Report No. 35871-GW Guinea-Bissau Public Expenditure Review (PER) Update Enhancing Growth and Fiscal Adjustment Through Civil Service Reform May 23, 2007 PREM 4 Africa Region Document of the World Bank PIP Public InvestmentProgram PER Public ExpenditureReview PPG Public andPublicly Guarantied PRGF PovertyReductionandGrowthFacility PROST PensionReformOption SimulationToolkit PRSP PovertyReductionStrategyPaper REER RealEffective ExchangeRate SMP Staff MonitoredProgram SSR SecuritySector Reform TC Treasury Committee TCTC TechnicalCommittee oftheTreasuryCommittee TEC Tarifa ExternaComum(Common ExternalTariff) TFP Total FactorProductivity TOFE FinancialOperationofthe Government TOT Terms ofTrade TVA Value AddedTax UNDP UnitedNation Development Fund USD UnitedStatesDollars WAEMU West African EconomicandMonetaryUnion Vice President: GobindT. Nankani CountryDirector: MadaniM.Tall SectorManager: AntonellaBassani Task Team Leader: Boubacar-SidBarry ... 111 Table of Contents EXECUTIVE SUMMARY ............................................................................................................................... IX 1 2.. INTRODUCTION ...................................................................................................................................... 1 MACROECONOMIC MANAGEMENT AND PERFORMANCESINCE THE PREVIOUSPER: A SITUATION STILL DIFFICULT .................................................................................................................. 2 ECONOMIC POLICY MANAGEMENT THE PREVIOUS PER SINCE ECONOMIC ...................................................................................... .......................................................................... 2 4 FISCALPOLICYMANAGEMENT THEPREVIOUSPER................................................................................ PERFORMANCE SINCE THE PREVIOUS PER SINCE ..;.. ....................................... 7 7 Fiscalpolicy management has improved since 2004.................................................................................... Governmentfiscal situation was in a dire condition in 2003 ........................... 8 Thereare still majorfiscal policy issues that remain unsolved.................................................................... 9 FISCALPERFORMANCE OUTCOMESINCETHEPREVIOUSPER AND .................................................................. 14 General Context.......................................................... .14 Revenue Performance ................................................. .................................................... Economic Analysis of Public Expenditure: Large WageBill and Low Pay . .15 18 20 Sources of Budget Deficit Financing................................................................................................. Sectoral Analysis of Public Expenditure: Still Dealing with Security Issues.............................................. MACROECONOMICPOLICIES IN2006AND MEDIUM-TERMPROSPECTS ................................................... Government Policiesfor 2006: Pursuing Macroeconomic and Structuring Reforms ................................ 22 Medium-term Prospects: Strengthening Governance and Enhancing Fiscal Sustainability and Growth by 2009............................................................................................................................................................ 26 CONCLUSIONOF POLICY RECOMMENDATIONS ................................................................................................. 3. IMPROVING THE FISCALSITUATION THROUGHDOWNSIZING THE CIVIL SERVICE 30 ..27 BRIEFDIAGNOSTIC OFTHECIVIL SERVICE ...................................................................................................... 30 RATIONALE AND JUSTIFICATION FORCOMPENSATIONPACKAGES ................................................................... 35 ESTIMATING THE 37 Social Cost/Benefit........ THECOSTBENEFITOFDOWNSIZING CIVIL SERVICE............................................................ ........................................................................... 37 Short-Term Fiscal impa ......................................................................... Long-Term Economic Impact OfDownsizing ...................................................................... ASSESSING POTENTIALPOVERTY IMPACTOFRETRENCHMENT WITHOUT SEVERANCE BENEFITS THE ............45 CONCLUSIONAND POLICY CONSIDERATIONS ................................................................................................... 46 4. DEBT SUSTAINABILITYANALYSIS ................................................................................................. 47 PUBLIC DEBT SITUATIONAS OF END-2005 ........................................................................................................ 47 MACROECONOMICFRAMEWORK OF THE DSA 48 DEBT SUSTAINABILITY ...................................................................................................................................... .................................................................................................. 49 CONCLUSIONAND STAFFSAPPRAISAL .............................................................................................................. 50 5. CONCLUSIONTO THE PER ................................................................................................................ 56 MAINMESSAGES PER .......................................................................................................................... THEWAY FORW ......................................................................................................................................... 56 OF THE 56 ARD REFERENCES ................................................................................................................................................... 58 ANNEXES ........................................................................................................................................................... 60 iv ListofBoxes: BOX 2.1:QUANTITATIVEAND STRUCTURALINDICATORSUNDERTHE SMP IN2005 ...................... 3 BOX 2.2: TASK AND STRUCTUREOFTHE TREASURY COMMITTEE 8 BOX 2.3: THEBUDGETCYCLE BOX 2.4: ANALYSIS OF REVENUEPRODUCTIVITYDURING 1980-2005............................................... ...................................................................................................................... ...................................................... 13 17 BOX 2.5: QUANTITATIVEAND STRUCTURALINDICATORSUNDERTHE SMP IN2006 .................... BOX 3.1: ADMINISTRATIVE CATEGORIESINTHE C M L SERVICE....................................................... 23 32 BOX 3.2: WAGE INEQUALITY:THE GIN1COEFFICIENTAND THE LORENZCURVE ......................... 34 BOX 3.3: SUMMARYFINDINGSOF THE COMPENSATIONBENEFITSTUDYBASEDON 1997DATA .................................................................................................................................................................... 39 BOX 3.4: METHODOLOGYOF THE DOSEMODEL ..................................................................................... 40 BOX 3.5 :LONG-TERMFINANCIALRETURNOF DOWNSIZING .............................................................. 44 BOX 4.1:MACROECONOMICASSUMPTIONSUNDERTHEBASELINESCENARIO ........................... -50 Listof Tables: TABLE 1.1:PRIORITYREFORMAGENDA .SEQUENCING. FEASIBILITY. AND IMPACT ONFISCAL TABLE 2.1: SELECTEDECONOMICAND FINANCIAL INDICATORS,2003-2006..................................... XIV PERFORMANCE .................................................................................................................................... 4 TABLE 2.2: DNTPAYMENTS,2004-05(INMIL .OF CFAF) 10 TABLE 2.3: ECONOMICCOMPOSITIONOF SPENDING ............................................................................. ......................................................................... 18 TABLE 2.4: COMPARISONOF THE GROSSCAPITAL FORMATIONWITHINWAEMU(IN% OF GDP) .................................................................................................................................................................... 19 TABLE 2.5: FINANCINGOF THEPIP, 2003-04 .............................................................................................. 19 TABLE 2.6: GOVERNMENTSPENDINGINSOCIALSECTORS (PERCENTAGEOF GDP AVERAGED OVER 1996-2004) ...................................................................................................................................... 20 TABLE 2.7: TREASURY CASH-FLOW PLANFOR2006(INMILLIONS OF CFA FRANCS) ................................................................................. .................... 24 TABLE 3.1:STRUCTUREOF EMPLOYMENTIN2002 31 TABLE 3.2: CIVIL SERVANTS BY CATEGORYAND WAGE ..................................................................... 33 TABLE 3.3: INEQUALITY INPUBLICWAGES, 1997AND 2005 ................................................................. 33 TABLE 3.4: PERFORMANCEOF ALTERNATIVE PACKAGES-DIRECT AND INDIRECT ................................................................................................................................ TABLE 3.5: ESTIMATESOF TOTAL SEVERANCECOSTFORTHE GOVERNMENTy2005 ................... APPROACHES, 2005 42 TABLE 3.6: POPULATION-BASEDPOVERTY MEASURESBY REGIONAND AREAS, 2002 ................42 45 TABLE 3.7: ESTIMATESOF POVERTYAMONG CIVIL SERVANTS ASSUMINGNO SEVERANCE PACKAGE ................................................................................................................................................. 46 TABLE 4.1: SUMMARYOFBASELINEDEBTSUSTAINABILITY INDICATORS(*) .............................. 50 TABLE 4.2: EXTERNALDEBT SUSTAINABILITY FRAMEWORK, BASELINESCENARIO, 2006-2026 (INPERCENTOFGDP,UNLESSOTHERWISEINDICATED) 1/ ........................................................ 51 TABLE 4.3: SENSITIVITYANALYSESFORKEY INDICATORSOF PUBLICAND PUBLICLY GUARANTEEDEXTERNALDEBT, 2006-26(INPERCENT) .............................................................. 52 TABLE 4.4: COUNTRY: PUBLIC SECTORDEBT SUSTAINABILITY FRAMEWORK,BASELINE SCENARIOy2003-2026(INPERCENTOF GDP, UNLESSOTHERWISEINDICATED) .................... 53 Listof Fimres: FIGURE2.1: KEY INDICATORSOF ECONOMICHEALTH. 2001-05( INANNUAL %CHANGE) FIGURE2.2: REAL GDP AND CASHEWPRODUCTIONVOLUMES. 1999-03(ANNUAL % CHANGE) 5 ............5 FIGURE2.3: GOVERNMENTDEFICITAND INFLATION. 2001-05.............................................................. FIGURE2.4: CHANGESINTT AND REER. 2003-05 (ANNUAL %)............................................................... 5 6 FIGURE2.5: CURRENT SPENDING. LEVELAND TREND. 2001-05 ............................................................ 6 FIGURE2.6: BUDGETAND EXTERNALBALANCES,2001-05 (IN% OF GDP) ....................................... 14 V FIGURE2.7: TOTAL REVENUEIN 2005 (INBILLIONCFA) ...................................................................... 14 FIGURE2.8: MONTHLY CONSUMPTIONTAXES, 2003 & 2004 (INBILLIONOF CFAF) ....................... FIGURE2.9: MONTHLY GENERALSALESTAXES, 2003 AND 2004 (INBILLIONOF CFAF) 15 FIGURE2.10: DOMESTICPRIMARYBUDGETBALANCE,2001-05 (IN% OF GDP) ............................... ...............14 15 FIGURE2.11: PUBLICEXPENDITUREBYMAINSECTORS IN2005 (IN% OF GDP) ............................ 15 FIGURE2.12: SOURCES OF DEFICITFINANCINGDEFICIT, 2004-05 (IN% OFTOTAL FINANCING)16 FIGURE2.13: PUBLICEXPENDITUREBYMAINSOURCESIN2005 (IN% OF GDP) ............................. 20 FIGURE2.14: COMPOSITIONOF DEFICITFINANCING, 2003-05 (IN% OF OVERALL BUDGET DEFICIT) ............................................................................................................................................................. 21 FIGURE2.15: PUBLICINVESTMENTPROGRAM(PIP), 2004-05 ................................................................ 21 FIGURE3.1: EVOLUTIONOF THEPUBLICSECTOR WAGE BILL, 1997-2005 ........................................ 30 FIGURE4.1: GUINEA-BISSAU:STOCK OF DEBT, 1999-2005 ..................................................................... 47 FIGURE4.2: INDICATORSOF PUBLICAND PUBLICLY GUARANTEEDEXTERNALUNDER ALTERNATIVE SCENARII, 2006-2026(INPERCENT) ................................................................................. 54 FIGURE4.3: INDICATORSOF PUBLICAND PUBLICLY GUARANTEEDEXTERNALUNDER ALTERNATIVE SCENARII, 2006-2026(INPERCENT) ................................................................................ 