcomes were noted in that period in relation to good governance and capacity building. D. CONSOLIDATEDINVESTMENT SOURCES BUDGET(BQ FUNDING 3.26 Financing requirements under the first PAP for the period 2001-2004 were estimated at US$475 million,including support for the balance o fpayments (US$121 million) and for the PIP (US$354 million or UM92 billion). Financingneeds exclusively related to priority investments-xpected to have a direct impact on the attainment o f its objectives-were estimated at UM 73 billion. It was planned that those needs would be covered by HIPC funds (UM 28 billion) and budget savings realized in the period 28 considered (UM 14 billion), leaving a net balance o f funding requirements equal to UM 31 billion (US$121 million). 3.27 Inthe period2001-2005, nationalresourcesallocatedto the BCIandconsisting o ftax andnon-tax revenue increased gradually from 34 to approximately 50 percent o f total BCI, or an annual average o f 44 percent. This increase-the result o f a deliberate political decision to finance infrastructure with the State's own resources-was funded from additional revenues gained from HIPC. Such resources were generally used as counterpart funds for projects drawing on external financing or to support productive sectors in the absence o f donors. As counterpart funds, they sometimes covered recurrent expenditures, such as operational costs o f project units, including contract employee wages, fuel, etc. In the period considered, the rate o f execution o f allocations financed with national resources through the State budget process was on average 93 percent. Figure 3.5: BCIFinancing Sources, 1992-2000and 2001-2005 100.0% 80.0% 60.0% 40.0% 20.0% 0.0% 1992-2000 2001-2004 Externalfinancing Source:MAED 3.28 Investment expenditures financed from external resources continued to increase in volume but decreased as a percentage o f total B C I inview o f the significant rise inthe share of national resources. In the period 2001-2005, the external financing envelope consisted o f loans (76 percent) and grants (24 percent). The rate o f execution o f these expenditures averaged 65 percent during that period and decreased to 56 percent in2005. That situation was mainly due to the following factors: 0 limitedcapacity for B C Iplanning and for preparing feasibility studies for the various projects; 0 diversity o f expenditure execution procedures agreed with the various donors and the difficulty experienced by local authorities inmasteringthose rules; 0 inadequate capacity inpublic procurement, and weaknesses inthe regulatory fiamework. 29 E. HIPCRESOURCES PRSPPRIORITIES AND 3.29 Additional resources available to Mauritania inthe framework o fHIPC debt relief increased from UM 2.2 million in 2000 to UM 7.6 million in 2005,44thereby reducing the State's need to borrow. Although only half o f the total amount o f available HIPC resources was spent, the share of BG14' expenditures funded from those resources increased in the period 2000-2003 from 3.5 to 10.8 percent. The additional resources in question enabled the State to take charge o f most o f the investment expenditures theretofore financed with external resources. 3.30 On the whole, the allocationo f HIPC resources seems consistent with PRSP priorities. As shown inthe tablebelow, HIPCresources executedintheperiod2001-2004contributedto the implementation of the first three PRSP pillars. Approximately 49 percent o f those expenditures were earmarked for pillar 2, and inparticular for the multi-sector projects aimed at fighting poverty implemented by the CDHLPCI; 38 percent for pillar 3, and in particular for the sectors o f education, health, water, and energy, whose allocations increased significantly; and 13 percent for pillar 1, exclusively for the transport sector. Moreover, the distribution was directed to areas where the poverty percentages were relatively lower.46 Table 3.2: Breakdown of Executed FIIPC Resources (in millionsUM) 2001 2002 2003 2004 Axe 1 :Stimulating broad-based growth 784 1,500 1,164 1,040 Growth sectors 784 1,400 1,164 1,040 New technology o f information and cornmunicati 0 100 0 0 Axe 2 :Anchoring growth in the economic enviro 2,205 3,180 4,222 3,625 Rural development 105 300 50 50 Multisectoral project for poverty reduction 2,100 2,880 4,172 3,575 Axe 3 :Improving access to basic services 1,264 3,324 4,692 4,145 Education 700 1,297 1,662 1,366 Health and social affairs 354 1,746 1,758 1,400 Water and energy 140 231 1,272 1,379 Youth and sports 70 50 0 0 Total 4,253 8,004 10,077 8,810 Source :MAED 44HIPC initiative resources are comprised inthe hnctional and economic classifications o f expenditures discussed earlier, save for expenditures executed by organizations o f mixed status, such as the Commission for Poverty andthe Commission for Food Security, which are responsible for the execution o f approximately 40 percent o f HIPC expenditures (cf. Chapter 3, section 3.5). 45This BGI contains only the investments financed by the state's own resources, executed through regular budget procedures. 46Refer to the report `Rapport sur la programmation, la mise en ceuvre et l e systbme de suivi pour les dtpenses PPTE, 2003' ofCMAP prepared for the committee incharge o fmonitoring and evaluation o fHIPC revenues. 30 F. CONCLUSIONSANDRECOMMENDATIONS 3.3 1 The analysis revealed that the priority action plan (PAP) drawn up under the PRSP was not used as a basis for planning investment allocations. In order to ensure consistency in pursuing the strategic objectives o f economic and social development, the following measures are proposed: 0 The PAP should identify all integrated activities necessary for addressing the country's economic and social development constraints inaccordance with sectoral strategies and MTEFs; 0 PAP implementation costs should be estimated by identifying their recurrent and investment components, assessing the projects executed, and determiningthe unit costs; 0 Every year, the PAP should be drawn up and implemented with sufficient flexibility to take into account any lessons learned from execution andpossibly to introduce new priority activities; 0 Priority activities should be distributed geographically with a view to addressing specific constraints faced by individual wiluyus; 0 The PAP should be the only basis for programmingthe B C I and the recurrent budget. 3.32 Alignment o f budget programming with the PAP requires systematic recourse to multiyear budgeting in order to allocate resources by strategic objectives rather than by needs and make sectoral departments accountable for the respective sectoral objectives. This approach calls for monitoring and evaluation tools that can provide information on the evolution o f the related indicators. Accordingly, capacity building for planning, monitoring and evaluation in the sectoral units o f the Ministryo f Finance (MF) and the Ministry o f Economic Affairs and Development (MAED) is crucial to the coordination o f the two processes. Support for both exercises under the PRECASP would help to gradually implement this recommendation, particularly through effective integrationof the MTEFprocess into the preparation o f the State budget and the BCI. 31 PART I1 ASSESSMENT OF THE PUBLIC EXPENDITURE MANAGEMENT SYSTEM 32 4. BUDGET PROGRAMMING, EXECUTION, AND PUBLICACCOUNTS 4.1 As noted in the chapter analyzing public expenditure, fiscal policy has been expansionary, with systematic long-standing recourse to off-budget spending which was not included in the budget and accounting reports. Such practices are commonly found in countries with severe shortcomings in their fiscal management system. However, it i s important to recognize that no matter how robust the system and procedures may be, political will i s still essential for the enforcement o f existing regulations. Inthe case o f Mauritania, the current system displays many weaknesses at all levels, as confirmed in a number o f studies prepared by the World Bank and other development partners, but the distortions noted have as muchto do with the irregularpractices usedby the authorities as with these weaknesses inthe system. 4.2 The analysis in this chapter will be based on the conclusions o f the 2004 PER. That report reviewed the conclusions and recommendations o f several World Bank studies on the subject, primarily the 1999 Country Procurement Assessment Review (CPAR), the 2002 Country Financial Accountability Assessment (CFAA), which looked at financial management and private sector accounting practices, the 2002 Report on Observation o f Standards and Codes47 (ROSC), and the 2003 Country Economic Memorandum (CEM), which identified strengths and weaknesses in the public expenditure management system inMauritania. This chapter will review the measures recommended for improvingpublic financial management, in order to assess how the situation has evolved, and propose new measures. A. PREVIOUSASSESSMENTS THE PUBLICFINANCIAL OF MANAGEMENTSYSTEM 4.3 The new authorities have undertaken an evaluation4* o f public financial management through the Inter-ministerial Committee on Governance, which was set up immediately after the coup d'Ctat o f August 2005. The objective was to identify the major obstacles to good governance and to propose realistic measures that could be implemented in part during the transition period in order to lay the foundations for a transparent and efficient system o f government. Infact, the shortcomings and problems identified in this self-evaluation are similar to those found in the external assessments. The report, published inNovember 2005, identifies the following weaknesses inbudget planning: lack o f a long-term vision; lack o f an integrated management o f foreign debt; limitedunderstanding and implementation o f the global MTEF and the sectoral MTEFs; and the strong centralization of budget planning. Regarding resource mobilization, revenues are constrained by inefficient tax administration, the existence o f sectors that are not effectively taxed, and poor coordination with the departments responsible for non-tax revenues (miningroyalties, fishing licenses, dividends, etc.). Among the prime weaknesses o f the current budget execution system are the fact that much spending is done off-budget and i s completely opaque; procedures are cumbersome and delays are long; there are too many execution procedures depending on the type o f financing; there is no system o f cash management and budget regulation; and there is n o reliable information on revenues and expenditures. The report identifies as the main factors impeding effective expenditure control and evaluation the fact that vertical controls and supervision o f public accountants are ineffective, supervisory bodies and boards o f directors add little value, the Supreme Audit Court does not function properly and i s not independent, and there i s no systematic tracking o f public expenditure. 47Preparedincollaboration with the IMF. 48Government evaluation is based o n Donors analyses such as CFAA, RONC, 2004 PER 33 4.4 The CFAA o f November 2002 identifies as major obstacles to proper financial management a lack o f updates o f laws and regulatory texts governing budgetary execution, the concentration o f payment authority at the MF, the frequent exemption from rules and procedures and a lack o f capacity within the control units. While the CFAA was based on false data, it did however underlined that the use and frequency o f exceptions not inline with regulations pointed to either an inadequacy o f the regulatory texts with regards to the reality o f the field, or a lack of strong financial discipline. Indeed, much o f the off- budget expenditures were executed based on these types o f arbitrary exceptions to the rules. The accounting system and the ex post control mechanisms were weak and suffered from the absence o f quality information. The CFAA presented an action plan based on these observations. However, only the recommendations concerning decentralization were implemented in 2005. The remaining recommendations related to the improvement o f budget planning, public accounting and control were only initiated at the end o f 2005. Box 4.1: HIPCAAP Indicators The HIPC Assessment and Action Plan (AAP) indicators for evaluating expenditure tracking capacities, were developed for monitoring and assessing the financial and fiduciary management system o f HIPC countries and cover all fields o f the expenditure circuit. The 16 indicators are divided into four sets, covering: (i)budget formulation, presentation, and comprehensiveness; (ii) execution; (iii) budget budget reporting; and (iv) the procurement system. 4.5 As the assessment o f the fiscal management system performed as part o f the HIPC initiative in 2001was based on erroneous information providedby the Mauritanian authorities, performance was rated satisfactory against 7 o f the 15 indicator^.