55 Annexes: ANNEX 1: FISCALAND ECONOMICDATA ................................................................................................. 60 ANNEX 2: STATUS OF IMPLEMENTATIONOF KEYPOLICYISSUES IDENTIFIEDINTHE2003 PER .................................................................................................................................................................... 78 ANNEX 3: ANALYSIS OF TAX REVENUEPRODUCTIVITY ANNEX 4: PUBLICSECTORWAGES ............................................................................................................. - TAX BUOYANCY AND ELASTICITY 81 . 84 ANNEX 5: COMPENSATIONPACKAGEANALYSIS ................................................................................... 89 Annex Tables: TABLE A1 1: CENTRALGOVERNMENT OPERATIONS.2001-06 (INPERCENTOF GDP)................... TABLE A1 2: CENTRALGOVERNMENTOPERATIONS. 2001-06 (INBILLIONS OF CFA FRANCS)..60 61 TABLE A1 3: SELECTEDECONOMICAND FINANCIALINDICATORS.2001-06 62 TABLE A1 4: BALANCE OF PAYMENTS. 2001-06 (INBILLIONOF CFA FRANCS) .............................. TABLE A1 5:MONETARY SURVEY. 2001-06 .............................................................................................. TABLE A1 6: BREAKDOWNOF GOVERNMENT REVENUE. 2003-05 (INBILLIONOF CFA) ..............64 TABLE A1 7: BREAKDOWNOF GOVERNMENT REVENUE. 2003-05 (INPERCENTAGEOF GDP) ....65 68 TABLE A1 8: BREAKDOWNOF GOVERNMENT EXPENDITURE. 2003-05(INBILLIONOF CFA) .....71 TABLE A1 9: BREAKDOWNOF GOVERNMENT EXPENDITURE. 2003-05(INPERCENTAGEOF GDP) ......... ................................... 63 7') TABLE A1 10:FUNCTIONAL CLASSIFICATIONOF CURRENT PRIMARYEXPENDITURE. 2005-0673 TABLE A1 11:PUBLIC INVESTMENTBY SECTOR. 2003-08 74 TABLE A1 12:PIPPROGRAMMED AND EXECUTED IN2003 (INBILLIONCFA FRANCSUNLESS ... ..................................................................... OTHERWISEINDICATED) -75 TABLE A1 . 13: .................................................................................................................... PIPPROGRAMMEDAND EXECUTEDIN2004 (INBILLIONCFA FRANCSUNLESS OTHERWISEINDICATED) 76 TABLE A1 . 14: ..................................................................................................................... OTHERWISEINDICATED)..................................................................................................................... PIPPROGRAMMEDAND EXECUTEDIN2005 (INBILLIONCFA FRANCSUNLESS 77 TABLE A2.1: LEVELOF IMPLEMENTATIONOF THE MAINRECOMMENDATIONSINTHE PREVIOUSPER ........................................................................................................................................ 78 TABLE A2.2: LEVELOF IMPLEMENTATIONOF THE MAINWAEMUBUDGETARYREGULATIONS SINCE THE PREVIOUSPER ................................................................................................................... 79 TABLE A2.3: STATUS OF IMPLEMENTATIONOF WAEMU CONVERGENCECRITERIAAS OF 2005 .................................................................................................................................................................... 80 vi TABLE A4 1:PUBLICSECTORWAGESAND CATEGORIESOF SKILLS TABLE A4.2: PUBLICSECTORWAGESBY CATEGORY OF SKILLS AND EMPLOYMENT.2005......84 . ................................................ 85 TABLE A4 3: COMPONENTSOFTHEGOVERNMENT'S WAGEBILLIN1997AND 2005 85 TABLE A5 1:COMPENSATIONPACKAGESUMMARY STATISTICS .. ...................................................... .................... 89 TABLE A5.2: PREDICTORSOF TOTAL LOSS-DIRECT APPROACH. 1997AND 2005 90 TABLE A5 91 TABLE A5 4: POVERTYMEASURESINTHE CIVIL SERVICEBY GRADE. 2005 ..3: ......................... PREDICTORSOF TOTAL LOSS-INDIRECT APPROACH. 1997AND 2005 .................................. ..................... 96 Annex Boxes andFimres: BOX A4. 1:WAGEAND POVERTY 86 BOXA4 2: DETERMINANTSOFTHE CHANGEINWAGES . ................................................................................................................ ...................................................................... 87 FigureA4 1:Number of Civil Servantper Category . 88 FIGUREA4.2: AVERAGE SALARY (CFAF PERYEAR)PERCATEGORY OF CIVIL SERVANTS ........88 FIGUREA5 1:ANNUAL LOSSINEARNINGSFROMSEPARATION IN1997 .......................................... 92 FIGUREA5 ..2:ANNUAL LOSSINEARNINGSFROMSEPARATIONIN2005 .......................................... FIGUREA5.3: TOTAL LOSSINEARNINGS AND BENEFITSFROMSEPARATION IN1997 FIGUREA5.4: TOTAL LOSS INEARNINGS AND BENEFITS FROMSEPARATION IN2005 ................94 ................93 95 vii ACKNOWLEDGEMENTS This Public Expenditure Review (PER) update is based on the findings o f a mission to Guinea-Bissau, headed by Boubacar-Sid Barry (AFTP4). The event took place in March 2006 as part of ajoint Bank-Fundmission to discuss the authorities' budget for 2006 and the preparation o f a Staff Monitored Program (SMP). The mission worked with a local team composed o f technical staffs o f the public administration: Ministry o f Finances (MOF), Ministry o f Economy (MOE), Ministry o f Health (MOH), and Ministry o f Education (MOEd). The team included: Mr. Jeremias Pereira (Director o f Conjuncture, MOF), Mr. Vasco Silva (Director General o f Planning, MOE), Mr.Flavio Gama (Advisor to the Minister of Finance, MOF), Mr. RomZo Varela (Director General o f the Budget, MOF), Mr. Luis Rosario (Director o f the Budget, MOF), Mr. Carfa Embalo (Sr. Budget Technician, MOF), Mr.Francisco Camala (Sr. Budget Technician, MOF), Mr.Pedro Correia (Former National Coordinator DRRP, MOF), Mr.Andre Paul0 Nhanca (Director General o f the Department o f Ex-combatants' Pensions, MOF), Mr. Mario Nasoline (Director General, Basic Education, MOEd), Mr. Ibrahima Diallo (Director, Basic Education, MOEd), Mrs. Joana Monteiro de Macedo Ferreira (Director o f Civil Service Pensions, MOF), Mr. Jose Miguel Dias (Technician o f Pensions, MOF), Mr. Guy Mandef (Computer Technician, MOF), Mr. JoZo Viriato Rodrigues (Director External Debt Unit, MOF), Mr.Adulai Jalo (Statistician, MOE), Mr. Bessa Victor da Silva (Demographer, MOE), and Mr. August0 Paul0 Silva (Director, MOH). The task team leader and main author o f this report i s Boubacar-Sid Barry. The other members o f the PER core team are Quentin Wodon, Edward Creppy, and Clarence Tsimpo (AFTPM), Luca Bandiera (PRMED), and Manrique Saenz (IMF). The peer reviewers are Anita M. Schwarz, Lead Economist (ECSHD) and Denyse E. Morin, Sr. Public Sector Specialist (AFTPR). Martin G. Rama, Lead Economist (EASPR) has kindly agreed to be other peer reviewer for the civil service severance package analysis. The report was prepared under the supervision of Robert R. Blake (Sector Manager, AFTP4), who offered conceptual guidance, provided critical analytical advice and ensured quality control and management support. Madani M. Tall (Country Director AFC14) and Iradj A. Alikhani (Country Program Coordinator, AFC14) supported the process andprovided usehl comments. Quentin Wodon, EdwardCreppy, and Clarence Tsimpo (AFTPM) prepared the severance package analysis in the third chapter, and Luca Bandiera (PRMED), and Manrique Saenz (IMF) carried out the debt sustainability analysis inthe fourth chapter. The team is grateful to Abdoulaye Bamba, fiom Columbia University, for usehl research assistance on civil service reform andpension. As part o f the PER, five background papers have been compiled in a volume titled "Public Sector Wages andRetrenchment inGuinea-Bissau" for publication inthe Word Bank Working Paper Series. The volume was prepared under the coordination o f QuentinWodon by a team comprising Boubacar-Sid Barry, Edward Creppy, Clarence Tsimpo, and Quentin Wodon. The analysis builds in part on previous work by Albert0 Chong and Martin Rama who published in 2001 a previous version o f the volume's chapter 5 based on earlier data. The volume benefited from generous support fiom the Trust Fund for Environmentally and Socially Sustainable Development (ESSD). Vlll . . a EXECUTIVESUMMARY 1. Introduction 1. Guinea-Bissau's large public sector wage bill poses a major threat to the country's macroeconomic stability: it hampers growth, limits the government's ability to service the domestic and external debt, and crowds out private investments. For this reason, the government decided in early 2006 to retrench more than 2,800 civil servants ina first phase, and about 1,600 military later. The objectives o f this PER update are to: (i) progress review inmacroeconomic and fiscal management since the previous PER; (ii) the issue of analyze compensation benefits inthe context o f the ongoing civil service reform; and (ii) update the debt sustainability analysis for Guinea-Bissau. Besides this introduction, the report includes three main chapters and a conclusion. The rationale behind the core issues andhow they fit in the overall structure o f the report is briefly outlined below. 2. MacroeconomicManagementandPerformancesince the PreviousPER 2. Economicpolicy managementsince the previousPER.Duringthe period 2003-05, the government experienced fiscal difficulties due to a high wage burden. Despite these obstacles, the authorities made progress inthe implementation of key reform policies starting in2004. These efforts allowed the adoption of a SMP with the Fundinearly 2005 andearly 2006. 3. Economic performance since the previous PER. Economic performance was sluggish throughout 2003 but started to improve in2004. By end-2005, the real GDP grew at the rate o f 3.4 percent due to improved economic policies. Agriculture contributed for much o f the growth performance. Inflation as measured by the average annual change in the CPI remained on average broadly under control below 2 percent over the period 2003-05. The country's external competitiveness showed signs o f improvement with the growth o f the TFP improving from an average negative growth during 2001-04 to a positive growth in 2005. However, output growth continued to be driven by a factor accumulation rather than a productivity gain, and agriculture remained the main source of growth. 4. Fiscal policy management since the previous PER. Government fiscal situation was in a dire condition in 2003, and severe dysfunctional aspects of the budget preparation and execution processes were still present. The monitoring and control o f the budget execution was also deficient andreports on budget execution were not produced as expected. However, fiscal policy management has started to improve since 2004 thanks to the implementation o f a strict cash flow management system by the TC. Inthe area o f revenues, tax collection improved thanks to the centralization o f the State's bank accounts at the Central Bank o f West African States (BCEAO), and the abolition o f the practice o f tax compensations andthe strengthening o fborder controls. These efforts were further reinforced bythe creationo fthe large taxpayers unit (LTU). 5. There are still major fiscal policy issues that remain unsolved. Although the TC has been successful in ensuring that spending is in line with the resources available, it has not been able to contain extra-budgetary expenditures. The proliferation of the Non-established i x Expenditures (DNT), for example, has a strong causal effect on the accumulation o f domestic arrears. The non observance o f the established budgetary rules represents a major source o f expenditure backlog. Also, the multiplicity o f government accounts at BCEAO complicates payment processes. Inthe area o f the control o f budget execution the internal and external control mechanisms remain weak, and would need to be reinforced. In addition, the multiplicity o f government accounts at BCEAO further complicates the budget control process. With regard to revenues, the tax exemptions granted by the customs would need to be revisited and managed in a transparent manner. Finally, the current public procurement system established since 2002 would need to bereinforced. 6. Fiscal performance and outcome since the previous PER. Continued enforcement o f customs control in2004 and 2005 contributed to improve revenues. Inaddition, improved fiscal management during 2004-05 limited non-salary current spending. As a result, the government's fiscal and external positions gradually improved. However, revenues remain highly dependant on cashew nuts' exports. This seasonality represents a major source o f treasury and cash flow difficulties because the government is more likely to accumulate arrears in other periods o f the year and wait until the time o f the cashew campaign to clear them. 7. The economic analysis o f public expenditure indicates that the public sector wage bill represents a relatively large portion o f current spending (40 percent), and nearly half o f it goes to the security sector (military and gendarme and police). As a result, wage spending crowded out expenditures on goods and services, and the government continued to build up debt service arrears with external creditors. Inthis context capital expenditure remained low and was mostly financed by donor-funded projects. The sectoral analysis o f public expenditure also point to the predominance o f security outlays. In 2005 for example, spending in the security sector represented 7 percent o f GDP compared to 4 percent in the economic sectors and 6 percent in social sectors. The government was able to maintain a minimum service delivery inthe social sector only thanks to support from external donors. However, the coordination o f external aid remained weak due to weak institutional organization and capacity. 8. Macroeconomic policies in 2006 and medium-term prospects. Continued economic recovery and improved fiscal management allowed the negotiation o f a Staff Monitored Program (SMP) with the IMF in early 2006. The objective o f the S M P is to stabilize public finances and limit the accumulation o fnew arrears. The launching o f the civil service reform and the pursuit o f the ongoing customs and tax reforms are expected to improve the government's fiscal space. The security sector reform (SSR) i s expected to start as soon as the census o f the armed forces i s completed. The government is committed to pursuing efforts to strengthen the financial sector. Efforts are also being pursued to improve the business climate, good governance, regional integration, and growth and fiscal sustainability. These reforms are expected to enhance macroeconomic stability in the medium-term, which is a necessary condition for achieving the HIPC completion point expected by 2009. 9. Conclusion and policy recommendations.The high level o f personnel expenditure leaves little room for social policies and economic growth. In order to improve the X government's fiscal space, the launching o f the civil service reform and the pursuit o f the ongoing tax reforms and sound fiscal management under the established cash flow management system will be required in the coming three to twelve months. For this reason the government would need to accelerate the pace o f implementation o f the ongoing reform agenda (see Table 1, further below). 3. Improving the Fiscal Situation through Downsizing the Civil Service 10. Brief diagnostic of the civil service. The public sector wage bill has significantly increased since 1997. As an illustration, there are more workers in the public sector today thanthere are inthe private formal sector. Skilled workers' wages increasedmore thanthose o f their unskilled counterparts did. Despite the increase in the wage bill, the comparison o f the data from the civil service censuses o f 1997 and 2005 indicate a decrease in the civil service payroll. This decrease seemed to have been caused by a reduction in skilled workers. The wage o f skilled workers increased relatively faster than that o f their unskilled counterpart, which contributed to increase wage inequality inthe public sector. 11. Rationale and justification for compensation packages. Compensation packages for separated public sector workers are not new and are often estimated using quantitative tools inorder to avoid unfairness and overspending. A survey o fthe cross-country experience with public sector retrenchment in developing countries and transition economies estimated at USD 2,400 the average spending per job separation. However, the figure was as high as USD 13,000 inthe civil service o f Senegal, USD 16,000 inthe miningsector o f Bolivia, and USD 17,000 in the public enterprises o f India, despite the fact that all three countries had a per capita income o f less than USD 1,000 per year. In the case o f Guinea-Bissau, civil service reform through separation packages is an issue that has been around since the mid- 1990s. 12. Estimating the cost of downsizing the civil service. The retrenchment analysis is based on the Downsizing Option Simulation Exercise (DOSE) model. It evaluates separation benefits based on the estimated loss that the separated public sector worker would experience. DOSE uses two approaches. First, the direct approach measures the total loss that the separated public sector worker would experience based on the present values o f the earnings loss the worker would experience in all the years from displacement until retirement, and the income loss over all the years between retirement and death. Second, the indirect approach, inturn, measures the total loss based on the value o f the predicted change in earnings. Based on these two approaches, three alternative packages identified with the letters A to C are estimated. Package A is a lump sum. Package B i s a multiple o f the employee's salary ingovernment. Inpackage C, the amount o f compensation is set inmonths o f salary per year o f service. Predictedearnings under the 2002 data suggest that inthe event o f a retrenchment some o f the discharged public servants are likely to earn less inthe private sector. The estimated losses under the direct approach tend to be larger than those under the indirect approach with the 2005 census data. At the highest satisfaction rate (70 percent), packages B and C are less expensive than the lump sum package for both the direct and indirect approaches. On fairness ground, package C i s likely to fare better than packages A and B regardless o f the approach considered. The proposed downsizing o f unskilled worker is likely to improve the quality o f the corresponding services; however the reform would xi have a relatively limited fiscal impact. Additional restrictive fiscal policies would be needed, including a reduction inthe size o f the armed forces and further downsizing aimed at workers inother categories. 13. Assessing the potential poverty impact of retrenchment without severance benefits.The low skilled workers targeted for retrenchment are on average poorer than any other person. Retrenching low skilled workers without providing them a severance pay is therefore likely to worsen their condition. 14. Conclusion and policy considerations. The average compensation benefit per satisfied worker is expected to be between CFAF1.6 and 2.4 million, which translates into a total program cost o f CFAF5 to CFAF6 billion (3 to 4 percent o f GDP). On the other hand, the gross salary saving from the downsizing reform i s expected to be C F A l.15 billion (0.72 percent o f GDP). If appropriate compensation i s needed to make the workers accept voluntarily the prospect o f downsizing, it is very likely that the package will have to be negotiated with public sector trade unions. Inthat case, the final compensation formula may not be the cheapest one, or the fairest one, but rather a compromise shaped by the bargaining power o f all the players involved. From this perspective, the proposed simulation approach should be seen as a tool to introduce some economic rationality in delicate political negotiations. 4. DebtSustainabilityAnalysis 15. External debt situation as of end-2005. Guinea-Bissau's large public sector wage billhas limitedthe government ability to service its domestic andexternaldebt. At end-2005, Guinea-Bissau's stock o f external public and publicly guaranteed (PPG) external debt amounted to USD 1billion, o f which USD 327 million were arrears. The debt to GDP ratio has decreased slightly over the last five years from 353 in 2000 to 332 percent in 2005. Since the PRGF went o f f track in 2001, Guinea-Bissau has not benefited from most o f debt relief pledged at decision point. As a result, the country could not service its external debt andaccumulated arrears to most o fits externalcreditors. 16. Macroeconomic framework of the DSA. The DSA baseline assumes that an international concerted assistance effort is in place by end-2006 to help finance urgently needed structural reforms in the public sector. Post-conflict assistance is assumed to be provided by the Fundby end-2006, followed, after two years, by a PRGF. Growth will be sustained by reforms in the electricity sector and investment in infrastructure which would loosen the bottleneck for private investment and agricultural activity. Growth averaging 5 percent will be sustained by exports heavily dependent on energy-intensive services. The external current account deficit is projected initially to deteriorate as more resources are assigned to public investment and poverty reduction, but this would be offset by a strong growth o f exports in the medium and long-term. The primary fiscal balance is initially expected to deteriorate as more resources are assigned to public investment and poverty reduction. In the medium term, however, stronger revenue performance and Civil and security service reforms would contain the public sector wage bill and improve the fiscal balance. Public external borrowing at concessional terms would be contained over time, while domestic debt and arrears would be progressively paid down. HIPC Completion Point xii is expected by end-2009. Total HIPC debt relief, as established at Decision Point in 2000, amounts to USD 413 million in net present value terms and would reduce debt service payments by about 7 percent o f GDP per year. 17. External debt sustainability. Guinea-Bissau is currently in a situation o f debt distress with accumulation o f external arrears. Evidence suggests that less concessional borrowing, exchange rate depreciation and lower growth o f exports are likely to increase the risko fdebt distress. 18. Conclusion and Staffs appraisal. The DSA points to a country in debt distress, which would be substantially reduced if the HIPC Completion Point is achieved by end- 2009. Continued political stability as well as a concerted international assistance effort are key elements for the country to achieve Completion Point and the associated debt reduction to sustainable levels. Even within this scenario, the country's capacity to sustain this debt level may be compromised by significant shocks to the nominal exchange rate, the growth rate, and the degree o f loan concessionality. Given that Guinea-Bissau will hit the one-third limit o f IDA HIPC interim reliefin2007, consideration should be given to the possibility o f extending IDA interim assistance as allowed by the Board-approved `Provision o f Interim HIPC Debt Reliefon IBRD Loans and IDA Credit'. 5. Conclusionto the PER 19. The main messagesof the PER.Public finance management has gradually improved since 2004, but the overall fiscal situation has remained very difficult. The proposed retrenchment reform is a positive step towards improving the fiscal stance; however, it will take time before the fiscal situation is fixed. Inthe meantime, it is vital that donors continue to support the government's reform agenda, and the authorities will need sizable external support inorder to implement both the civil service reform andthe SSR. 20. The way forward. While the proposed fiscal consolidation reform represents a step forward, it will need to be supplemented by further expenditure rationing and revenue enhancing policies in the near future. Two key elements o f interest in that regard are the reform o f the public sector pension system and the transition from the current General Sales Tax (IGV) system to the Value Added Tax (TVA). Duringthe next PER exercise, it would be useful to work together with the government PER team to explore ways to address these issues. In the area o f pensions for example, it will be useful to develop a pension management tool such as the Pension Reform Option Simulation Toolkit (PROST). PROST was developed by the World Bank and has been applied inmany countries. The objective is to develop a PROST model for Guinea-Bissau, which can be then transferred to the authorities. Inthe process o f adopting the TVA, it would be usehl to analyze the potential fiscal impact andthe poverty or distributive effect o f the TVA system, and explore mitigating factors for the poor. xiii x x x x x x > .3 X x x X x x x x x -E & e e 'G 1. INTRODUCTION 1.1 With the June-July, 2005 presidential elections, Guinea-Bissau completed its post-conflict transition. However, the country is still facing severe socioeconomic challenges. The public administration has not recovered from the enormous destruction caused bythe 1998 war. Yet the public sector remains the main employer o f the economy as the private sector i s still ailing from the nearly-a-billion dollar damage ininfrastructure caused by the conflict. This situation is hrther complicated by weak domestic private investments and the absence o f foreign direct investments due to the fragility o f the political and security situation. All these factors contributed to hinder economic performance and hence the capacity o f the State to deliver basic social services. According to a recent Bank poverty assessment in Guinea-Bissau, nearly two-thirds of the population lived with less than USD 2 per day in 2002, and the majority o f the poor lived in rural areas. For people living in Bissau, the poverty incidence was relatively higher for low skilled public sector employees who represent about 30 percent of the work force inthe sector. 