^' That assessment also proposed a plan of actions to overcome problems in the public financial management circuit at all levels. Those measures, however, were not implemented, and indeed the new economic statistics, which allowed a more accurate assessment o f the situation in 2001 and subsequent years, showed that the system's weaknesses worsened through the end o f 2005. The 2006 AAP assessment, conducted as part o f the G8 multilateral debt cancellation process, found the same shortcomings as those in the Inter-Ministerial Committee's self-assessment. Infact, using the new budget data for Mauritania, the 2006 assessment maintains the same rating for only 2 of the 16 indicators. The two indicators rated relatively satisfactory relate to inclusion o f donor-financed investment spending in the budget, and the use o f economic, administrative, and functional classifications. Yet even for these two indicators, it was noted that the authorities could make further progress. The measures recommended in the 2001 assessment have not been implemented, which explains why the situation has stagnated. 4.6 The 2004 PER, which was also conducted on the basis o f erroneous information, nevertheless highlightedthe main weaknesses o fthe public financial management system inMauritania, and offered a series of recommendations aimed at improving the situation. The system's shortcomings are to be found at various levels, and affect the entire expenditure circuit. Since there was no World Bank support to aide the implementation o f the main reforms proposed in the 2004 PER (which in fact required capacity buildinginaddition to political commitment), few changes havebeen made. The status o f implementation ofthe recommendations made by the PERi s attachedinthe appendix. 49There are now 16HIPC AAP indicators, with the addition of one rating the public procurement system. 34 4.7 With respect to budget comprehensiveness, the PER recommended that mining and fishing revenues," which were being spent off-budget, should be integrated into the LdF. There has been no progress on this point. Noting the existence o f a separate budget that prevents a comprehensive overview o f public spending, the PER recommended gradual integration o f the LdF and the BCI in a consolidated document that includes all sources o f financing for both recurrent and investment spending. While some improvements were noted in the presentation o f the 2006 LdF, this recommendation has not yet been implemented. Moreover, on the basis o f new information, the following analysis shows that the budget i s still far from comprehensive, and that the recommendation i s still relevant. Similarly, with regards to budget presentation, the h c t i o n a l classification has not been instituted as recommended in the 2006 PER. Furthermore, the review highlighted the need to continue and reinforce the MTEF process. No progress hadbeen made on this score to the end o f 2005. 4.8 In terms of expenditure execution, the main recommendation called for the progressive deconcentration o f payment authority, which was centralized in the Ministry o f Finance, to the line ministries, and then to the regions, inorder to shorten the various stages o f the expenditure circuit. This i s infact the only recommendation on which progress was made in2005. The analysis ofthis subject inthe present chapter examines progress and possibilities for improvement one year after this deconcentration to line ministries was launched. On the other hand, the computerization that i s supposed to accompany this reform for the 3 pilot ministries was not implemented until September 2006. The government-wide rollout o fthe computer system is plannedfor 2007. 4.9 Withrespect to improving expenditure tracking at the payment level, the PERpointed to the need to produce regular treasury balances and to reconcile budget and accounting operations, recognizing that the data on budget execution at the payment level are far from comprehensive. Unfortunately, these recommendations had not been implemented as o f August 2005. There had been no effort to produce comprehensive budget and accounting reports or to perform regular reconciliation between the treasury accounts and the balance on the State account with the BCM. 4.10 Interms ofbudget reporting and control, the PERfound that, inspite ofwell-defined rules and procedures, the effectiveness o f the internal control and audit system i s thwarted by the many different participants involved in the system, and by the lack o f regular monitoring and appropriate sanctions. It also found that the other control bodies have limited capacities. Strengthening the capacities o f these bodies was deemed an appropriate measure, and has been implemented in part through assistance fi-om the German and French cooperation programs. However, control remains inadequate at all levels. 4.11 The remainder o f this and the following chapter will re-examine each stage o f the public expenditure management system in light o f the new information on budget execution, and progress in implementing the PER recommendations. Some recommendations that may have appeared essential are less so today, while new problems have emerged. A percentage of fisheries agreement revenues is spent directly by the Fisheries Ministry,and a portion of mining revenues i s spent by the Ministryof Mining and Industry. 35 B. IMPROVING BUDGETPRESENTATION Composition and scope of the budget 4.12 The 2004 PER highlighted the problems posed by the existence o f two budget documents, prepared and executed by different bodies. The Government budget comprises two main documents: (i) the State Budget (LdF), issued by the Directorate o f Budget and Accounts (DBC) o f the Ministry o f Finance (MF); and (ii) the Consolidated Investment Budget (BCI), issued by the Ministry o f Economic Affairs and Development (MAED). The LdF i s the budget for the central government and the deconcentrated services. However, it does not include revenues and expenditures o f public institutions or local governments. The communes51have their own revenues (land taxes, property taxes and local license fees), supplemented by central government transfers (from the Regional Development Funds).52A portion o f government revenues flows directly to the Ministry o f Fishing and the Ministry o f Mining. All government resources should be recorded intheir entirety under revenues and expenditures inthe LdF. 4.13 The scope o f the LdF i s limited, as far as the general investment budget (BGI) i s concerned, to domestic financing. The summary balance sheet presented in the LdF does not include external financing for investment. It shows all operating and investment expenditures financed from the government's own resources, as well as the special accounts. The general investment budget (BGI) i s not consolidated with the general operating budget for individual ministries. Since 2006, it i s simply added to the general operations expenditures o f each specific department. The LdF contains the following components: (i) revenues, (ii) debt, (iii) public ordinary operating expenses, (iv) self-financed capital spending, and (v) special accounts. The LdF presents a summary o f the BCI, to provide an overview of external financing. Duringpreparation o f the 2006 LdF, operating budget items previously shown inthe BCIwere recorded under operations for eachMinistryinthe LdF. 4.14 The BCI summarizes annual investments recorded in the Public Investment Program (PIP), both those financed by the government and those financed externally. The government-financed portion i s presented in the LdF via the general investment budget (BGI). The BCI also covers amortization o f the debt and transfers to the PES,notably the National Industrial and Mining Company (SNIM). It includes all loan-financed and grant-financed investments. Decree no. 86-178 o f October 29, 1986 created the annual BCI. Strictly speaking the BCI does not have a legal status but simplyrecapitulates the investment financed by domestic and external resources and the amortizationo fpublic debt. 4.15 The existence o ftwo budget documents, prepared and executed according to different procedures, increases the risks o f inadequate macroeconomic control and contributes to ineffective resource allocation. B C Irevenues and expenditures are exempt, inpart, from the public accounting rules intended to ensure that public funds are properly used. The MF cannot readily keep track o f investment expenditure financed by external resources, which are executed through mechanisms established in the bilateral agreements with donors. Those resources are in fact deposited in special accounts held at commercial banks. The financial execution procedures used are those selected by the donors, who can effectively control expenditure, making sure that the available funds are properly allocated and spent in a timely manner. 51Mauritania has 216 communes. 52The Regional Development Funds are also annexed to the La. 36 Budget classijications used 4.16 Budget preparation and execution uses administrative and economic classifications. It i s on this basis that the DBC establishes the payment authorization officer's administrative account (compte administratif de I'ordonnateur). A specific chart o f accounts defines the classification for PES,EPICS, and local governments. The administrative classification i s divided into "titles" (ministries) and "chapters" (directorates). Ministries are identifiedby a two-character serial coding, as are the directorates and services within each chapter. The BGI sub-chapter covers investment projects. The administrative classification currently in use contains no geographic information on the administrative office receiving these allocations. The economic classification i s divided into "parts," "articles," and "paragraphs." The "parts" describe the general type o f expenditures, the "article" presents the nature o f the expenditure, and the "paragraph" specifies the character o f the expenditure. Neither o f these classifications makes any reference to the location o f expenditure byregion. 4.17 A start was made in2001 at introducing a functional classification, but it has never been used in the presentation o f the LdF.An attempt has now been made to reclassify the 2006 LdF on a functional basis. This exercise will pave the way for the presentation o f the 2007 LdF with a functional classification, in addition to the other two classifications. The fbnctional classification, divided into classes, groups, and subgroups, determines the function o f expenditures and i s independent o f the administrative classification, which may vary over time and therefore does not allow for long-term analysis. 4.18 A sectoral classification is used in the BCI, which has a presentation different from that o f the other classifications. Sectors, which are the basic categorization unit in the BCI, are not connected to ministries, which are the basic categorization unitsin the LdF. Nonetheless, the MAED computer system has been harmonized, allowing for a direct link between this classification and the LdF. This means that the BCIcan be presented according to the administrative and economic classifications. c. OPTIMIZING THE MTEFAPPROACHINTHE BUDGET PREPARATIONCYCLE 4.19 The 2004 PER stressed the importance o f improving the MTEFprocess and integratingit into the budget preparation calendar, but no steps had been taken inthis direction by the end o f 2005. The budget i s formulated "on the spot," without any clear schedule identifying the different steps and defining the responsibilities o f those involved in the exercise. It i s the MF and the MAED that jointly direct the process o f preparing the government budget. The conventional budget cycle has four main phases: (i) preparation o f the macroeconomic framework; (ii) preparation o f the budget framework, (iii) o f holding budget conferences; and (iv) adoption o f the budget. As o f 2002, the medium-term expenditure frameworks have been integrated into the budgetpreparation process. 4.20 The first stage o f the budget preparation cycle consists o f projecting the main macroeconomic aggregates on which government priorities and the budget forecast are based. The Planning and Research Directorate o f the MAED, working with the BCM and the MF, prepares this macroeconomic and financial framework. Macroeconomic and financial forecasting shows a number o f weahesses. Forecasts are not adequately performed, given the absence o f forward planning and the limitations o f the tools used. The second stage o f the process is to set the budget framework, which i s the responsibility o f the DBC. This directorate estimates the level o f tax and non-tax revenues and o f external financial support for the coming year (n+l).Next, the thresholds o fthe broad expenditure categories are set, inorder o fpriority. 