1.2 Confronted with low productivity, mounting salary arrears and other difficulties inherent to a bloated civil service, the government decided in early 2006 to carry out a reform o f the public service. The reform is seeking to retrench more than 2,800 civil servants in a first phase, and about 1,600 military later. The civil service component will target essentially workers in the categories corresponding to the low skilled labor force. The ultimate goal o f the reform is to subcontract the targeted activities to private firms to be selected through competitive bidding, and allow these firms to hire the retrenched workers. The implementation o f the reform i s expected to improve the fiscal situation through better public services and a reduced the wage burden. 1.3 The proposed reform is certainly not sufficient to significantly reduce recurrent expenditures and reach a sustainable fiscal balance. The reform will need to be extended to the security sector and the upper categories o f the civil service. It is urgent that the wage bill is brought to a sustainable level in order to secure financial resources for the delivery o f social services and the payment o f the external debt service so as to secure a positive net capital inflow for projects. Since 2001, a less than satisfactory macroeconomic performance has prevented Guinea-Bissau from taking full advantage o f the interim assistance under the Highly Indebted Poor Countries (HIPC) initiative. As a result, the country has been accumulating debt service arrears vis-a-vis a number o f creditors, thereby limiting the possibility to tap into additional concessional assistance. This situation contributedto increase the level o fdebt distress. 1.4 The remainder o f the report is organized as follows. The second chapter will review progress in macroeconomic and fiscal management since the previous PER. The third chapter will analyze the issue o fcompensation benefits inthe context o fthe ongoing civil service reform. The fourth chapter will update the debt sustainability analysis for Guinea-Bissau. The concluding chapter will summarize the main messages o f the PER andhighlightkeypolicy issues to be addressedinthe near future. 1 2. MACROECONOMIC MANAGEMENTAND PERF'ORMANCE SINCE THE PREVIOUS PER: A SITUATION STILLDIFFICULT This chapter reviews progress in macroeconomic and public finance management since the previous PER. Since 2003, Guinea-Bissau's political situation has improved but remains fragile. Public finance management has also gradually improved but the overall fiscal situation remains difficult. Economic pe~ormancewas sluggish throughout 2003, but started to show signs of improvement in 2004 and continued to recover in 2005. While inflation was on average kept under control at 1.2 percent during 2003-04, this rate has more than doubled to reach 3.4 percent by end 2005. The large public sector wage bill, whichfireled both the budget and external account deficits, was the main contributing factor to the inflationary pressure. I n this context the government ofen accumulated salary and other domestic and external arrears. Also, the scarcity of financial resources provided a breeding ground for ad hoc public finance management behaviors beside the established cash flow management system. Enforcing the established budgetary rules as well as building capacity in expenditure management and control, and enhancing revenue collection will be critically needed. Equally important is the reform agenda agreed upon with the donor community in the context of the 2006 SMP. Thesuccessful implementation of the SMP will certainly pave the wayfor a renewed supportfrom the Bank and other donors. ECONOMIC POLICY MANAGEMENTTHE PREVIOUS PER SINCE 2.1 The government experienced fiscal difficulties due to a high wage burden. During 2003, the government implemented an 8-fold increase in the salary o f the military, which contributed to raise the public sector wage bill from 7 to 13 percent o f GDPbetween2002 and 2005. As a result, the domestic primaryfiscal deficit (which was about 5 percent o f GDP in2002) increased by nearly 2 percentage points o f GDP by end- 2005. The immediate consequence o f this situation has been recurrent delays in salary payments andthe accumulation o f domestic and external arrears. 2.2 Despite these fiscal difficulties, the authorities made progress in the implementationof key reformpoliciesstarting in 2004. By end-2003, the government prepared a budget for 2004, which served as a basis for the preparation and implementation o f a donor-supported emergency economic management plan (EEMP)'. Under this plan, donor resources were channeled through a UNDP-managed emergency fund to finance salaries and other urgent expenditures in social sectors. Moreover, in order to ensure transparent and efficient management o f public finances, the government established an independent Treasury Committee (TC) to monitor revenue collection and expenditure management on a cash-flow basis. This measure contributed to improve fiscal discipline by ensuring that spending programs were consistent with the resources available. 2.3 These efforts allowedthe adoptionof a staff monitoredprogram(SMP) with the Fund in early-2005. The improvement in the financial situation under the EEMP allowed the negotiation o f a SMP for the period April-December 2005. The performance under the SMP was mixed'. The targets for government revenue and the domestic financing o f the budget for end-December were met by a considerable margin, and the domestic primary budget deficit (commitment basis) was lower than programmed. 'In September 2003, abloodlessmilitary coup forced PresidentYala to resign. Subsequently, anew governmentwas appointed which *endorsed thecould EEMP and organizedparliamentary elections in2004 and Presidentialelectionsin2005. The SMP not be assessedas planned at end-2005 due to the change of government that took place following the presidential elections. The assessmenttook place inMarch 2006 concomitantlywith the preparationof the 2006 SMP. - 2 - However, inthe face o f shortfall inexternal budget support, the domestic financing target was in part met by placing short-term certificates in the financial markets o f WAEMU. Overall, two out o f the five quantitative indicators o f the SMP (ceiling on salaries and accumulation o f domestic arrears) were missed: the indicative targets for the wage bill in the last quarter o f the year were exceeded, and new domestic arrears accumulated during the same period. Progress on structural reforms was also mixed, reflecting significant technical constraints. As a result, three out o f ten structural indicators were missed (see Box 2.1). Box 2.1: Quantitative and Structural Indicators Under the S M P in2005 Quantitative indicators End-June End-Sept. End-Dec. (inbillionofCFA franc) Act Prog Act Prog Act Prog Domestic financingof the budget 2.3 2.1 -5.5 -2.2 -5.6 -3.7 Domesticprimary budgetbalance(commitment -4.5 -5.6 -0.4 -9.1 -11.0 -12.9 basis) Revenue 3.6 14.0 18.8 20.6 28.0 26.8 Wage bill 5.0 10.2 10.0 15.0 22.4 20.5 Accumulationof domestic arrears 1.8 0 4.7 0.0 4.2 0.0 Externalpublic debt, maturities<=l year 0 0 0 0 0 0 Externalnon-conces.public debt maturities>l 0 year 0 0 0 0 0 Structural indicators -Indicators largetaxpayer unit within the MOF under the 2005 SMP Target date Status as of 03/06 Establisha End-June2005 Done - Implementbanderolesystemfor alcoholic beverage& cigarettes End-June2005 - Transfer the Partially done collectionof customsduties and taxes to the CustomsOffice inthe port of Bissau End-June2005 Done - Completediscussionswith all Ministries and other budgetary unitsonthe budgetfor 2006. End-Sept.2005 Done - Approvalby the Councilof MinistersoftheOrganicLaw of theMOF andthe Ministry of Economy. End-Sept.2005 Not done (ready but still awaiting approvalby Council ofMinisters) - Rehabilitationof the taxpayer master file. - Implementationofthe Done End-Dec. 2005 Depar!ment . new organizationalchart of the Customs Not done (lack of End-Dec. 2005 - Implementationofthe censuses capacity) of the military andpara- military. Partially done - Reconciliationof End-Dec. 2005 (para-mil. Census changesinthe treasury accountsat the completedinBissau) BCEAO with expendituresauthorizedby the Treasury Committee, and immediateliquidation of such expenditures. Continuous Done - Collect fullamountof tax due on: (i) importsofpetroleum products; (ii)rice imports; and(iii) cashew nuts exports Continuous Done - 3 - ECONOMIC PERFORMANCE SINCE THE PREVIOUS PER 2.4 Economicperformancewas sluggish throughout2003 but started to improve in2004. Followingthe recession o fthe period 2001-2003, real GDP grew by2 percent in 2004, thanks to better than expected agricultural performance mainly in the cashew sub sector. The recovery was sustained in2005, as the economy is estimated to have grown at the rate o f 3.5 percent in2005 (see Table 2.1). With the population growing on average at 3 percent in 2005, the negative trend observed since 2001 on the per capita GDP growth rate was reversed. This performance was made possible thanks to the implementation since 2004 o f a set o f structural reforms, which allowed improving public expenditure, enhancing revenue collection through greater customs control, and improving the business climate through the simplification o fthe tax code3. Table 2.1: SelectedEconomicandFinancialIndicators,2003-2006 2003 2004 2005 2006 Actual Proj . RealSector (annual percentagechange) RealGDP at market prices -0.6 2.2 3.5 4.2 RealGDP per capita -3.6 -0.8 0.2 1.6 Consumer price index (annual average) 1.6 0.8 3.4 1.o Export volume 7.3 15.8 9.5 7.7 Import volume -12.9 40.9 35.5 9.9 Terms of trade (deterioration-) -6.0 -3.9 14.4 0.1 Realeffectiveexchange rate(depreciation-) -2.3 0.6 0.4 ... (changeinpercent of beginning-of- Money and Credit periodstock of broadmoney) Creditto government(net) -2.4 -7.4 -6.2 -5.7 Creditto the rest ofthe economy -1.8 -1.3 2.5 4.4 Broadmoney -65.3 44.0 19.9 8.5 Investmentand Fiscaland Externalaccounts (inpercentofGDP) Grossinvestment 12.6 13.2 14.6 17.2 Ofwhich: governmentinvestment 11.1 11.1 10.0 11.6 Domesticprimary balance -4.7 -6.5 -6.5 -2.8 Overallbudgetbalance(commitmentbasis) Includinggrants -15.7 -15.0 -12.0 -8.1 Excludinggrants -23.4 -32.2 -20.4 -19.2 Externalcurrent account(including official current transfers) -2.8 2.4 -7.0 -6.9 Externalcurrent account (excludingofficial current transfers) -12.0 -7.4 -11.1 -13.0 Externalcurrent account (excl. off. Tr. other than fishing licenses) -6.2 -5.4 -5.4 -10.1 Memorandumitems Nominal GDP at market prices(inbillions ofCFA ) 137.1 142.6 158.8 172.3 Nominal stock ofdebt (e.0.p; excl. arrears, inmillions of USD ) 1,362.9 1,565.6 1,347.6 1,302.2 Nominal stock of arrears, e.0.p (, inmillions ofUSD) 355.0 462.3 517.0 528.5 NPV of external debvexports of GNFS (in percentof exports) 635.7 477.1 359.8 320.1 Source: Guinea-Bissau authorities and IMF staff estimates and projections. 'For example, the industrial contribution or business profit tax activities (whether permanent or occasional) -- was --a direct tax levied on income from commercial and industrial reduced from 39 to 30 percent and the government i s planning to further reduce it to 25 percentby 2007, inline with the recommendations of the Bank's Foreign Investment Advisory Service (FIAS). - 4 - 2.5 The growth performance in Figure 2.1: Key Indicators of Economic health, 200145 2005 was largely attributable to (in annual YOchange) improved policies. The government 6.0 1 .._ fiscal stance improved noticeably over 2.0 . the past two years. The overall budget deficit (commitment basis including V grants) as a share o f GDP, which represents nearly 16 percent in 2003, -6.0 - decreased to 15 percent and 12percent in -8.0 . 2004 and 2005, respectively. Better -10.0 - III Real GDP growthrate public finance management in 2005 .12,0 ~ allowed availing more financial resources to the rest of the economy. The growth rate of the credit to the rest of the economy, which has been negative since 2002, rose by 2.5 percent in2005 (see Figure 2.1). Consequently, private investment as a share of GDP more than doubled passing from 1.5 percent to 4.6 percent between 2003 and 2005. 2.6 Agriculture contributed for much of the growth performance.As in Figure 2.2: Real GDP and Cashew Production the past, growth continued to be driven by Volumes, 199943 (annual YOchange the agricultural sector with raw cashew 6.0 -r -40 production playing the leading role. The 4.0 agricultural sector4 continued to be the 2,0 o,o main source o f growth in Guinea-Bissau. -2.0 --20 The sector represented about 60 percent of -4.0 -- GDP and grew on average by more than 3 -6.0 -- -- -10 percent per annum over 2003-05, at the -8.0 - --20 RealGDP GrowthRate(left axis) I improvement are construction, Figure 2.3: Government Deficit and Inflation, 2001-05 transportation and commerce. 4.0 T T 0.0 3.5 -- -2.0 2.7 Inflation as measured by the 3 3.0 -- -4.0 average annual change in the consumer 8 2.5 -- -6.0 0 c price index (CPI) remained broadly -g s 2.0 -- -8.0 sO under control during the past three 8 -10.0 .E 1.5 -- years. The average inflation rate was about -12.0 1.0 -- 2 percent over the period 2003-05. On an -14.0 0.5 -- annual basis, however, the rate increased -16.0 0.0 , I ,, I 1 -18.0 fiom 0.8 percent to 3.4 percent between II I 2004 and 2005 (see Figure 2.3). This volatility may be explained by the 'Agriculturecomprisesfoodcropping, fishingandforestry. - 5 - combined effect of the 2003 salary increase and the recent oil price hike on the government budget deficit and consequently on the aggregate demand given the fact that the public sector i s the main employer inthe economy5. 2.8 The country's external competitiveness showed signs of improvement. Guinea-Bissau's current Figure2.4: Changes in TT andREER,2003-05 (in%) - - account balance (excluding official 20.0 1.0 transfers other than fishing licenses) 15.0 -- -- 0.5 improved by 1 percentage point o f GDP -- 10.0 -- 0.0 between 2003 and 2004 to stabilize at 5 -- -0.5 percent o f GDP by end-2005. At least two 5.0 -- -- -1.0 factors may have contributed to this 0.0 -- -1.5 outcome: since 2004, Guinea-Bissau's -5.0 -- 2904 2005 -- -2.0 terms o f trade (TOT) and real effective -10.0 - exchange rate have been moving - --- - --2.5 T e r n of trade (deterioration -)left axis in opposite direction, which indicates that - the gain in external competitiveness MER(real depreciation-)rightaxis associated with a real depreciation o f a factor accumulation rather than a Figure2.5: TFP GrowthRate, 2001-05 productivity gain, and agriculture 1.00% 1 remains the main source of growth. Real 0.00% / -/ I I I / I I GDP grew by about 2 percent on average -1.00% - during 2003-05. Agriculture remains the -2.00% - main source o f growth with a share o f about 2001 2002 ,4003 2004 2005 -3.00% - 60 percent o f GDP and an average real / growth rate o f nearly 3 percent over the same period. As indicated in Figure 2.5, assuming an output elasticity o f labor ranging between 0.65 and 0.75 - as suggested by similar empirical studies on developing countries - capital and labor Lower food and rentprices contributed to party offset the increaseinthe overall CPI. The measure of the REER used is the index constructedby the Fund, which defines the real exchange rate as the relative price of non-tradableto tradable goods. A decline of the index impliesarealdepreciation inthe value of the local currency. 'Theinternationalprice of cashewnuts went upby morethan 25 percent in2005, which contributedto completelyoffsetthe adverse *shockTOT are from the oil crisis. The determinedexogenouslybasedon the internationalprices of exportedand importedgoods. - 6 - have increased on average by about 3 percent each during2001-05. At the same time, the total factor productivity (TFP)' contracted by about 1 percent. A similar estimation performed at the time o f the previous PER indicated that the production factors and the TFP grew by 2 percent and minus 3 percent, respectively. This trend clearly shows that while growth continues to be driven by factor accumulation, there are signs o f productivity improvements (as evidenced by the decreased negative average growth in the TFP since the previous PER). FISCALPOLICYMANAGEMENT SINCETHE PREVIOUSPER Governmentfiscalsituationwas in a dire conditionin2003 2.10 The fiscal situationwas difficultin 2003. As the two precedingyears, 2003 was characterized by continuing deterioration o f public finance management. Following the dissolution o f the Parliament by President Yala in November 2002, the Government operated without a formal budget in2003 because there was no legislative power to vote and approve a budget proposal. Government expenditures were therefore carried out on the basis of the previous year's budget without a relation with the real needs o f the sectors. As a result, public finance management deteriorated significantly during the first nine months o f the year 2003: the Government accumulated salary arrears and was unable to ensure the delivery o fbasic social services. 2.11 Severe dysfunctional aspects of the budget preparation and execution processes were still present in 2003. There were major weaknesses in the areas o f budget preparation, execution, and control and reporting (see Annex 2, Table A2.1). In the area o f budget preparation, revenue and expenditure forecasts lacked rigor and strategic planning. In the absence o f a medium-tern forecasting framework, the authorities simply relied on the previous year's forecasts, which were often irrelevant to the present situation. Inregard to budget execution, tax compensations - the deduction o f outstanding government liabilities from the amount o ftax to be paid bybusinesses -were a frequent practice at the customs and other revenue departments. The direct collection o f revenue at the source by various Ministries and sometimes by the armed forces was another major source o f tax evasion. These practices significantly hampered government revenue collection effort. They also contributed to deteriorate the fiscal situation as the corresponding resources were not channeled through the budget, and therefore were not used for the intended purposes. Expenditure execution did not follow budget execution rules: in2003 all non-wage spending (about 75 percent o f total expenditure) were carried out without a prior commitment. This situation referred to as `non established expenditure' (Despesas NZo Tituladas, DNT) represented the main source o f extra- budgetary spending (for example the DNT for the month o f April 2003 alone accounted for more than CFAF300 million representing about 6 percent o f the spending on goods and services for the whole year). Although the year 2003 was an exceptional case due to the'absence o f a formal budget, this practice is a violation o f the `normal' budgetary The TFP is estimatedusingaCobb-Douglassproductionfunction with constantreturns to scale underperfect game of the form Y = AKb La,O< a,b F= 0.00 Roor MSE = 0.15 Obs.26 The elasticity o f a tax i s defined as the relative change in revenue from tax compared with the relative change in the tax base (proxied with the GDP or the GNI), holding the structure of the tax (tax rates, brackets, coverage, exemptions, and deductions) (Barth and Hemphill, 2000). Tax elasticity can also be estimated using the dummy variable (D) technique, which is an econometric method (Manasan and Querubi, 1986) o f simultaneously adjusting for discretionary revenue effects inequation (1). logT, = a, p,logr; + yID, + p,D, log + r, (2) IogT, = 0.09 + O.OSlogY,-1.310, +O.lZD,.logY, (2.1) (0.27) (0.03)' (0.57)' (0.06)' Adj. R' = 0.51 Prob > F= 0.00 Root MSE = 0.14 Obs.2 The estimated coefficient is the tax elasticity net o f discretionary tax policies. A tax system i s elastic ifit has an elasticity value greater than 1, suggesting that tax revenues are increasing at a higher rate that the tax base without discretionary changes intax policy (i.e. adoption o f new taxes or increases intax rates). A tax elasticity greater than one may be desirable ina tax system and should be encouraged incountries inwhich government expenditure tend to increase more rapidly than GDP. The tax system is likely to be elastic with respect to the tax base when taxes are levied on growing economic sectors and when tax rates are progressive. The tax system will be inelastic if taxes are specific rather than ad valorem, andor taxes are not collected promptly. This last point i s especially important in the case o f high inflation, as an unduly long lag between the assessment and collection o f taxes would erode the real value o f tax revenue (the so- called Tanzi effect). An elastic tax system is also useful from the point o f view o f economic growth, which generally calls for sustained rise in spending on social and economic infrastructure maintenance. Ifthese expenditure increases are not accompanied by rising revenues, they can lead to undue reliance on deficit financing, either external or domestic. Inthe case o f Guinea-Bissau the tax buoyancy and elasticity o f total, tax revenue i s estimated using equation 1 and 2 above. With regard to buoyancy, the results indicate an estimated coefficient less than one for total tax revenues (B,I0.1 inEquation 1.1).This result suggests that total tax revenues have increased less rapidly than the national income. This seems to suggest that discretionary tax policies may have had a deleterious impact on total tax revenues. The results may also justify the low tax-to-GDP ratio that the country experienced over the past decade (on average less than 12 percent). The results indicate that when the series o fthe actual tax collected is adjusted for discretionary tax policies, the coefficient o f the total tax revenue decreases slightly and remains inelastic (PI I0.08 in Equation 2.1). This result suggests that total tax revenues are increasing at an even lower rate that the tax base without the discretionary changes intax policy. Source:Staffestimates based on LDB -17- EconomicAnalysisof PublicExpenditure:Large Wage BillandLow Pay Table 2.3: Economic Compositionof Spending 2003 2004 2005 2.32 The public sector wage bill In billion of CFA represents a large portion of current Current expenditure 35 42 44 spending, and remains relatively large Ofwhich: Personnelexpenditure 14 16 21 compared to the resource base. In Goods and services 4 5 8 Guinea-Bissau, the bulk o f the workforce in Interest 8 9 7 the formal sector is employed by the public Capitalexpenditure 18 28 17 sector. As shown in Table 2.3, Government I npercentageof GDP unlessotherwise(*) wage spending represented on average Current expenditure 26 I O more than 40 percent o f current spending [431 and more than 10 percent o f GDP over the {39} Ofwhich: Personnelexpenditure (26) past three years (compared to less than 30 Goods and services 3 and 8 percent, respectively, during 2001- Interest I l l6} 02)". On average, salary levels in Guinea- (221 Bissau's public sector are generally lower 13 10 Capitalexpenditure (34) (39) than in WAEMU. As an illustration, the Memorandum items(billion of CFAF) average wage inGuinea-Bissau is about one Nominal GDP 137 143 159 fifth of the average public sector wage in Total Revenue 31 49 41 Senegal. The relative importance o f the Source:MOF (*)The figures in [ ] { } and ()are inpercentageof: total wage bill is therefore largely due to the size revenue, current expenditure, and total expenditure o f the payroll as opposed to salary". respectively. 2.33 Nearly half of the wage bill goes to the security sectorz3.The sector employs nearly 7,000 personnel, of which about 5,000 are active military personnel. While this size i s similar to the average in WAEMU, the number o f security personnel per population head remains relatively high: there are 5.04 soldiers per 1,000 inhabitants in Guinea-Bissau compared to a regional average o f 1.23. Moreover, wage scales in the upper ranks o f the m y are higher than their equivalent in the civil servicez4. The combination of these factors contributed to raise the security sector wage spending to more than45 percent o fthe total wage billby end-2005. 