37 4.21 Generally speaking, once the macroeconomic and budget frameworks are in place, the MF- MAEDjointly issuecirculars to the ministerial departments. The departments submit their financial needs to the MF-MAED, estimated in light o f their sectoral objectives. The departmental proposals may be taken from their sectoral MTEF, if this i s available and up-to-date. On the basis o f these responses, the MFandMAEDperformthe necessary trade-offs among the sectoral departments. 4.22 Following approval o f the PRSP (Poverty Reduction Strategy Paper) in 2001, the government adopted the MTEFapproach in order to operationalize its development strategy. The sectoral MTEFs are powerful tools, particularly when they are consistent with a global MTEF that can track the financing constraint. The principles underlying the medium-term expenditure framework are the following: prepare a coherent and realistic budget framework that enhances macroeconomic stability and allows projections o f revenues, expenditures, deficit, and debt over a three-year horizon; 0 set up a system o f resource mobilization that i s consistent with sectoral and intersectoral strategic priorities and budgetary constraints; 0 foster better participation and greater autonomy o f administrative structures in the elaboration o f the budget and promote more efficient use o f resources programmed within a global and rigorous framework o fbudgetary discipline. 4.23 Between 2002 and 2004, eight MTEFs were prepared for the following sectors: education, health, rural development, transport infrastructure, urban development, energy, fisheries, and water and sanitation. This sectoral programming exercise was not as useful as it might have been, given the absence o f a global MTEF determining the macro budget framework and setting the sectoral envelopes. T o remedy this situation, the first global MTEF was prepared in 2003 for the period 2004-2006. The 2004 LdFwas prepared onthe basis o fthis document. An analysis o fthe global MTEFandthe sectoral MTEFs shows that the system has a number o f weaknesses that should be corrected in order to make the three- year budget programming exercise more useful. 4.24 Apart from the weakness o f the macro-budgetary framework forecast in the global MTEF, the main shortcoming o f the current MTEF has to do with the intersectoral allocation o f spending. Over the period 2004-2006, the portion o f sectoral spending in total spending has remainedremarkably constant in comparison with past trends. It should be noted that the sectoral envelopes are allocated by the MAED and the MF on the basis o f budget priorities. However, in the absence o f a reliable analysis o f the relationshipbetween sectoral priorities and objectives, on the one hand, and sector budgets on the other, the committee responsible for preparing the MTEF allocates funding on the basis o f past trends, the inflation rate, and existing policies. There are sometimes significant discrepancies between the overall envelopes by sector (shown inthe global MTEF) and the needs that emerge from the sectoral MTEFs. 4.25 An evaluation o f the performance o f past public policies should be useful for programming the MTEFs and the budget. Such an evaluation does not take place in Mauritania, however, for several interrelated reasons: monitoring and evaluation capacities are very weak at all departmental levels, which means that there i s frequently no evaluation o f actual policy outcomes; the statistical information system i s unreliable; reflecting the first two problems, there are many gaps in the definition of objectives and indicators; 38 while there are occasional evaluations o f strategies at the sectoral level, their outcomes are not necessarily reflected inbudget decisions, and they are not sufficiently disseminated. 4.26 Continuity in the MTEF process is not guaranteed, because o f a number o f problems in implementing the approach, run by operational directorate^^^ that are preoccupied with day-to-day concerns. They focus on the preparation of the document for only a short period, without any real involvement from the sectors. Moreover, the MTEF approach has not been integrated into the budget preparation cycle or into the preparation and monitoring o f development strategies. This situation reflects the departments' weak capacities and the inability o f the technical infrastructure to bringcoherence to the process. 4.27 For the time being, then, the process i s limited to producing documents, and there has been no attempt to plan more accurately either the resources committed or the results expected. In short, what i s lacking is: (i) coherence, a calendar, and a precise working method; (ii) development o f sector overall the strategies that will provide a basis for medium-term planning; and (iii) proper involvement o f sectoral managers inthe process. 4.28 Mastering the MTEF approach, however, i s an important step in the transition towards performance-based program budgeting. Tunisia experience presented inthe box 4.2 can be considered as good example for Mauritania. Consideration i s now being given to introducing program budgeting in Mauritania, starting with basic and secondary education. However, the prerequisites for the effective use o f this approach are not yet in place. Those prerequisites relate to several levels: budget preparation, questions about the personal and financial liability o fpublic accountants, empowering program managers, transparency inbudget information, a clear determination o f programs as well as objectives and concepts, reorganization o f the administrative structure and clarification o f decision-making channels, and a review of the accounting system with introduction o f analytical accounting in association with accrual accounting. s3The DBC, the DPE, and the CMAP are the main entities responsible for coordinating the MTEF approach. These three entities all have very heavy workloads that prevent themfrom focusing on this problem. 39 Box4.2: Tunisia's ExperiencewithPerformance-BasedBudgeting andDecentralization ofBudgetExecution Performance-based budgeting. Tunisia's approach to performance budgeting could serve as an example for Mauritania in improving its multi-year budget planning and the presentation o f the budget. A management unit was set up w i t h the Ministry o f Finance to steer the work o f revising forecasting methods, the t r a c h g o fbudget execution, and evaluation o f expenditure using the management-by-objectives method in order to link strategic objectives to the means needed to achieve them. This is a restricted, high-level unit with a clearly defined mandate that runs for five years. The advantage is that this unit i s devoted full-time to budget reform, whereas the t e c h c a l directorates that are supposed to work o n this issue remain preoccupied with day-to-day concerns. The unit's work is overseen and evaluated by a commission within the Ministry. Its tasks include: (i) preparing a functional and programmatic budget classification; (ii) establishingperformance standards and indicators; (iii) up a program for tracking and evaluating the setting achievement o f objectives; (iv) designing staff training plans in management by objectives; (v) assistance to other ministries in implementing management by objectives; and (vi) preparation o f the new budget structure, o n this basis. Budget deconcentration. The Tunisian budget system is highly deconcentrated and could offer an example for reform, both vertically and horizontally (to the regions). The bulk o f operating and investment expenditure in Tunisia is managed at the local level in a deconcentrated form, or in a form close to decentralization: (i) centralized expenditures are managed at the central level (the ministry) by a central authority (payment authorization officer or ordonnateur); (ii) deconcentrated expenditures are managed at the local level (gouvernorat) by a State authority represented at the local level (Governor or regional directors o f ministries); (iii) allocations to the Governor for financing deconcentrated expenditures are known as "delegated appropriations," andthey are listedinthe equipment budget classification as "national interest appropriations"; (iv) decentralized expenditures are managed at the Regional Council level and are executed by the Governor, as representative o f the region; (v) allocations to the region for financing decentralized expenditures are referred to as "transferred appropriations," and are listed in the budget classification as "regional interest appropriations." Source: CFAA of Tunisia, World Bank, 2004 D. IMPROVING BUDGETEXECUTIONPROCEDURES 4.29 As noted in the introduction, the deconcentration o f budget execution recommended in the 2004 PERbegan to be implementedin2005, with success in some areas and gaps in others. It is time to draw some lessons from the first year o f this deconcentration inorder to consolidate and expand this reform. 4.30 Until the end of 2004, responsibility for executing the LdF was concentrated in the MF. The Minister of Finance alone had the authority, as ordonnateur unique, to order the payment o f budgeted expenditures. H e delegated this power to commit and authorize expenditures to the Budget Director. Line ministries thus had no responsibility for executing their department's budgets, since they could not initiate the spending process or authorize any o f the stages in the expenditure circuit. According to existing legislation, the "ordonnateur" should not only supervise the pace at which expenditures are committed, but should also judge whether particular expenditures are justified. Clearly, the DBC does not'have the capacity to pass suchjudgment, and so it approves decisions taken elsewhere. 4.31 This centralization o f budget execution within the MF means that execution times are too long. The budget circuit o f departments not involved in the deconcentration o f budget execution causes excessive delays between expenditure commitment and payment54. A review o f the 13 steps that constitute this budget circuit shows, on one hand, that there are many parties involved and, on the other 54Military expenditures are directly initiatedby the minister o f defense 40 hand, that controls are redundant. Moreover, the current system o f control entails some risks. The Budget Director and his assistants are supposed to review all accounting documents several times and authorize them personally. This imposes a tremendous workload on this directorate, and means that a significant portion o f expenditures i s not tracked in real time, and that various management procedures are used. In other words, this lengthy budget circuit i s one o f the reasons why closing and opening dates for successive fiscal years mustbe extended. Box 4.3: FiscalManagementMeasuresIntroduced in2004 Deconcentration o f budget execution. Certain articles o f the general public accounting regulations bearing on budget execution were amended by Decree no. 096-2004 of September 8, 2004. An order was adopted specifying the ministries to which budget execution was to be transferred. To date, most o f the regulations for implementing the reform have been adopted, in particular the decree instituting the deconcentration o f budget execution and the decree creating the new financial control system inpilot ministerial departments. Deconcentration o f financial control. Decree no. 023-2004 o f March 13,2005. instituted financial control inministerial departments with budget execution for their own budgetary allocation. The application order on financial control is in the process of approval. Financial control at the ministerial level: Decree no 023-2005 of March 13. 2005, mandates financial controls in departments qualified to execute expenditures. The implementation is being finalized complementarily with the execution o f the RACHAD software. Comuuterizing the suuulies and eauiument expenditure chain. A draft decree regulating computerized management o f spending on supplies and equipment and repealing certain provisions o f Decree no. 74-187 on automation of public expenditures has been prepared but not yet adopted. Its adoption is a prerequisite to effective implementation o f the RACHAD software. 4.32 At the beginning o f 2005, Mauritania launched a reform o f budget execution, involving gradual deconcentration o f budget execution for supplies and equipment to certain pilot ministries. This reform is currently limited to supplies and equipment expenditures, a large portion o f which i s included in the "common expenditure^"^' managed directly by the Ministry of Finance. It i s based on redistributing tasks within the budget execution framework, with the line ministries as the principal payment authorities (ordonnateursprincipaux), a reinforcedrole for financial control, and better tracking bythe Directorate o f Budget and Accounts (DBC). The deconcentration o f budget execution for supplies and equipment is currently limitedto six pilot ministries, selected for their size and the importance o f the sector policies for which they are responsible. These are the Ministry o f Health and Social Affairs, the Ministryo f Basic and Secondary Education, the Ministry o f Higher Education and Scientific Research, the Ministry o f Rural Development and Environment, the Ministry o f Water Resources, and the Secretariat o f State Responsible for New Technologies (SETN). 