2.34 Wage spending crowded out expenditures on goods and services. Expenditures on goods and services have experienced a downward trend over the past few years. From an average o f 20 percent o f current expenditure (or 5 percent o f GDP) during 2001-02, they decreased to about 17 percent (or less than 4 percent), inthe same units, during2003-05. 2.35 The government continued to build up debt service arrears with most external creditors.AAer the suspension o f the HIPC interim assistance by the Fund in 2002 (following the failure of the PRGFprogram during its first year o f implementation), the Paris Club and other bilateral and multilateral creditors (excluding the Bank and the ''Theratiosto total revenueare inthe same order of magnitude. 22A civil service census camed out inend-2004estimatedthe number of civil servants at around 12,000. 23The security sector includesthe military and the paramilitary (police and gendarmes). 24For example a Lieutenant Coloneland a Generalearn CFA420,OOO and CFA672,000, respectively, per month.On the other hand, a Director Generaland aMinister are paidCFAF240,OOO and CFAF451,200, respectively,per month. - 18- AfDB) also stopped their assistance. As a result, all debt service payments due to these creditors (excluding the Bank, the IMF andthe African Development Bank (AfDB))went into arrears. As o f end-2005, the stock o f external debt service arrears reached nearly USD300 million (about 20 percent increase innominal terms on average over 2002-05). Moreover, the government continued to contract new debt obligations mainly on concessional terms. As a result, the external debt stock (inNPV terms) reached more than USD750 million (representing over 280 percent o f GDP or more than 1000 percent o f exports) by end-2004. 2.36 Capital expenditures remain low. On average, capital spending represented about 34 percent o f total expenditure during 2003-04. This allowed reaching an investment ratio of 13 percent o f GDP over the same period. However, this performance was lower compared to the investment level within WAEMU: the average gross capital formation in WAEMU during 2002-03 was about 18 percent o f GDP compared to 13 percent for Guinea-Bissau (see Table 2.4). Table 2.4: Comparison of the GrossCapital Formationwithin WAEMU (in YOof GDP) CountryName 2003 2004 Average 2003-04 Benin 18.6 20.3 19.5 Burkina Faso 18.7 19.1 18.9 Cote d'Ivoire 10.1 10.8 10.5 Guinea-Bissau 12.6 13.2 12.9 Senegal 20.7 23.4 22.0 Togo 18.9 18.0 18.4 Mali 23.9 19.7 21.8 Average WAEMU 17.6 17.7 17.6 m:WorldBank (GDF & WDI Central) Table 2.5: Financing of the PIP, 2003-04 2003 2004 2005 2.37 Government capital spending is Inbillion ofCFA unless otherwiseindicated * mostly financed by donor-funded 18 28 17 projects. The government's public Investmentexpenditure [341 1391 3 P O2I 2 investment program (PIP) represented Domesticallyfinanced {17) (7) about 38 percent o f total spending during 15 26 15 2003-05 (see Table 2.5). Nearly 90 percent Externallyfinanced (83) {93) {88) PIP by Sector, (in% of externallyfinancedinvestment) o f this amount was financed by resources Economicsectors 26.5 33.7 46.1 from donor-supported projects, o f which Social sectors 29.1 23.1 21.2 about 60 percent are donor-financed project Other 44.4 43.3 32.8 in social and economic sectors (education, m: Total 100 100 100 health, agriculture and infrastructure). * The figures MOF While it i s generally assumed that about 60 in [ ] and { } are inpercentage of: total expenditure and capital expenditure,respectively. percent o f the PIP is executed inthe course o f a fiscal year, such information can hardly be confirmed by actual data due to the absence o f adequate aid monitoring mechanisms. - 19- Sectoral Analysis of Public Expenditure: Still Dealing with Security Issues 2.38 Expenditures with the security sector remain relatively higher than in the Figure2.13: PublicExpenditureby Main Sectors in 2005 (in YOof GDP) other sectors. Security sector-related spending rose from 3 percent o f GDP in 1 7.0 2002 to 7 percent by end-2005 (see Figure *.O 5.6 n 2.13). This development contributed to . crowd out other domestically financed expenditures. For example, spending on institutional and economic services experienced a decrease o f about 1percentage Institutional sectors 49cOoomic sectors point o f GDP over the same period. 2.39 Donors support allowed maintaining a minimum level of expenditure in social sectors. Expenditures in social sectors (education and health) remain low compared to regional and international standards. Over the past decade, social spending represented 4 percent o f GDP on average inGuinea-Bissau compared to 5 percent for the average developing count$ (see Table 2.6). The bulk o f the government's spending went to salary payments with little or no resources left for goods and services (supply, maintenance costs) andcapital goods. Table 2.6: Government Spending in SocialSectors (percentage of GDP averagedover 1996-2004) ~ ~~ Health Education Total Benin 1.8 3.5 5.4 Angola 2.1 2.7 4.8 Cote d'Ivoire 0.9 4.1 5.1 Ethiopia 1.4 4.3 5.7 Ghana 0.1 0.5 0.7 Guinea 0.5 1.9 2.4 Guinea-Bissau 1.6 2.8 4.4 Kenya 1.5 6.2 7.7 Lesotho 2.9 10.1 13.0 Mali 2.0 2.7 4.7 Mauritania 1.8 4.3 6.0 Mauritius 2.1 3.8 5.8 Mozambique 2.9 4.6 7.5 Rwanda 0.6 2.8 3.4 Senegal 1.3 3.4 4.8 Uganda 1.2 3.6 4.8 Average all countries above 1.6 3.8 5.4 Median all countries above 1.6 3.6 5.0 Average WAEMUE (excluding Burkina Faso and Togo) 1.5 3.5 5.0 Source: World Bank WDI database, 2005 25Onlycountriesfor which dataoneducationandhealthare availableover 196-04are considered. - 20 - Sources of Budget Deficit F ~ ~ ~ ~ c i n g 2.40 The ~ ~ ~ ~ e r n e~s se ~nnt~ ~ ~ l l y relied on arrear a & c ~ ~ ~ l ~ and t i o n ~ ~ to closeS ~ ~ t ~ ~ ~ ~ A s h i ~ h l ~ing ~ t ~ ~ tion o f a ~ e a-r ~ external cred~~ors- rep relatively lasge c o r n p o ~ e o ~ t deficit during ~ ~ 0 3 - 0On average, mear 4 . resented more than half er the period under ~ o ~ s j d e ~ 8A t ~ o ~ . ~ ~ a r - b ~ - matysis indicates that arrears ~ e a r 100percent of the budget deficit in 2003, but this ratio was ~ r Q u ~down to h t 15 percent by e ~ ~ d - 2thanks5to the i ~ ~ l e ~ e ~of thea cash oflow ~ ~ ~ a g e n ~ e n t ~ ~ ~ t ~ ~ n system staring in 2004. al resources financed more its on average over the indi eriod, On a yearly basis, t to more than 90 percent 3 and 2005, and the net d o ~ ~ financing~was s t ~ negat~vet ~ o ~ ~ htheuperiod. Also, as the expected external b ~ d g e t asupport has o t ~ failed to ~ ~ a ~ ein~2 ~~ ~due ito~d ~e f f i ~ ~ lto ires~oren ~ a e ~ Q e ~ o ~ ot ~ ~bc~and~ t y ~l 5 ~ e s n I good gQve~ance,rhe ~ ~ v e ~relied on short-term b o ~ o ~In October .2005, the ~ e n t ~ ~ n ~ ~ omade reco~ to the~reg~ona~ rn ~ et for short-term ~~ o ~ ~ e ~~paper n ~ ~ n ~ efor t t the first time, The a ~ ~ ~ o rcollectedsCFAF 6 ~ j ~ ~ e j l (about~ 4~percent of GDP) by l ~ issuing 8 G ~ ~ ~ rea~as^^hbill that was mostly subscribed by banks outside Bissau". o t Over the n ~ e term, d ~the~ ~~ o ~ ~ intends toe issuet treasury bills for l j ~ u ~ d ~ t ~ ~ ~ n p ~ ~ o sands also s u p p o ~~ n ~ emarket~d e ~ ~ l o p inthenregion. e i a ~ e ~ 2.41 The exe rate of external ced capital a p e g o ~ y e ~projectse ~ndica~ethat the ~ ~ ~ ~ annual e x ~ c u t ~ orate of the PIP n d below SO percent during ~~~~- Figure 2.15). This number shou~d however be ~ ~ n s i d ~~r e~d u ~becj o ~ ~ ~ y some ~ r ~ j e c t - r e l 8~t ~~dp e ~ d i tar r e ~ u r~corded in forthe budget, Even when a~justed the non-recorded e~pe~~dj~ures,the rate of e ~ e c u t i oo~f ~the PSP is ~ e ~ e r a lnot expected to exceed GO pe l y 2,112 I ~ ~ donor co ~ ~ ~ is ~criticalo for~ ~ n ~ ~ the much needed ~ ~ ~ ~ ~ ~ ~ n a ~ c iresources for the budget. The c o ~ r d ~ ~ ~ aoft iexternal aid remains weak due a l o n fo a lack of ~nstitut~Qna~~ r g a n i ~ a t iand~ capacity. In ~ ~ n c ~ pthe ,task of aid o l e -21 - coordination is ensured by three departments: the International Cooperation Directorate (DCI) o f the Ministry o f Foreign Affairs for external grants, the External Debt Unit (DDE) o f the MOF for external loans, andthe Directorate General o f Planning(DGP) o f the Ministry o f Economy for the PIP and the management o f the various projects. In reality, these departments operate separately and there is often no coordination between them. Hence, there is practically no follow up with donors and therefore there is little information on the execution o f external grants and projects. At the DDE, significant progress has been made recently to improve external debt management. However, it still lacks basic communication means (a telephone line and a computer linked to internet), which if made available would improve communication with donors and more specifically the timely payment o f the external debt service. The D C I does not have a mechanismwhich would allow the follow up o f external grant commitments with donors. The DGP ensures the management o f projects and the preparation o f the PIP, which allows determining the investment budget. 2.43 The government will need to take a number o f actions in order to facilitate the dialogue and engagement with the donors. First, the government should finalize the national poverty reduction strategy paper (PRSP) which represents the consolidation o f all sectoral programs and strategies for the medium-term, and will serve as ultimate dialogue instrument for engaging donors. Second, the government should establish a General Directorate who will be in charge o f ensuring coordination with donors, centralizing all external resources, andworking with the TC to ensure that these resources are effectively channeled through the budget (ideally, this body would be placed under the MOF and would comprise relevant staffs o f DCI, DDE and DGP). Finally, the government would need to pursue reforms aimed at improving public expenditure and financial accountability (PEFA) so as to increase confidence in the budgetary process, and gradually pave the way for a programmatic assistance approach in the medium-to- long-term. MACROECONOMIC IN2006ANDMEDIUM-TERMPROSPECTS POLICIES GovernmentPoliciesfor 2006: PursuingMacroeconomicandStructuringReforms 2.44 Continued economic recovery and improvedfiscal managementallowed the negotiationof a SMP with the IMFin early 2006. Followingthe presidentialelections, the government prepared a budget for 2006 and an economic program, which was approved by Parliament in March 2006. Against this background, the joint Bank-Fund mission that visited the country in March 2006 reached an agreement with the government to extend the S M P for the period April-December 2006 (see the quantitative andstructural indicator o fthe SMP inBox 2.5). - 22 - Box 2.5: Quantitativeand StructuralIndicatorsUnder the SMP in2006 Quantitativeindicators End-June End-Sept. End-Dec. (inbillionofCFA franc) Prog Prog Prog Domestic financing o fthe budget -6.1 -6.6 -11.8 Domestic primary budget balance -3.4 -6.0 -6.4 (commitment basis) Revenue 16.3 24.9 35.9 Wage bill 10.0 14.6 19.2 Accumulation o f domestic anears 0 0 0 External public debt, maturities <=lyear 0 0 0 External non-concessional public debt maturities >lyear 0 0 0 Prior Actions -- Parliamentary Target date approval o f organic laws o f Ministries End-April2006 Approval o f organic laws o f the Ministries by the Council o f Mid-May 2006 Ministers - Parliamentary approval o fthe government budget fully consistent with the Memorandumo fEconomic andFinancial Mid-May 2006 Policies for 2006 -StructuralIndicatorsrecruitment Launch a test-based programo fhigh level civil -- Eliminate servants End-May 2006 all discretionary tax exemptions Implement banderole system for alcoholic beverages and Continuous cigarettes -- Retrenchment End-June 2006 Implement military census o f 2834 civil servants and 1629 military from End-June 2006 the payroll -- Introduce End-June 2006 Enact the new electricity bill for the energy sector an automatic adjustment system for petroleum End-Sept. 2006 product prices - Bring End-Dec. 2006 3 State-owned companies inthe commercial and production sectors to the point o f sales End-Dec. 2006 m:IMF. 2.45 The objective of the SMP is to stabilize public finances and limit the accumulation of new arrears. The macroeconomic policy framework underlying the S M P aims at (i) sustaining the growth recovery that started in 2004; (ii) containing the inflationary pressure observed since last year, to preserve recent gains inreal per capita income; and (iii) enhancing poverty reduction by accelerating the pace o f social sector spending within the scope o f the existing budget. Inorder to achieve these objectives, the government is committed to pursuing prudent fiscal policies through the established cash flow management system, and avoiding the buildup o f new domestic arrears. Short-term fiscal policies will thus focus on avoiding the accumulation o f new domestic arrears over 2006 as a whole, within the framework of a day-to-day cash flow management system (see Table 2.7). In this regard, a circular has been issued by the Prime Minister to strengthen the mandate o f the TC. - 23 - Table 2.7: Treasury Cash-Flow Plan for 2006 (In millions of CFA francs) Jan. Fcb. Mar. Apr. May Jun. Jul. Aug. Sep. Oct. Nov. Dec. Total Inpercent Actual Actual Actual Actual Pmj. Pmj. Pmj. Pmj. Pmj. Pmj. Pmj. Pmj. Pmj. ofGDP Inflows 1,748 3,243 4,876 3,293 8,763 7,488 4,259 2,612 4,012 1,792 1,620 15,144 58,850 34.1 Revenueandgrmts 1,494 1,080 4,828 2,254 7,826 4,738 4,259 2,361 1,978 1,792 1,620 7,574 41,804 24.3 Revenue 1,494 1,080 2,528 2,254 4,218 4,738 4,259 2,361 1,978 1,792 1,620 7,574 35,896 20.8 Taxes 929 837 1,607 1,285 2,359 3,085 3,028 1,876 1,226 1,180 1,260 1,841 20,513 11.9 Nontaxrevenue 565 243 921 969 1,859 1,653 1,231 485 752 612 360 5,733 15,383 8.9 BudgetSupport 2,300 3,608 0 0 0 0 0 0 0 5,908 3.4 Bank financing 254 2,163 48 0 0 0 0 0 0 0 0 0 2,465 1.4 BCEAO(Treasuryaccount)(useofdeposits =+) 254 163 48 0 0 0 0 0 0 0 0 0 465 0 3 CommercialBank(BAO) 0 2,000 2,000 1.2 Other Financing 0 0 0 0 300 0 0 251 0 0 0 0 551 0.3 Privatization 0 0 0 0 3 0 0 0 0 2 5 1 0 0 0 0 551 0.3 Identifiedbudget support 0 0 0 1,039 637 2,750 0 0 2,034 0 0 7,571 14,030 8.1 outflows 4,056 4,106 4,313 10,011 4,540 4,042 4,511 3,838 3,855 4,517 3,832 7,479 59,100 34.3 CurrentRimary Expenditure 2,862 3,550 2,897 2,118 3,350 3,350 3,350 3,350 3,350 3,350 3,350 3,323 38,197 22.2 Wages andSalaries 1,838 1,940 1,688 1,528 1,527 1,527 1,527 1,527 1,527 1,527 1,527 1,500 19,182 11.1 Remuneration 1,613 1,656 1,465 1,366 1,334 1,334 1,334 1,334 1,334 1,334 1,334 1,307 16,743 9.7 Bonuses 70 88 71 36 39 39 39 39 39 39 39 39 578 0.3 Embassies -personnel spending 2 18 50 5 36 36 36 36 36 36 36 36 361 0.2 Food 153 178 102 I21 118 118 118 118 118 118 118 118 1,500 0.9 Goodsandsmices 494 547 584 118 484 484 484 484 484 484 484 484 5,613 3 3 of which: Embarsier I 18 45 2 35 35 35 35 35 35 35 35 346 0.2 Transfers 470 832 548 429 959 959 959 959 959 959 959 959 9,952 5 8 Socialsecurity(includepensions) 275 376 389 369 315 315 315 315 315 315 315 315 3,932 2.3 Othm 195 456 158 60 644 644 644 644 644 644 644 644 6,020 3.5 Other currentexpenditure 60 231 77 43 380 380 380 380 380 380 380 380 3,450 2.0 Memo: despesasnao tifuhdas (end- ofperiod) 399 Capitalexpenditure 241 324 114 74 400 400 400 400 400 400 400 541 4,097 2.4 PublicInvestment(domstically financed) 241 324 114 74 400 400 400 400 400 400 400 541 4,097 2.4 PIP 40 8 49 0 270 270 270 270 270 270 270 270 2,255 1.3 Capitalgoods 201 316 65 74 131 131 131 131 131 131 131 272 1,842 I 1 Paymentofdomesticmars andcomplimcntarypcriod 910 144 272 356 0 0 0 0 0 0 0 -1,182 500 0.3 ComplimentaryPeriod 910 272 -1,182 0 0.0 Currentyear 0 0 0 Previousto 2005 144 356 500 0.3 BankFinancing 0 0 1,006 1,349 707 82 718 0 81 652 0 688 5,282 3.1 BCEAO 0 0 245 817 0 82 718 0 81 652 0 540 3,134 1.8 Debt to BCEAO 0 0 93 94 0 0 94 0 0 96 0 60 437 0.3 BCEAO--othns(includesarrears-IMFaccount) 0 0 0 723 0 0 624 0 0 556 0 467 2,370 1.4 Interestondomstic debt 0 0 152 0 0 82 0 0 81 0 0 13 327 0.2 CommercialBank(BAO) 761 532 707 0 0 0 0 0 0 148 2,148 1.2 Other Financing 0 0 0 6,000 0 0 0 0 0 0 0 3,900 9,900 5.7 Treasurybills 6,000 6,000 3.5 RegionalCommercialBanks 3,900 3,900 2.3 Debt Smice to Multilaterals 43 88 24 I15 83 210 43 87 24 115 82 210 1,124 0.7 Float -2,308 -862 563 -6,718 4,223 3,446 -252 -1,226 156 -2,725 -2,213 7,665 -250 -0.1 IMF'r podsuthcritier' estimsla Earedonthe 2006 budget 2.46 The launching of the civil service reform and the pursuit of ongoing tax reforms are expected to improve the government's fiscal space. The high level of personnel expenditure leaves little room for social policies and economic growth. The government recognizes that the current level o f wage bill (75 percent o f revenues) is unsustainable and could endanger the fragile macroeconomic stability. Duringthe 1990s, Guinea-Bissau made progress inreducing excessive public sector staffing under a donor- financed retrenchment program. However, the program was not completed due to the conflict. Subsequently, hiring policies loosened after the war and total public employees (civilian and military) passed from about 17,900 in 1997 to an estimated 21,000 in 2003 and 19,00027in 2005.**The government is committed to launching the first phase of a "A civil service census was conducted in end-2004 and the final data were published in early 2005. The census revealed a total of nearly 12000civil servants. The size of the security is preliminarily estimated at 7,000 people (5,000 military and 2,000 police and gendarme). - 24 - civil service reform program in the second half o f 2006. The reform aims at retrenching about 2,800 civil servants among career or rank staff (categories 0 to Z)29.It is expected to lead to a net decrease inthe wage bill equivalent to about 1percentage point o f GDP. The overall impact o f the reform i s limited due the fact that the severance and other retirement packages to be paid to the retrenched workers will increase expenditure on transfers3oby an amount close to the salary decrease. With regard to tax reforms, the process of simplifying the tax system was started in2005 with the elimination o f several low-yielding taxes, the unification o f corporate income tax rates, andthe introduction o f a minimumtax for companies that are unable to provide a balance sheet. Moreover, inline with the recommendations o f the FIAS, the government reduced the effective marginal tax rate from 39 to 30 percent in2005, andhas further reduced it to 25 percent in200631. 2.47 The security sector reform(SSR) is expectedto start as soon as the census of the armed forces is completed. The SSR is a necessary condition for securing lasting peace andmoving ahead for other reform agenda. Suchreform would also have a positive impact on the public finances due to relatively high share o f security-related spending in the budget (as indicated earlier, security sector salaries are about half o f the total wage bill). Following a mission by the United Kingdom Security Sector Development Advisory team in October 2005, the government issued a decree creating an inter- ministerial committee on the SSR inFebruary 2006. Subsequently, a steering committee chaired by the Minister o f Defense was set up to orient the work o f a technical team and approve its proposal o f reforms before submitting them to the cabinet. The government intends to organize a donor Roundtable in the last quarter o f 2006 to seek donors' assistance for its overall development program including the SSR. Meanwhile, some donors have already pledged support to the program. In order to mobilize the much- needed support for the SSR, it is crucial to complete the census o f the military and paramilitarybefore the donor Roundtable inNovember 2006. 2.48 The government is committed to pursue efforts to strengthen the financial sector. Untilrecently, only one commercial bank (Le. BAO) was operating in Guinea- Bissau. In early 2006, two new banks established local agencies in Bissau: Bank de 1'Union (BDU) and the Regional Solidarity Bank (BRS). The BRS in particular is specialized in the financing o f micro-projects. It is expected that two other commercial banks from WAEMU will take advantage o f the single licensing law and open branches inBissau by end-2006. Prudential supervision of these banks will be provided through the BCEAO and WAEMU's Banking Commission. The banking system remains reasonably sound, although excessive credit concentration and continued short-term lending to a cash-scrapped government have contributed to raising the vulnerability o f ~ ~ 0 . 3 2 2.49 Efforts are being pursued to improve the business climate. The government prepared an action plan for simplifying procedures for trade and investment. The plan is The reduction of public sector employeesbetween 2003 and 2005 i s believed to have been the result of the introduction of more control and WesternUnion-basedwage paymentsstarting in2004. 29Categories0 to Z correspond to low-skill workers (about 30 percentof the payroll). 'OThe governmentplans to prepareaprogramof economic reinsertionfor the retrenchedpersonnel. The programwill include training inmicro-entrepreneurial activities,andprovidingastartingcapitalfor suchactivitieswith the supportofthedonor community. ''Thisratereduction is expectedto affect revenues only startingin2007. 32Nonetheless,the capitaladequacyratio (CAR) of BAO remainedwell above the 8 percentnormat end-March, 2006. - 25 - being implemented under the Bank-financed project for the Rehabilitation and Development o f the Private Sector (PRDSP). Recent initiatives aimed at removing obstacles to private investment include (i) a review o f the Investment Code, which is expected to be completed by June 2006; (ii) adhesion to MIGA and other FDI-guarantees agencies such as FAGACE; (iii)the introduction o f a one-stop shop for potential investors; (iv) the preparation o f a draft law on build-operate-transfer (BOT) procedures for projects requiring public/private partnerships; (v) elimination o f administrative barriers; (vi) elimination o f most license fees (alvara) requirements for the industrial and commercial sector, including for imports and exports; (vii) identification and elimination o f remaining constraints for private sector involvement in strategic sectors such as cashew nutproduction andprocessing, the fishing sector, andrice and fruit and vegetable production; and (viii) the simplification o f the fiscal system to promote private sector activity. In order to sustain these initiatives, the government will need to pursue the ongoing efforts aimed at: (a) bringing labor and commercial laws in line with the provisions of the Organization for the Harmonization o f Trade Legislation in Africa (OHADA); and (b) completing the pri~atization~~o f the remaining public enterprises. The implementationof these reforms will certainly contribute to attract new investments inthe country, particularly inthe boomingcashew processing business. Medium-termProspects:StrengtheningGovernanceandEnhancingFiscal SustainabilityandGrowthby 2009 2.50 Promoting good governance. The government believes that good governance will be achieved through political stability and the enhancement o f fiscal accountability. The UNsystem has already taken the lead inproviding assistance inpromoting political stability, supported by the Economic Community o f West African States (ECOWAS), the Community o f Portuguese Speaking Countries (CPLP), and other international partners. Key immediate issues are the SSR and capacity building, but also improving good governance and transparency, the lack o f which was a major factor behind the civil war. To enhance fiscal accountability, the government is committed to (i) rules and apply regulations in government operations; (ii)strengthen the political and democratic institutions, including at the local level; (iii) apply measures to increase the efficiency and effectiveness o f the government administration; and, (iv) strengthen the judiciary system; and (v) reinforce the means andindependence ofthe control departments to allow a better control o f budget execution. Moreover, in order to improve transparency in public finances, the MOF is committed to continue publishing monthly summary fiscal tables, which started in July 2005. More generally, the National Committee for Social Cooperation-which is chaired by the Prime Minister and includes trade unions and employer organizations-was revived, and will meet regularly to discuss economic developments andreform policies. 