4.33 This partial deconcentration o f budget execution was accompanied by a modification in the organization o f financial control, which until now has been centralized in the Prime Minister's Office (Primature). Prior to the 2005 reform, a financial controller in the Prime Minister's Office was responsible for overseeing the regularity o f all expenditures, as well as the recording o f commitments. The maintenance o f the accounting function at this level was deficient. Now, four financial control officers (FCOS),~~ reporting to the MF, have been placed in the departments benefiting from the deconcentration. This reform o f financial control is an important stage o f the overall reform o f the public 55 These common expenditures represented about 20 percent of the budget in 2004, but no more than 12percent in 2006. 56Two new FCOshave been assigned to the Departments ofWater Resources and ofHigher Education. 41 finances, and should contribute to its success. These financial controllers participate in overseeing execution o f the LdF, with respect both to appropriations and staffing. Their field o f responsibility has thus been split off from that of the current financial control system, which remains responsible for all departments to which budget execution for their appropriations has not been devolved. 4.34 Placing financial controllers inthe pilot ministries without reforming the current system leads to problems. The first problem i s that there are two categories o f financial controllers, without any apparent link.The ex ante control over budget executionby financial controllers for the departments inwhich they are placed i s not directly linked to the financial control office inthe General Secretariat o f Government. The second problem has to do with organization o f the financial control office, which the Ministry o f Finance has not clearly specified. In addition to these problems with financial control, the relationship between the General Secretariat o f the line ministries and the financial directors i s not sufficiently clear for instituting a payment authorization unit. It has also been found that the line ministries do not yet have sufficient capacity to carry out this decentralizedmission. 4.35 As part o fthe reform, each ordonnateur is supposed to prepare an "annual initialbudget planning document" (DAPBI).This DAPBIshould set out the conditions under which the LdF i s to be executed, and i s supposed to be in place before pilot departments receive their allocations. The FCO reviews the DAB1to ensure that the document as a whole is consistent and that the forecast expenditures and staffing levels are realistic, and to analyze the consequences o f the forecast charges for the public finances and for the coverage o f obligatory expenses. However, this document bears no direct relation to the cash management plan, which i s regularly updated inthe Treasury. 4.36 The simplified expenditure circuit resulting from this reform reduces the number o f steps from 13 to 6. The commitment slip (bon d'engagement) issued by the Information Processing Units in each spending ministry becomes, after validation by the respective ministry's FCO, the confirmation form (titre de confirmation) that allows the order to be placed. Following endorsement by the FCO, the spending ministry issues the payment order. The DBC no longer performs checks but centralizes all the payment orders and transmits them to the Treasury, which does all the necessary verification and will then either make the payments or reject the expenditure as irregular or insufficiently justified. The regulations have not set any timeframe within which the parties involved must perform the necessary checks. The concentration o f payment orders within the DBC could pose a bottleneck to efforts to simplifythe circuit. Moreover, without computerization o fthe expenditure chain, the reformitselfhas not shortened expenditure execution times as planned. 42 Box 4.4: RecentMeasuresTakenAfter the Coup d'Etat of August 3,2005 Following the coup d'ktat o f 2005, the new government took steps to rectify the fiscal situation by preparing a supplementary budget that captured existing off-budget spending commitments. All public servants were reminded o fthe basic rules o f financial management: ODening o f sub-accounts inthe current account o f the Treasury with the BCM for public establishments and projects with government counterpart funding: Circular no. 05 o f September 12, 2005, requires that disbursements o f counterpart funds, subsidies or any other funding to public entities (EPA, projects etc.) mustbe done through sub-accounts o fthe Treasury account at the BCM. Reminder o f the basic rules o f routine bookkeeping. In the Circular o f September 21, 2005, the MF reminded all regional treasurers, embassy accountants, collection officers and accountants o f public establishments about the importance o f observing basic accounting rules inhandling expenditures o f the State and local governments and incollecting public revenues. Reminder o f provisions relating to budget closing dates. A circular o f October 3, 2005, reminded ministers and State secretaries o f the closing dates for the fiscal year. As a result, the accounts for 2005 were closed within the regulatory deadlines. Review o f the "chancellery rate" (taw de chancellerie). The recognition o f foreign exchange losses related to expenditures abroad (in particular, by embassies and students on bursaries for foreign study) has undergone a major change with adoption o f a benchmark exchange rate for 2006 that is close to the BCM's rate. This explains the sharp increase inbudget allocations to the Ministryo f Foreign Affairs and Cooperation and the Ministry o f Higher Educationand Scientific Research. Public expendituremanagement at the local level 4.37 There are serious shortcomings in the management o f decentralized funds at the local There i s no systematic collection o f information through the Treasury and the level o f accountability i s very low. The administrations concerned receive notification o f their allocations at the beginning o f each year. That notification follows the same circuit as the commitment slip. The regional representatives o f the MF (the regional treasurers) are responsible for verification and payment. These expenditures are regularized, i.e. a payment order i s issued, only ex post, after control by the verification unit. That unit, located within the General Treasury, includes representatives o f the budget office, the financial control office, and the general treasurer. This cumbersome and time-consuming approach also leads to delays in preparing documents for closing the accounts. The verification unit takes a long time to perform its work. It is an anomaly that all transactions conducted at the local level should be treated as provisional operations, despite the involvement o f so many officers and representatives o fthe central government. Expenditures executedby the two commissariats 4.38 The 2004 PER highlighted the problem o f hybrid entities that spend money outside the conventional budget circuit, and the need to bring them within that circuit. There was no consensus on this measure among the authorities, and there seems to have been no movement in this direction. The entities in question, namely the Commissariat on Human Rights, Poverty Reduction and Integration (CDHLPCI) and the Commissariat on Food Security (CSA) continue to spend public funds that are allocated to them off-budget. Their task i s to deliver public services to the poorest and most vulnerable population groups, and they do so through parallel funding channels. It i s true that these entities disburse their funds more quickly on average (24 days) than do the ministerial departments that must work through the conventional budget circuit. For example, it takes the Ministry of Rural Development and ~~ ~~ 57 wilayas, moughataas, diplomatic missions 43 Environment (MDRE) six times as long to execute expenditure as it takes the CDHLPCI. Yet this slow performance should not lead the authorities to allow increasing amounts o f poverty reduction funds to be exempt from the controls imposed on other budgetary expenditures. On the contrary, the budget circuit should be improved in order to shorten the delays, and the public funds transferred to these agencies should be gradually brought within the regular expenditure system, in accordance with current regulations. Respectingbudget cut-off dates 4.39 The 2004 PER didnot dwell on the issue o f extensions to budget cut-off dates, but this has been a recurrent problem for several years: in fact, two budgets from different fiscal years can continue to be executed at the same time, because o f a supplementary period that runs beyond the cut-off date set by the organic law. According to current legislation on the closure date for budget operations, all expenditures are supposed to be adjusted or posted by March 15 o f the following year. On several occasions, expenditures were allowed to be authorized for payment after the end of the fiscal year (December 31), whereas the law requires that payment authorization must stop on December 15. The fact that appropriations cannot be carried over, together with delays in the procurement process, explains in part why investment expenditures are authorized after December 31 as a charge to the budget for year n-1. Indeed, payments have been made after March 15 o f the following year. This situation frustrates proper budget execution, because for several months in any given calendar year there may be two parallel budgets inexecution. 4.40 Whatever the justification for these irregular practices, they clearly diminish the transparency o f budget execution and o f government operations. This situation poses problems o f two kinds for the Treasury. First, in terms o f the financing and cash management plan, expenditures for fiscal year n-1 are paid during fiscal year n+2, whereas the financing forecasts do not cover these operations. This impedes the preparation o f summary statements such as the Government Flow o f Funds Table (TOFE), since budgetary execution data are supposed to be documented for each fiscal year. The second problem relates to the production o f accounts. This procedure generates considerable delays in closing the accounts and producing the budget execution law (act approving budget execution) on time. Because o f this situation, the Treasury must work with a provisional balance sheet for an abnormally longperiod o f time. 4.41 Recent measures have been taken to close the accounts for fiscal year 2005 by the regulatory cut- o f f date. The government that took office after the coup d'Ctat o f August 3,2005 is committed to putting an end to the irregular procedures. Ina circular o f October 3, 2005, the Minister o f Finance reminded all ministers and secretaries o f State o f the legal provisions governing the closure dates for budget transactions. H e asked them to respect those closure dates for all their budget allocations. As a result o f these instructions and provisions, notifications were closed by October 30, 2005, commitments by November 20, 2005, and payment orders by December 15, 2005, consistent with Decree 84-189 o f August 21, 1984. In 2005, supplementary days were taken as accounting days in compliance with the regulations. E. COMPUTERIZINGTHE EXPENDITURE CHAININ SUPPORT OF DECONCENTRATION 4.42 The 2004 PER stressed the importance o f deconcentrating budget execution and o f supporting this change with suitable software. However, the first year o f the deconcentration reform has not been accompanied by computerization o f the expenditure chain as planned. Each entity still runs its own application, which i s not necessarily compatible with those o f the other units. Data transfers from one system to another are still done largely by hard copy or diskette. The main responsibility o f the Directorate of Information Technology (ITD) i s to record public expenditure transactions. The current computer system dates from 1975 and has not been upgraded since. It i s an outmoded system that does 44 not allow for ongoing management o f operations and cannot provide an overview o f the situation for proper cash management. This old system does allow the ITD to inform the DBC o f the production o f confirmations and payment notices. The computer system theoretically lets the D B C perform regular checks on the availability o f budget funds at the stages o f commitment and payment order. This information was used to prepare the monthly budget execution statement (TOFE) and inprinciple serves as the basis for verifications by the DBC and financial control office. 4.43 In order for payment authorization to be successfully deconcentrated, the execution o f expenditures on equipment and supplies was reformed with the introduction o f a computerizedsystem for managing these outlays. The system uses a central database that contains all data relating to budget preparation and execution. The new Rbeau Automatid de la Chaine des Ddpenses (Automated Expenditure Chain Network, "RACHAD") i s run by a software package that integrates all budget execution procedures, as well as parameters relating to the country's fiscal environment. This application has two components: the "budgetary allocations" component, which is exclusively in the hands o f the Ministry of Finance, allows for management of the allocation of appropriations, and monitoring and regulation o f budget execution; the "spending entity" component, reserved to the ministerial departments, allows for the execution o f expenditures following the commitment, verification and payment authorization stages. 4.44 The current product does not include a payment module that would connect payment officers to the same network as other players in the circuit, for purposes o f malung payments. After testing and validation, RACHAD was delivered inNovember 2005. According to current information, the application i s capable o f deployment in all ministerial departments located within the Nouakchott administrative zone, where fiber-optic networks already exist. However, budget management for 2006 began with the outmoded 1975 system still in use. This makes it unlikely that they will be any prompt improvement in budget execution interms o f reducing the time lags inthe circuit or the supplementary extension to cut- off dates. F. IMPROVINGPUBLICACCOUNTINGAND REPORTING 4.45 None o f the 2004 PER recommendations on public accounting and reporting had been implemented as o f the end o f 2005. Although a general circular from the DTCP describes developments in the government accounting framework and clearly defines the way the accounts and procedures are supposed to work, the accounting system is in fact a hybridone: double-entry bookkeeping i s used at the central level, but single-entry in the regions other than Nouakchott. Mauritania's chart o f accounts, created in 1992, is intended to describe and control operations with the general budget, the special accounts, and the supplementary budgets, and to provide information for budget management and control purposes. It draws heavily on the French chart o f accounts, and includes nine classes.58Despite this, the chart o f accounts i s not sufficiently detailed to track expenditures by type through to payment. 4.46 A proper public accounting system should allow the DTCP to establish a monthly cash balance and to prepare annual summary statements o f the Treasury, together with substantiating documentation. Prior to August 2005, these important statements were not regularly produced, and information on budget execution at the payment stage was therefore far from comprehensive. Apart from the problem with the hybridaccounting system, some payments were made directly by the BCM and were not recorded inthe treasury accounts. Moreover, the "chancellery rate" used in embassies did not reflect the real value o f the ouguiya (UM).Efforts to improve the situation were begun in2001, but have produced no tangible results interms ofthe financial statements. 58 Three (3) classes of accounts are not operational: class 6, "Expense account," Class 7, "Revenue account," and Class 8, "Special account." 45 4.47 A treasury balance was produced at the end o f 2005 and was used to reconcile treasury records with those o f the Central Bank o f Mauritania. This balance, which constitutes a summary statement o f government accounting operations for a given period o f time, i s computerized. It i s automatically updated whenever a new transaction is centralized. The summary statement can be used to determine the net situation (of debits and credits in the opening balances, budget operations, totals and balances. There are some gaps inthe summary statement, which does not cover all revenues inthe budget execution accounts. Moreover, the opening balances from the budget execution accounts are not carried forward to the projectedbalance sheets. 4.48 Ithas not been possible to produce comprehensive summary accounts, because of the limitations in the accounting method used and irregular practices. The summary account should demonstrate the government's net position and all transactions for the fiscal year that affect that net position (stocks and flows). Compiling itrequires double-entry bookkeeping at all levels. According to current regulations, the Treasury's summary account, together with substantiating documentation, i s supposed to be sent to the Cour des Comptes for audit and for preparingthe Budget Execution Law.59The DTCP i s hampered in its production by the lack o f full information on State assets and liabilities. Given this situation, there i s no comprehensive operating account that can be reconciledwith the administrative account. G. CONCLUSIONS AND RECOMMENDATIONS 4.49 The above analysis shows that the public financial management system suffers from inadequacies at all levels, despite efforts made in the last few months. The system as it stands does not allow rational management o f the substantial revenues expected,60 nor does it constitute a basis for development o f a modern economy. Most o f the recommendations in the 2004 PER are still relevant. They are repeated here, with due regard to the new challenges revealed in the analysis o f current procedures for budget programming, execution, and accounting. They have been thoroughly discussed with the Mauritanian authorities and there i s overall consensus on them. Moreover, most o f these measures are covered inthe report o f the inter-ministerial committee on governance and are reflected inthe new PRSP plan o f action. Their implementation will require capacity building and will be taken up, in part, by the Public Sector Capacity BuildingProgram (PRECASP).61 4.50 The existence o f two budget documents prepared and executed in different ways does not . contribute to efficient budget execution. Therefore it i s recommended : to merge the two budgets (LdF and BCI) into a single document, over the medium term; to conduct a detailed study within the PRECASP in order to determine the best approach for . integrating all State financial transactions into a single document as a move toward harmonizing expenditure execution. to make the budget more comprehensive by including all the PES(for example the University o f Nouakchott, the National Statistics Office, etc.) and the major EPICs such as the CNSS, which has a large budget6* ~~ 59The Budget Execution Laws for 2001 and2002 were prepared without a comprehensive operating account. 6o Additional resources generated by oil production, cancellation of multilateral debt, and official development assistance. 61A US$13 millionproject financed byIDA, with a fiscal component ofmore thanUS$5 million. 62The other EPICs, which are involved inindustrial and commercial activities, have financial autonomy and are subject to private sector accounting rules, and would therefore remain outside the general public budget 46 4.51 Interms ofbudget presentation, itis importantthat it shouldallow spending onsocial programs . andpoverty reduction to be tracked: .. inthe short term, the administrative andeconomic classifications shouldbe improved; introduction o f a geographic component into the administrative classificationi s urgent inorder to make operational the regional poverty reductionprograms. it is important to continue to use the functional classification for tracking poverty reducing and social sector spending. 4.52 The government should reinforce its current policy, which is to prepare a PRSP and a global MTEF and, within that framework, a series o f sectoral MTEFs that constitute operational action plans. Mauritania might consider the approach taken by Tunisia (box 4.2) in this area. The following measures are proposed: . to set up a temporary unit within the Ministry o f Finance to work solely on this reform, instead o f . leaving it to various operating directorates that may be preoccupied with their day-to-day activities; to put in place a procedural mechanism to ensure the system's sustainability, by linking the .. PRSP, the sectoral strategies, the global MTEF and the sectoral MTEFs with the preparation o f the State budget; consideration should be given to institutionalizing the budget process, usingthe MTEF approach and performance-based program budgets. the global and sectoral MTEFs should be updated annually. Preparation o f the sectoral MTEFs . should be extended to all sectors based on the country's economic and social development priorities, as set forth inthe overall strategy and the sectoral policies. the quality o f medium-term forecasting o f sectoral MTEF should be improved so that objectives can be reliably linked to available (or any incremental) resources. . 4.53 Introduction o f performance-basedprogram budgets will require : amendments to budget terminology to introduce classification by program, . amendment o fthe accounting and financial tracking system to include the new classification, and training for staff to understand clearly the goals and the methodology o f program budgeting. The program budget approach should be introduced on a pilot basis in three well-run sectoral departments, such as those for the basic and secondary education, health, and water and sanitation. 4.54 Success with the reform started in2005 for deconcentratedbudget execution will require: specifying more closely the powers o f the financial controller attached to the General Secretariat o f Government vis-a-vis those o f the financial controllers attached to the Ministryo f Finance, and on the other hand, designating the entity within the Ministry o f Finance responsible for coordinating the work o f the financial controllers. In Morocco, for example, the controllers belong to the Expenditure Commitments Inspection Directorate (CED), an institutionally and hierarchically independent body located within the Ministry o f Finance. The Inspection Directorate i s headed by a Comptroller General. The training o f future financial controllers shouldbeginpromptly, beforethe reformi s extended to other departments. 47 . statutory time limits should be set for each control procedure in order to prevent excessive payment delays. 4.55 When it comes to deconcentrating expenditure commitment and authorizationto the regions, this should be done gradually, and only after budget execution has been fully deconcentrated to the line ministries. As noted above, the line ministries would become "ordonnateursprincipaux," and the regional directors of the ministries must become "ordonnateurs se`conduires," reporting to the CdC, which i s supposed to enforce budgetary regulations. Expenditures covered by notifications o f appropriations could be considered as definitive transactions and recorded as such by the local public accountant. This reform should be accompanied by the creation o f financial control offices at the regional level (wilaya, moughutaa,or the ministerial level). The financial controllers should send monthly reports to the DBC on the expenditures committed and authorized by the ordonnateurs se`condaires at the regional level. This information would allow the D B C to settle the local administrationaccounts inreal time. Inaddition, the Treasury would benefit from a revision o f the system for centralizingpayments made by the decentralized accounting offices (there is seldom any reconciliation between the statements o f expenditures executed at the decentralized and at the central levels). This would allow for up-to-date accounting, the regular production o f statements on expenditure transactions from the local managers, and the filing o f substantiating documents. 4.56 The introduction o f RACHAD in the ministries i s essential for malung any significant improvement in the effectiveness o f expenditure, and for achieving the objectives o f deconcentrating payment authorization in terms o f reducing budget execution times. It i s also important to determine and apply time limits for each officer inthe expenditure chain, in order to limit the time taken to complete the transaction. Development o f a payment module linking RACHAD to the Treasury would help streamline the circuit and ensure that expenses are paid on time. Over the medium term, introduction o f compatible computer systems for the entire budgeting process (revenues and expenditures) and for the administrative and accounting phases will speed up budget preparation and execution, improve formal controls, and make data available more promptly. Compatible software must be used for budget execution in all ministries (MF,MAED, and the line ministries) involved inthis task. 4.57 The,length o f the "supplementary period" should be enforced, strictly consistent with existing regulations. This period should only be used to settle expenditures authorized before the end o f the fiscal year. 4.58 With respect to budget accounting, the DTCP needs to be reorganized to split off the Treasury . functions. As a short-term measure : a unit should be set up, headed by a chief accountant who would be responsible for expenditures. A comprehensive reorganization could be envisaged with a view to creating (i)a treasury and government accounting directorate to manage the Treasury, keep accounts, and produce reports; . and (ii)for external Treasury offices with? a network o f government accountants answering to the central directorate. as for the production o f accounts, double-entry bookkeeping should be extended to all external . treasury offices, links should be established between treasury accounting and budget accounting, and reconciliation o f the BCM and treasury accounts should be further improved. buildingon progress made inrecent months, the balance o fTreasury accounts should continue to be regularly posted. Accordingly, all revenue should be recorded in budget execution accounts, and the opening balances o f those accounts carried forward to the projectedbalance sheets. 48 5. BUDGET CONTROL, AUDITS AND PUBLIC EXPENDITURE REVIEWS 5.1 The purpose o f this last chapter i s to review the evaluationo f the budget control system inlight o f developments since the 2004 PER was finalized. The 2004 PER revealed problems with the different budget controls. Apart from the recommendations for strengthening the capacities o f the financial controllers, the Cour des Comptes and Parliament, which have been carried out in part, a number o f the inadequacies examined earlier still persist. The various control bodies have now been joined by a new institution, the General Inspectorate o f State (Inspection Gdnkrale d 'Etat, IGE), which has not resolved the problem o f overlap and lack o f coordination among the control institutions. The existence o f these multiple control structures63has not, in the recent past, assured sound management o f the public finances inaccordance with existingregulations, nor has itproduced accountability inthe use ofpublic resources or lead to imposition o fthe penalties stipulated incase o f irregularity or fraud. A. IMPROVINGINTERNALAND EXPOSTCONTROL Financial control 5.2 Normally, the financial controller is supposed to play an important role in ensuring the proper use o f public funds. Untilthe end o f 2004, financial control did not lie with the MF, but was placed under the General Secretariat o f Government (SGG) in the Prime Minister's office. The transfer o f authority for budget execution required the introduction o f a financial control office reporting to the Minister o f Finance to oversee the new ordonnateurs. However,, the imperatives o f budget management and regulationmake it necessary to return financial control to the Ministryo f Finance. 5.3 Decree no. 023-2005 o f March 13, 2005 instituted financial control over those ministerial departments that had the authority to authorize payments according to their budget allocations. That decree was to be followed by a number o f regulations, now in preparation, for application in the ministries and inentities that receive public funds. The area o f responsibility o f this new financial control has been split off from that o f the centralized financial control within the SGG, which remains responsible for all departments to which budget execution for their allocations has not been devolved, as well as for common expenses and payroll expenses. What i s novel about this new financial control system i s that it supervises budgetary compliance andno longerjudges the appropriateness o f the expenditure. 5.4 The establishment o f these new financial controllers raises concerns o f two kinds, relating directly to the oversight function. The first problem i s that there are now two categories o f financial controllers, without any apparent link. The ex ante control over budgetary execution by financial controllers for the departments in which they are placed is not directly linked to the financial control office in the General Secretariat o f Government. The second problem has to do with organization o f the financial control office, which the Ministry o f Finance has not clearly specified. In addition to these problems with financial control, the relationship between the Secretary General o f the line ministries and the financial directors i s not sufficiently clear for institutinga payment authorizationunit. 63The control institutions are the Financial Control Office (FCO), the Internal Inspectorate of the Treasury, the Inspectorate General o f Finance (IGF) o f the MF, the Central Directorate for Combating Economic and Financial Crime, and the Inspectorate General o f State (IGE) inthe Prime Minister's Office. The CCM (Central Procurement Commission) and the CDMs (Departmental Procurement Commissions) also play a role inoverseeing procurement expenditure. 49 Internal control in the Treasury 5.5 The Directorate o f Treasury and Public Accounting (DTCP) has an internal inspection office that conducts simple controls over cash balances and reporting. There i s in fact some conhsion between the duties devolved to this treasury inspectorate and the tasks o f the IGF. As with the IGF, the treasury inspectorate prepares an annual work program covering the main fields o f supervision. For example, in 2005, on the basis o f the annual program for fiscal year 2005, the inspectorate conducted audits at ROSSO, Nouadhibou, and the Ai'ouri-Nema corridor. Implementation o f the supervision program was limited by the fact that the office has only 11agents and i s inadequately equipped. To make it more efficient, it will have to be strengthened both technically (clearer definition o f tasks vis-a-vis the IGF, a procedures manual) and interms o f resources (required travel). General Inspectorate of Finance 5.6 The General Inspectorate o f Finance (IGF) is supposed to conduct ex post audits o f the state budget execution and to investigate instances o f bad management. This office goes into action at the request o f the Minister o f Finance. Inaddition to dealing with emergencies, the office also has an annual work program that i s submitted to the Cow des Comptes for information. In the past, it suffered from shortages o f material and human resources, but was recently strengthened with the recruitment o f additional staff. The results o f its audits and investigations are not published, and it is not known whether they are usedto make public expenditure more efficient. B. UPDATINGTHE PROCUREMENTCODE 5.7 The Central Procurement Commission and its subdivisions, the departmental procurement commissions, are important institutions for the evaluation and control o f procurement activities. The Procurement Code, adopted in 2002,@ laid the basis for a new regulatory framework more in keeping with accepted principles and requirements in the field o f procurement. The fact that the procurement thresholds have been raised has allowed the departmental commissions of the sectoral ministries to participate in procurement. However, there have been several shortcomings in implementing the 2002 code. Apart from the fact that the execution and ex ante control structures are overloaded and do not have sufficiently qualified and motivated staff, some regulatory provisions of the 2002 code are not consistent with the internationally recognized basic principles o f procurement. The limitations in the current legislation relate, among other things, to a lack o f clarity in the application o f the rules and the incompatibility of certain functions, inparticularthose o f regulation, contract award, control, and appeal. 5.8 As part o f administrative modernization, a review o f the current procurement code (CMP) has been undertaken inorder to enhance the country's absorptive capacity by making the procurement process swifter and more transparent. That review, which began in 2005 with an institutional grant6' from the World Bank, should correct the weaknesses highlighted inthe updated 2002 CPAR. The new government guidelines are designed to attract more interest in government procurement on the part o f private suppliers, both domestic and international, and to reduce the time and costs associated with the procurement o f goods and services by the Mauritanian government. The proposed reform program addresses three aspects: 0 establishing an institutional framework that will take into account the need to keep separate the different functions o f procurement, i.e. regulation, contract award, control, and recourse; the preparation and adoption o f regulations for application; and the publication and distribution o f standardbiddingdocuments and contract forms; 64Decree 2002-08 of February 12,2002 and Order R 0540 o fMay 15,2002 on thresholds. 65An IDFofUS$349,000 was provided inDecember2004 to assist the government. 50 0 bringing fiscal provisions governing procurement into line with internationally recognized practices for trade and imports; and finally; 0 strengthening capacities and structures through a permanent program o f training in public procurement, and support for the procurement regulatory function, including monitoring and evaluation o f the system's performance. 5.9 The proposed action plan calls for a complete overhaul o f the CMP and establishment o f a complete, consistent, and functional system to re-assert the principles of transparency and efficiency in the award o f public contracts, by June 2006. It stresses the need to reorganize the institutional framework inorder to promotethe five essential pillars o faproperly functioning procurementsystem: 0 establishment o f a regulatorybody; introduction o f an ex ante procurement control system separate from the contract award function, inorder to guarantee transparencyinthe allocationofcontracts; 0 provisions for legal appeal; 0 existence and application o f sanctions; and 0 systematic publication o fprocurement plans as well as bid evaluation proceedings. C. COMBATINGCORRUPTION 5.10 The Mauritanian authorities appear currently to attach great importance to the fight against corruption. The preparation and implementation o f an anticonuption strategy i s called for inthe report o f the Inter-ministerial Committee on Good Governance, and in the Poverty Reduction Strategy Paper. This determination i s also reflectedinthe establishment o f the Inspectorate General o f State (IGE).66 5.11 The IGEreports directly to the Prime Minister (PM). It has a general and permanent mandate for supervision, investigation and inspection, dealing among other things with the proper management o f public affairs and the evaluation o f public policies and programs. It was created following the coup d'Ctat o f August 2005, and has been given an important role inimplementingthe new authorities' commitments to good governance. It conducts its work in accordance with an annual program, and upon referral by the Prime Minister. Because it i s a new entity, it does not yet have sufficient human and material capacities to carry out its activities fully. The results o f its investigations are submitted to the Head o f Government (the PM). 5.12 The Central Directorate to Combat Economic and Financial Crime was created in July 2004. It falls under the Ministry o fthe Interior, Posts and Telecommunications. It i s involved, among other things, ininvestigating tax fraud andpublic corruption (it isnot involvedinpreventingmoney laundering). D. REINFORCING JUDICIALANDPARLIAMENTARY CONTROL 5.13 The Cow des Comptes (CdC), created in 1993, i s making gradual progress in accomplishing its tasks as defined by law. The provisional report onjudicial control for 2001 i s awaiting response from the Treasurer General, and a start has been made on the report for 2002. With respect to the Budget Execution Laws, those for 2001 and 2002 have been adopted by parliament. On the other hand, the Budget Execution Laws for 2003 and 2004 have posed great problems, inpart because o f missingdocumentation and the difficulty o f locating off-budget expenditures. To remedy this situation, a formula will have to be found, either in the form o f a new loi d'apurement (law closing books) or a simple fact-finding report. 66The IGEwas createdinSeptember 2005. 51 The lack o f a unit responsible primarily o f producing updated account balance sheets and preparing the necessary statements i s a source o f major delays. Concerningjudicial control and the Budget Execution Law, the CdC depends on prior input from the Treasury. For this reason, strengthening the Treasury must include a strengthening o f control. In addition, the publication in 2006 o f the full annual report will consolidate efforts to improve the quality o f controls. The CdC also plans to institute performance audits, including sanctions for faulty management. 5.14 Parliamentary control was the responsibility o f the two chambers, which were dissolved after August 3, 2005. Untilthat time, however, the NationalAssembly hadbeenreceivingtechnical assistance, since September 2004, from the GTZ, including training in public finance issues. The training cycle, which had started with training in budget preparation, was interrupted by the coup d'Ctat. The report on the Budget Execution Law was subjected to in-depth debate in the Finance Committee and the Audit Court, resulting in a long list o f recommendations to the government. This was followed by extensive debate inthe plenary session. E. MAINPROBLEMSRELATED TOAUDITS 5.15 Audit results are not published inMauritania, regardless o f the entity producing them. Moreover, the work o f the Auditor General i s rarely subjected to external quality control. There has as yet been no resort to private sector auditors, who might be able to verify government transactions on behalf o f the Auditor General. However, externally funded projects, including those financed from World Bank loans, are audited by private sector auditors. 5.16 The audits have revealed anomalies and irregularities o f various kinds. Problems relate most often to the quality o f documentation, failure to follow proper competition procedures, splitting o f contracts, and timing inconsistencies inthe expenditure stages. Some irregularities are caused by unfamiliarity with the laws and lack o f training. Others arise from violations o f regulations, which disrupt the proper functioning o f public finances. 5.17 For example, the absence o f legally required identificati01-1~~on documents submitted by suppliers i s one o f the inadequacies noted. This practice has two direct consequences for the public purse: it facilitates tax evasion, as it supports unidentified suppliers over those who fulfill their tax obligations; it frustrates competition, since without identification it is difficult to determine inter-business relationships inorder to ensure that there i s real competition amongbidders. 5.18 In fact, the main problem encountered during the audits is the absence o f substantiating documentation on public expenditures. This shortcoming in terms o f record-keeping makes it difficult to prove competition during procurement procedures, or to substantiate expenditures such as salaries, bonuses, or scholarships. This problem goes beyond mere record keeping, however, and will have to be addressed in detail. If it persists, the government will be incapable o f rendering accounts, and this will impede achievement o f its commitments to good governance, and to combating corruption inparticular. F. MONITORING EVALUATION PUBLICEXPENDITURES AND OF 5.19 Regardless o f the sector, the expenditure trackmg and evaluation system i s defective in Mauritania. The rules do not require the MFto undertake a systematic evaluation o f a given percentage o f spending for programs and projects. Moreover, the results of ad hoc evaluations are o f little use in 67 Address, businessregistrationnumber, tax number. 52 budgetary decisions, and are not sufficiently disseminated. There are no surveys o f users and clients inthe various ministries to evaluate their degree o f satisfaction with the services offered. Nor are any investigations such as expenditure tracking surveys conducted to determine possible bottlenecks. The first survey o f this lund i s now being prepared in the health sector. Current practices in the system for allocating public resources and taking budget decisions provide little incentive for the sectoral departments to evaluate the policies implemented. 5.20 A committee was established in 2001 to monitor the programming and execution o f HIPC funding. That committee includes the MF, the MAED and several sectoral departments. It produced quarterly and annual reports on budget execution over the period 2001-2004, intended primarily for donors. This certainly marked a beginning, but the tracking and evaluation o fpublic spending needs to be broadened to other sources o f financing and to all policies. Moreover, the outcomes o f these evaluations should be made more widely known. 5.21 With respect to the traclung and evaluation o f external financing handledbythe MAED,there are number o f problems. The MAED does not publish comprehensive information on donor support or any evaluation o f the use o f these resources for policy purposes. We notice a lack o f information regarding actual disbursements which means that financial monitoring i s exclusively based on disbursement requests sent to donors. These requests can be denied or delayed by moths while exchange rate fluctuations continue affecting the accuracy o f the financial data. This i s one o f the factors impacting negatively the planning of investment spending. Additional problems are the unavailability o f data regardingphysicalmonitoring and the lack o f efficient monitoring tools. 5.22 The Mauritanian government, and inparticular the MF and the MAED, do not have the capacity to conduct a rigorous examination o f budget proposals based on cost effectiveness and cost-benefits analysis. This fact explains the problems found at the budgeting stage. G. CONCLUSIONS AND RECOMMENDATIONS 5.23 The strengthening o f control and supervisory institutions i s a priority on the government's agenda for improving the effectiveness and transparency o f the management o f public assets. Therefore, it i s recommended to continue the following measure: . it would be desirable for the various supervisory institutionsto work together inestablishing their . . annual programs, so as to avoid both duplication o f effort and unnecessary inconvenience to the institutionsthey supervise. greater coordinationwould also allow for better coverage o f supervision. given the resource constraint, there i s an urgent need to look closely at the main objectives o f supervision and the expected results. In some cases, amendments to legislation and regulations and revisions to procedures could diminish significantly the possibilities for embezzlement or corruption. 5.24 When it comes to prosecuting crimes detected by the supervisory institutions, it would be useful: (i) to create a division within the Public Prosecutor's office to deal with "economic and financial crimes"; and (ii) as for parliament, to reinforce external oversight by creating new channels between parliament and the CdC.68Submission o f a mid-term budget execution report by the MF to parliament would also be useful to the supervisory institutions and would contribute to transparency. 68Through legislation, debate on the comprehensive annual report inthe presence o f the President o f the CdC, the possibility of having parliament refer matters to the CdC, etc. 53 5.25 With respect to the procurement system, if the reform program is to succeed to promote competition, to prevent corruption, and to foster transparency and accountability in management o f the public purse, it will have: (i) to involve all stakeholders at the various stages o frevising the code to ensure that the procurement offices will feel ownership over it; and (ii) the rules to be established should be enforced impartially so as to help create a stable and well-regulated environment. 5.26 The audit and control system will only function efficiently if the public administration makes the considerable efforts needed to improve the record keeping system. This will require training for staff in charge o f the supporting documentation that must accompany any public expenditure at each stage o f the budget circuit. Respect for the basic rules i s fundamental inthis area. 5.27 The monitoring and evaluation capacities o f the MF and the MAED must be strengthened so that the various officials involved can determine the extent to which the resources employed have achieved the expected objectives. An effective trackmg and evaluation system, involving all stakeholders, could provide important indications when it comes to deciding whether to expand the scope o f a program, to modify it, or to terminate it. The monitoring o f HIPC expenditures should be extended to all poverty reduction spending, and shouldinclude an evaluation o f its impact. 54 ANNEXES 55 Annex 1:Selected Macro-Economic Indicators (2001-2006) Table 1. Mauritania: Selected Economicand FinancialIndicators.2001-06 2001 2002 2003 2004 2005 2006 Revised Est. Proj. (Percentagechanges; unless otherwise indicated) NationalIncome and prices GDP at constantprices 1/ 2.9 1.1 5.6 5.2 5.4 19.3 Non-oil GDP at constant prices 2.9 1.1 5.6 5.2 5.4 6.7 GDP deflator 1/ 7.9 7.8 2.5 11.5 18.9 44.4 Non-oil GDP deflator 7.9 7.8 2.5 11.5 16.9 8.8 Consumerprice index (periodaverage) 7.7 5.4 5.3 10.4 12.1 6.5 Consumerprice index (end of period) 4.1 8.4 2.9 16.1 5.8 8.1 External sector Exports of goods, f.0.b. (percentage change in US. dollars) 1.6 -7.7 -4.4 38.0 39.6 213.2 Of which: non-oil 1.6 -7.7 -4.4 38.0 39.6 27.9 Importsof goods, f.0.b. (percentage change in US. dollars) 2/ 2.4 -7.3 25.7 70.3 50.3 -12.9 Importsof goods, f.0.b. (percentage change in U.S.dollars) -7.0 -7.6 22.1 33.7 14.7 17.4 Officialtransfers(in percentof GDP) 7.5 6.1 7.2 4.1 3.8 2.5 Currentaccount balance(in percentof GDP) 2/ -11.o 3.7 -13.6 -35.0 -50.0 3.5 Currentaccount balance(in percentof GDP) -6.0 8.3 -7.0 -12.4 -7.9 -9.1 Overall balance(in percentof GDP) -6.9 -1.8 -9.5 -7.3 -4.0 4.8 Official reserves Gross official reserves (in millionsof US$,end-period)3/ 40 74 32 39 70 152 In monthsof followingyear's imports of goods and services 2/ 0.9 1.4 0.4 0.5 0.7 1.4 Money and credit Moneyand quasi-money 34.3 23.8 25.5 13.5 14.6 15.8 Currencyin circulation 17.6 17.7 47.5 -3.4 14.8 12.8 (In percentof non-oil GDP) Consolidated government operations Revenueand grants 24.7 34.8 35.4 32.9 26.4 43.0 Revenueand grants (excludingoil) 24.7 34.8 35.4 32.9 26.4 26.1 Idem,excludinggrants 20.7 30.4 30.7 29.7 24.4 23.6 Oil revenue4/ ... ... ... ... ... 17.0 Expenditureand net lending 35.0 37.6 47.2 37.7 33.5 32.6 Overall balanceincludinggrants -10.3 -2.9 -11.8 -4.8 -7.0 10.5 Overall non-oilbalanceexcludinggrants -14.3 -7.2 -16.4 -8.0 -9.1 -8.9 Overall non-oil balanceincludinggrants -10.3 -2.9 -11.8 -4.8 -7.0 -6.5 Memorandum items: OuguiyalUS5exchange rate (endof period) 264.1 268.7 265.6 256.2 268.6 ... Exports, f.0.b.(in millionsof U.S. dollars) 355 328 314 433 604 1,892 Imports. f.0.b. (in millionsof US. dollars)21 415 383 468 625 718 843 NominalGDP (in billionsof ouguiya) 287 312 338 397 497 857 Nominal non-oil GDP (in billionsof ouguiya) ... ... ... ... ... 577 NominalGDP (in millionsof US. dollars) 1,122 1,150 1,285 1,495 1,871 3,189 Population(in millions) 2.57 2.63 2.69 2.76 2.82 2.89 GDP per capita (in US. dollars) 437 437 477 542 663 1,103 REER (12-monthpercentagechange; end of period) 5/ 4.5 -9.3 -13.4 8.1 8.7 0.3 Sources:Mauritanianauthorities;and Fund staff estimatesand projections. I /The large increase in the GDP deflatorin 2006 (andthe relativelylow increasein real GDP for this year) reflectthe authorities'choice of 1998 as the baseyear for the calculationof nationalaccountsat constant prices. Oil priceswere at a low in 1998. 21 Excludingoil explorationlproductionand other mining (copper,gold)-relatedactivities. 31Excludingoil account. 4/ Includingoil signaturebonuses. 51 In 2006: estimates through end-May. 56 Figure A1.2: RealExchangeRate (1999-2005) 120 110 100 90 80 70 I 60 50 40 ~ I t-NEER--otREER Source: IMF 57 Annex 2: Fiscal Revenues inMauritania Important fiscal reforms have been introduced among the various actions undertaken since 1995. Several o f those reforms aim at reducing tax rates and simplifying tax procedures in order to establish a favourable environment for business activity and investment. The following table presents an evolutionary comparisono frates beforethe reforms o f 1995 and the current situation. Natureof the tax Beforethe 1995 I Situationin 2006 trial and commercial Source: MF Direct fiscal reform is responsible for 3 objectives: simplification, growth and fairness o f the tax system. Itplays a critical role inensuringmacro-economic stability, correctingincome inequalities andpromoting the development and diversification o fthe private sector. Started in 1995, the reform continued duringthe years 2001-2004 with the following initiatives: Gradual reduction o f the tax on industrial and commercial profits (BIC): the tax rate has gone down from 40% to 35% in2001,25% in2002 and 20% in200 before going back to 25% in2004; Full deduction on minimumpresumptive tax (IMF) imposed on BIC: This deduction went from 25% to 50% in2000, then 75% in2001 and 100%in2002; Extensiono f deficit report periodto 5 years instead of 3; Introduction o f option o f regressive depreciation: This enables companies to reduce or increase their cash flow by deducting all deficits incurred inthe first years o f operation; Standardization o f value-added tax (TVA) rates at 14%(2001), with a 0% rate on some products; Implementation o f reimbursement processes for TVA on investment; Improvement o f services to large taxpayers with the set-up o f the large taxpayer unit (Direction des Grandes Enterprises); Reformo f the tax on wages and salaries (ITS)69: The scale was simplified (fi-om 11to 2) and the rate reduced (from 40% to 15%) with an exemption o f amounts under 10,000 UM; Simplification o f the general income tax (IGR) with brackets reduced from 11 to 6 and the maximum applicable rate lowered from 40% to 39%. 69In2006, reforms areplannedto bringback the maximumrateto 35% with anexemptionupto UM21.OOO UM. 58 The reformo f external finances went on withthe implementation o f the followingprinciples: Reduction o f custom rightson equipments to a common rate o f 5%; 0 Reduction o f holding time on merchandise in optional storage and the limitation o f the type of product allowed to be held; The applicationo f a "regime special de remise de droits" (drawback) ; 0 Improvement o f evaluation practices by the inclusion inthe finance law o f 2002 o f new measures on customs evaluation (in June 2002, the customs administration adopted legal measures in line with GATT andnow use the transactional value as the evaluation tool for importedproducts); 0 Set up o f a reimbursement process for exporting companies. Despite the implementation o f the listed improvements, public finances continue to endure major challenges: a permanent weakness in terms o f information on contributors, existence o f sectors that are not being taxed, a lack o f monitoring o f non-compliant companies, low collections fi-om the real estate tax, a lack of fully computerized systems at the port and at the Customs head office, insufficient means to fight fraud, etc.. 59 Annex 3: Evolutionof Poverty and Social Indicators 1990 1996 2000 2004 Poverty Incidenceo f poverty at national level 56.6% 54.3% 51.O% 46.7% Incidence o f extreme poverty at national level 34.1% 27.9% Poverty depth 19.3% 15.3% Giniindex 34.0% 39.0% 39.3% Incidenceo f poverty inrural area 66.1% 59.0% Education Gross primary enrollment rate 46.0% 82.0% 87.0% 95.1% Gross primary enrollment rate for girls 78.8% Proportion o fpupils starting grade 1who finishgrade 5 55.0% 48.8% Proportion of girls starting grade 1who finish grade 5 51.0% 44.0% Adult literacy rate 42.8% 42.5% Health Life expectancy at birth 54 57 Infant mortality rate (per thousand) 87 106 Infant -chdd mortality rate (per thousand) 137 122 135 123 Proportion o fbirths attended by qualified health personnel 56.9% Maternal mortality rate (per thousand) 747 HIV/AIDSprevalence rate 0.5% Proportion ofpopulation living within 5 kilometers ofhealth services 67.0% 73.0% Malnutritionrate 47.6% 32.0% 30.2% Water, sanitation and electricity Proportionofpopulation with access to a potable water source 60.4% 63.7% Rate o f water supply through an indoor tap 35.0% 45.0% Proportion ofpopulation with access to improved sanitation 48.0% 49.0% Proportionof population with access to electricity 18.0% 23.8% Sources :ONS 60 Annex EPCV Dakhle Source: ONS 61 Annex 4: EvolutionofPublic Expenditures Public expenditures according to the economic classification and modality of execution (as a % of GDP) 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 Total expenditures 24.2 24.6 23.0 23.9 25.1 30.7 35.0 37.6 47.2 37.7 Current expenditure 15.9 16.1 16.9 16.3 16.8 20.5 20.9 22.4 31.1 24.4 Compensation of employees 4.6 4.4 4.9 4.4 4.2 4.6 4.5 5.1 4.7 4.3 Budget 4.0 3.9 4.1 3.7 3.3 4.0 3.9 4.1 3.9 3.5 Offbudget 0.6 0.5 0.8 0.7 0.9 0.6 0.5 1.o 0.9 0.8 LDA 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.1 0.1 Change 0.6 0.5 0.8 0.6 0.9 0.6 0.5 0.9 0.8 0.7 Good and services 4.6 4.6 4.9 5.4 5.4 6.9 7.0 8.9 10.4 9.3 Budget 4.1 4.0 4.1 4.4 4.2 4.7 4.8 6.7 6.8 6.2 Offbudget 0.5 0.6 0.8 1.1 1.2 2.2 1.8 1.5 3.2 2.5 LDA 0.0 0.1 0.2 0.2 0.4 0.7 0.3 0.5 1.4 1.5 Emergencyplan 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.5 0.0 BCM 0.0 0.1 0.1 0.3 0.2 0.7 0.6 0.4 0.1 0.0 Change 0.3 0.2 0.4 0.3 0.4 0.5 0.7 0.3 0.8 0.7 Deficit onpensionaccount 0.2 0.2 0.2 0.2 0.2 0.3 0.3 0.3 0.3 0.3 Arrears (PU+Ent. Publ.) 0.0 0.0 0.0 0.0 0.0 0.0 0.4 0.8 0.5 0.6 Subsidies 1.5 1.5 1.6 1.6 1.7 2.3 2.1 2.5 7.8 2.4 Budget 1.2 1.2 1.2 1.2 1.2 1.4 1.3 1.5 1.5 1.4 Offbudget 0.2 0.3 0.4 0.4 0.5 0.8 0.8 0.8 5.6 0.8 LDA 0.0 0.1 0.1 0.1 0.2 0.4 0.2 0.3 1.9 0.2 Emergencyplan 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 3.3 0.3 BCM 0.0 0.0 0.0 0.0 0.0 0.0 0.3 0.3 0.0 0.0 Change 0.2 0.2 0.3 0.2 0.3 0.4 0.3 0.1 0.3 0.3 Arrears (PU+Ent. Publ.) 0.0 0.0 0.0 0.0 0.0 0.2 0.0 0.2 0.7 0.2 Military expenditure 2.3 2.6 3.O 2.1 2.7 3.5 4.6 3.2 4.9 4.7 Budget 2.0 1.9 1.8 1.7 1.6 1.6 1.5 1.5 1.2 1.o Off budget 0.3 0.7 1.2 0.4 1.o 1.9 3.1 1.6 3.6 3.7 LDA 0.0 0.2 0.2 0.2 0.3 0.7 1.7 1.o 2.8 3.O BCM 0.2 0.5 0.9 0.1 0.6 1.1 1.3 0.5 0.7 0.6 Change 0.1 0.1 0.1 0.1 0.1 0.2 0.1 0.1 0.2 0.1 Interests 2.1 2.2 2.1 2.5 2.6 3.0 2.7 2.5 2.7 3.O External 1.7 1.8 1.8 1.9 1.9 2.2 1.9 1.9 2.0 1.8 Domestic 0.4 0.4 0.3 0.6 0.6 0.8 0.8 0.6 0.8 1.2 Hors budget 0.4 0.4 0.3 0.6 0.6 0.8 0.8 0.6 0.8 1.2 Intereston banking system 0.1 0.1 0.1 0.1 0.2 0.4 0.4 0.2 0.5 0.9 Intereston treasurybills 0.3 0.3 0.3 0.5 0.4 0.4 0.4 0.3 0.3 0.3 Special account 0.9 0.7 0.4 0.3 0.2 0.2 0.0 0.2 0.5 0.6 Capital expenditures and net lending 5.5 4.3 4.5 5.3 5.6 8.4 10.6 13.3 13.4 10.9 Capital expenditures 5.5 4.3 4.5 5.3 5.5 7.9 10.4 13.2 13.0 10.9 Budget 0.9 1.2 1.8 1.8 1.5 3.7 6.4 7.8 7.5 6.2 Offbudget 4.5 3.1 2.7 3.5 4.0 4.3 4.0 5.4 5.5 4.7 Net Lending 0.0 0.0 0.0 0.0 0.2 0.5 0.2 0.1 0.0 0.0 Other expenditures 2.9 4.2 1.6 2.3 2.7 1.8 3.5 2.0 2.7 2.4 Overall balance excluding grant -7.3 -2.2 -2.6 -4.7 -5.7 -8.9 -14.4 -7.2 -16.4 -8.0 Source : MF 62 Table A4.1: Number of Civil Servants by Ministry and Service (1998-2003) Ministries and Services 1998 1999 2000 2001 2002 2003 Presidency Office 223 219 216 213 209 207 Prime Minister & General Office 94 91 88 86 83 79 Interior Ministry and Telecommunications 2503 2457 2441 2503 2550 2517 Justice Ministry 869 846 844 840 854 865 Economics and Development Ministry. (MAED) 144 141 141 140 138 160 Ministryo fFinance(MF) 1270 1230 1203 1164 1142 1116 Commerce & Tourism Ministry 155 154 152 152 150 148 Fishery and Maritime Affairs Ministry 113 112 112 110 109 109 Equipment & Transport Ministry 143 140 149 143 138 126 Rural Development and Environment Ministry 712 689 695 688 678 651 Hydraulics and Energy Ministry 62 61 60 56 59 56 Culture and Islamic Studies Ministry 93 88 87 84 80 74 National Education Ministry(MEN) 12375 13154 14176 15808 17599 19312 Health and Social Affairs Ministry(MSAS) 3466 3594 3657 3847 3984 4265 ForeignAffairs Ministry 88 85 81 75 72 66 Communication& Relations with Parliament Ministry 79 78 77 76 75 71 Mine and Industry Ministry 66 65 65 65 63 61 Public Service, Youth and Sport Ministry 390 384 377 370 361 353 C o w des comptes 68 67 67 65 64 63 Secretary of State for Feminine Condition 97 96 90 85 82 113 NorthAfrican Affairs Secretary o f State 13 , 12 12 12 12 12 Secretary o f State for Literacy 181 178 175 175 170 165 Secrttariat d'Etat a l'Etat civil 41 41 41 41 41 39 ForeignDelegation/ 22 22 22 22 22 22 Total 23267 24004 25028 26820 28735 30650 Source: GRIM (2004~) 63 Table A4.2: EconomicClassificationof Off-BudgetExpenditure(in UMBillion) 1995 19% 19m 1998 1999m 2001 m 2003 2004 Totalo w 9.0 13.5 11.4 127 18.2 23.1 39.5 35.9 69.5 . 50.6 - m o f m w 1.1 1.0 1.7 1.5 2.2 1.6 1.5 3.1 3.0 3.1 InpKaiQgOftctalconpasafioneqdim 13.3 11.6 161 14.9 21.0 13.4 120 19.5 18.4 18.1 Gx&wd&cs 0.9 1.3 1.7 25 3.0 5.6 63 7.0 124 12.3 Inpxerbigoftcralgadsandsmksexpdituts 10.7 13.6 164 19.6 22.1 31.4 31.4 25.0 35.2 333 Shi&eSandTll3lKk& 0.5 0.9 0.9 1.3 2 4 23 3.1 21.3 4.1 In~oftciall&eswdandtl.insfatseqdibl 15.4 18.2 25.8 25.6 30.5 39.4 39.5 39.3 80.8 42.7 MlitayKperdltuts 0.5 1.5 25 1.0 2.6 4.9 8.8 5.0 122 14.8 I n ~ o f t o t a l n i l i t a y B q H d m 122 28.1 40.4 20.0 39.2 53.7 667 51.1 74.5 79.2 wmt 0.7 0.7 0.7 1.4 1.6 2 1 22 1.7 2 6 4.7 Inparertap:oftctalirterest 17.7 161 15.0 23.7 25.2 27.1 28.7 225 28.5 39.1 ChFitaleqmxhm 0.1 0.2 0.5 0.2 0.2 0.7 7.7 9.4 7.5 1.8 In- ofcapitaleqdim 4.3 7.4 129 5.0 6.4 7.2 422 38.6 29.4 7.4 Mldbgwd&lWh3lj m 0.1 0.2 1.4 0.2 I n ~ O f t o t a l ~ l ~ w d . r e s h U aou a nt i - - 100.0 100.0 100.0 100.0 ctkTBcpaditlw 5.3 8.4 3.5 5.3 7.2 5.9 10.5 64 9.2 9.6 S0ure:MF 64 I_ u s fei W .m 0 a8 L CE] .B E a, & GE a B E a, c I 58 3 B 0 U 0 8 E L 3 Ba 9 E 4- 0 cd E 2 @3 3 P 1 ru 0 ru 0 m cu 0 References Eifert, B., Gelb, A., & Tallroth, Nils T., 2004, The Political Economy o f Fiscal Policy and Economic Management in Oil-Exporting Countries, Documents de recherche sur les politiques 2899, (World Bank Washington, DC). GRIM, 2006, Presentation o f Budget according to economic and functional classification (1992-204), Direction ofbudget and accounts (Nouakchott, la Mauritania). GRIM, 2006, Report of interministerial committee in charge o f good governance (Nouakchott, Mauritania). GRIM, 2004a, Public EpendituresReview 1992-2002, MAEDand CMAP (Nouakchott, Mauritania). GRIM, 2004b, Third Progress Report of the Poverty Reduction Strategy Paper (PRSP) (Nouakchott, Mauritania). GRIM, 2003, MediumTerm Expenditures Framework (MTEF), Ministryo fFinance, (Nouakchott, Mauritania). IMF (International Monetary Fund), 2003, Mauritania - Statiscal annex country report No03/315, IMF (Washington, DC). IMF,2003, VersunerCformedeI'imposition directe, FAD,PloycopiC(WashingtonDC). IMF, 2001, Une politique fiscale pour les pays en dCveloppement, Dossiers Cconomiques No 27, Vito Tanzi & HowellZee, (Washington DC). IMF, 1995, Unproductive Public Expenditures : A Pragmatic Approach to Policy Analysis, DCpartement des finances publiques, Pamphlet Series No 48 (Washington DC). IMF & GRIM, 2002, Rapport sur l'observation des normes et des codes: Module transparence des finances publiques, IMF, Washington, DC. MEN, 2001, National development program for education (Programme national de dkveloppement du secteur tducatif) 2001/2010, (Nouakchott, Mauritania). MSAS (Ministry of Health-Ministbre de la SantC et des Affaires Sociales), 2002, SantC et Pauvrett en Mauritania :Analyse et Cadre stratkgique de lutte contre la pauvretC (Nouakchott, Mauritania). MSAS, 1997, PlanDirecteur de la SantC et des Affaires Socialespour la pCriode 1998-2002, Ministere de la SantC et des Affaires Sociales ;1997 OED (DCpartement d'kvaluation des ophations), 2004, Rapport de rCCvaluation des rCsultats A long terne - RCpublique islamique de Mauritania -Programme d'ajustement du secteur des Entreprises publiques (1990-1994) CrCdit 2166-MAU7World Bank (Washington, DC). ONS (Office national de la statistique), 2000, Mauritania : EnquCte dkmographique et sanitaire, (Nouakchott, Mauritania). 69 PEFA (Public expendituresand financial accountability), 2005, Cadre de mesures de a performance de la gestion des finances publiques, SecrCtariatPEFA, World Bank (Washington, DC). PEFA (Public expenditures and financial accountability), 2003, Aperip des nonnes internationales en matikre de comptabilitC, d'audit et de contrble interne dans le secteur public, SecrCtariat PEFA, World Bank (Washington, DC). Pradhan, Sanjay, (1996) Evaluating Public Spending : A Framework for Public Expenditure Reviews, World BankDiscussionDocument 323, World Bank (Washington, DC). UNDP (UnitedNations Development Programme), 2005, Report on Progress of MilleniumDevelopment Goals Achievement (Nouakchott, Mauritania). World Bank, 2006, Mauritania-Country Assistance Strategy CompletionReport ;(WashingtonDC). World Bank, 2004, Public expenditures review, Focusing public expenditure on growth and poverty reduction, Africa Region(Washington, D C -Green cover). World Bank et IMF, 2004, Public expenditure management in Mauritania: Evaluation and action plan, AFTP4 et MDE, World Bank et IMF (Washington, D C -nonpublic). World Bank, 2004, Tunisia, Evaluation on public finances management, MENAregion, Washington DC. World Bank, 2003, Mauritania: Accelerate growth simulating by the private sector - CEM-; PREM 4; Afkca Region, WorldBank,7 November,2003. World Bank, 2003, Morocco, Evaluation on public finances management, MENA region, Washington DC. World Bank and GRIM, 2003, Country Financial Accountability Assessment (CFAA), Vol I.and II., WashingtonDC. World Bank, 2002, Mauritania-Country Assistance Strategy ;(Washington DC). World Bank, 2002a, Mauritania - Aide-mCmoire & Recommendation of the President Regarding HlPC Initiative, (Washington DC). World Bank, 2002b, Country Procurement Assessment Review(CPAR), Volumes I. Africa & II., Region (Washington, DC). World Bank, 2001, Turkey, Public Eexpenditure and Institutional Review : Reforming Budgetary Institutions for Effective Government, Report No 22530-TU, Europe and Central Asia Region (Washington, DC). World Bank and IMF, 2001b, monitoring poverty reduction expenditures within HIPC framework Evaluation and actionplan, FADet AFMMR, World Bank and IMF, (Washington, DC). World Bank, 1998, Public Expenditure Management Handbook, PREM, World Bank, (Washington, DC). World Bank, 1994, Mauritania: Public Expenditure Review, Report No10973, World Bank, (Washington, DC). 70