2.51 Promotingregionalintegration. The promotion ofregional integration is one o f the central elements o f the government's economic policy. The accession to WAEMU in 1997 was shortly followed by the civil war, and Guinea-Bissau made little progress in bringing macroeconomic indicators in line with those o f the region. Nevertheless, the 33Startingwith the 2006 budget, the government plans to privatize some 14 companies with total market value of CFAF2.6 billion includingtwo hotels.The process is expectedto be carried out ina transparentmanner, usingpublicbiddingprocedures. - 26 - government remains committed to the implementation o f policies required to meet the objectives under WAEMU's Convergence, Growth and Solidarity Pact. Given its prolonged economic crisis, Guinea-Bissau was unable to respect none o f the eight convergence criteria set byWAEMU at end-2005 (see Annex 16). Little progress has also been achieved with respect to the other regional initiatives. While the common external tariffwas adopted in2000, the country continues to avail itselfo fthe safeguardmeasures allowed under the CET. The five directives on the public finances andthe directive on the harmonization o f indirect taxes have still not been applied. However, the government intends to request technical assistance intax administration, with a view to introducing a single rate-VAT in2007. 2.52 Promoting growth and fiscal sustainability. Growth and fiscal performance is expected to slightly improve in the medium-term. The real GDP growth rate is expected to improve moderately to reach an average rate o f 5 percent in the medium-term. The main assumptions are that improved governance and convergence toward regional convergence criteria, and the pursuit o f the reconstruction effort in key areas such as infrastructure and rural de~elopment'~ will enhance growth in the private sector. Inthe fiscal area, although the authorities do not have a medium-term expenditure framework, it is expected that the government's fiscal space will improve as a result o f a decrease inthe wage bill resulting from the public service reform. As a result, the fiscal primarydeficit i s on average expected to decrease from its current level o f more than 5 percent o f GDP to below 2 percent during2007-09, andimprove progressively beyond 2009. 2.53 PRSP, Emergency Post-Conflict Assistance (EPCA), PRGF and HIPC completion point. Provided that the PRSP is finalized in the second half o f 2006, and donors pledge the needed resources at the November 2006 Roundtable, the implementation o f the government's poverty reduction strategy is expected to take place during the period 2007-09. The reform program o f the PRSP -- enhancing governance, accelerating economic growth, and improving service delivery mainly for the most vulnerable -- provides an adequate framework for addressing the country's current challenge o f ensuring a smooth transition from post-conflict to long-term development. The current SMP i s expected to pave the way for a Fund-supported EPCA by end-2006. Subsequently, satisfactory implementation o f the PRSP and EPCA will herald an eventual Fund-supported PRGF and the completion point o f the HIPC by 2009. This would eventually make it possible for Guinea-Bissau to benefit from the Multilateral Debt ReliefInitiative (MDRI). CONCLUSIONOFPOLICYRECOMMENDATIONS 2.54 Yet as argued inthis chapter, the high level o f personnel expenditure leaves little room for social policies and economic growth. In order to improve the government's fiscal space, the launching o f the civil service reform and the pursuit o f ongoing tax reforms and the pursuit o f sound fiscal management under the established cash flow management system will be required. For this reason the government would need to implement the following actions: 34The Bank i s currently preparing a multi-sector infrastructure credit , and will support a government community development programinFY07.The EuropeanUnionis alsosupportingvarious infrastructuredevelopmentprogram. - 27 - Inthe comingthree months: e Finalize the PRSP e Strengthen the mandate o f the TC by enforcing the implementation o fthe circular issued bythe Prime Minister inthat effect. e Before the liquidation o f expenditures, ensure the control o f the legality, regularity, andmorality o f expenditure by the DCF. e Close all government accounts that are not under the control o f the MOFs, and centralize all payment operations at the DGT bank account held at BCEAO. e Impose by the law that all withdrawals andpayments made from the special accounts (fisheries, road fbd, etc.) be subject to the approval o fthe Minister o f Finance e Close all the individual bank accounts held bythe Ministries at BCEAO (except the special accounts) and centralize all deposits andpayments at the account(s) o f the centraltreasury at BCEAO. e Enhance transparency inthe management o f exemptions by: (i) recording exemptions as part o f customs revenue andregistering the counterpart tax spending inthe budget; (ii) andprohibiting the issuance o f exemptions on a discretionary basis. e Complete the census o f the militaryandparamilitarybefore the Roundtable. Inthe coming12 months: e Pursue prudent fiscal policies through the established cash flow management system, and avoid the buildup o f new domestic arrears. e Increase the level of scrutiny o f the T C so as to reduce the DNT to the minimum agreed upon inthe context o fthe SMP by ensuring that: (i) The expenditures submittedfor clearance under the DNTprocedure are consistent with budgetary laws andprocesses; (ii) The T C coordinates with the Budget Department andthe Treasury to regulatethe flow o f expenditure commitment andpayments; (iii) Temporarily stop new commitments when there is evidence o f a highrisk o f expenditure backlog due to liquidityproblems. e Improve the control o f budget execution by: (i) aligningtheregulatoryframeworkgoverningthefinancialcontrolsystem with the norms inWAEMU; - 28 - (ii) training the financial controllers andproviding them the needed incentives to carry out their duties; (iii) providing more autonomy to the DCF by establishing it as an independent Directorate withinthe MOF; 1 (iv) allowing the IGFto carry out routine as well as unexpected control visits; (v) andproviding adequate training to the M P s to allow them to better understand andcontrol, government accounts. Improve the management o fpublic procurements by: updating the hardware and software installed at the DGCP duringthe pilot procurementphase; reinforce the role o fregulation devoted to the CNCS and ensure better coordinationwith the DGCP; redefining the role andobjective o fthe CNCS interms o f designing procurementpolicies, organizingtraining programs, maintainingthe information andcommunication management system incoordination with the DGCP, andconducting audits; reinforcing the capacity o fDAFs inprocurement management; enhancing the partnership between the DGCP andthe private sector through the organization o fregular information and counseling seminars; posting o fpublic tenders inlocal newspapers andmedia on a regular basis; enforcing mandatory annual audits on at least 15 percent o f publicly procured contracts for which the costs are above a threshold predetermined bythe finance law; andextendingthe new procurement system infive additionalministries to include inpriority defense and security, economy, and fisheries. Enhance aid coordination by: (i) implementing the PRSP, which will serve as ultimate dialogue instrument for engaging donors; (ii) andestablishing a GeneralDirectorateto beincharge o f ensuringaid coordination. Alignlabor andcommercial lawswith OHADAnorms. - 29 - 3. IMPROVINGTHE FISCAL SITUATIONTHROUGH DOWNSIZINGTHE CIVIL SERVICE" Guinea-Bissau 's public sector wage bill increased rapidly in thepast years. The rise was largely due to an increase in nominal wagesfirst for militay personnel, and thenfor skilled workers of the categories A to N. By contrast, thepayroll of the civil service decreased between 1997 and 2005 as evidenced by the data fiom civil censuses carried out in those years. The decrease is in part attributable to the tightening of the control of salay payments, which may have helped to clean the payroll of `ghost workers'. Thefiscal dificulties facing the country today are largely attributable to the country's huge wage burden. I n order to address the issue the government, in preparing the 2006 budget, has designed a retrenchment reform program aimed at discharging more than 2,800 low skilled workers in a first phase and little more than 1,600 military in a second round. This chapter is an attempt to estimate the potential cost of eventual separation packages for the low skilled civil servants. The estimates are based on the DOSE model and suggest that the average compensation benefit per satisfied worker could be between CFAF 1.6 and 2.4 million, which translates into a one-time total program cost of CFAF5 to CFAF6 billion (3 to 4 percent of GDP). On the other hand the gross annual salay saving fiom the downsizing reform is expected to be about CFAI.15 billion (0.72 percent of GDP). However, a major political caveat still remains: while appropriate compensation is needed, the ideal package will have to be negotiated withpublic sector trade unions on the basis of compromises and the bargaining power of all the players involved. From this perspective, the simulation approach proposed in this chapter should be seen as a tool to introduce some economic rationality in delicatepolitical negotiations. BRIEFDIAGNOSTIC THECIVIL SERVICE OF 3.1 The public sector wage bill I - 160%, Figure3.1: Evolutionofthe PublicSector Wage Bill, 1997-2005 has been T l6''% Wages/taxes (left axis) increasing since 140%- -Wageshevenues(left __ 14.0% axis) 1997, In nominal 120%- - .. ,WagesIGDP (right axish __ 12.0% terms, the average annual salary for 100%- __ 10.0% all workers 80%- __ 8.0% increased from FCFA 382,480 in 60%- __ 6.0% 1997 to FCFA 40%- __ 4.0% 594,839 in 2005, which represents 20%- __ 2.0% an increase in 0% I I 0.0% I I I I I nominal terms o f , I , I I I , I 56 percent 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 between the two Source:Barry and Wodon (2006), usingGuinea-BissauLive DataBase (LDB), World Bank. years. At the aggregate level and relative to total output, the wage bill has increased from 3 percent of ~~ 35This chapter is based on Chong et al., 2006 (which i s an updated version of Chong and Rama, 2001. The paper i s published in Wodon, Q., editor, 2006, Public Wages and Retrenchment in Guinea-Bissau, mimeo, World Bank, Washington, DC. It is worth mentioning that the findings of the analysisare intendedto help inthe policy dialogue aroundthe government's ongoingbroadercivil reformprogram. The governmenthas designed a retrenchmentprogram aiming at ensuringthe training of the retrenchedworkers and their subsequent reinsertion in the privatesector.The team understoodthat, inlight ofprevious experiences, the government does not intendto providecashcompensationpackagesto retrenchedworkers. The underlyingjustification to this decisionis that cash transfers would most likely be used for consumptionrather than investmentpurposes. These issues are beyond the PER scope. Therefore, the analysis will not elaborate on them. Rather, the analysis will focus on the benefit-cost analysis underlying the process in order to provide useful inputto the government's retrenchmentprogram. - 30 - GDP in 1997 to 13 percent in2005 (see Figure 3.1). Over the same period, the inflation rate decreased from 49 percent in 1997 to 8 percent in 1998, after which inflation remained low. Inreal terms, the average annual public sector wage increased by nearly 6 percent per year over the nine years between 1997 and 200536,although with differences between different categories o f employees. Thus, on average, public servants benefited from an increase in their average wages, while the well-being o f the population as a whole decreased duringthe same periodby about 33 percentage points. 3.2 There are more workers in the public sector today than there are in the privateformal sector, and for every worker inthe private formal sector, there are close to 15 workers who are self-employed (see Table 3.1). The capacity o f the private formal sector to absorb retrenched workers from the public sector could be limited, unless economic growth is stronger and the State divests from activities that could be carried out by the private sector. Especially ifone were to focus on unskilled public sector workers, perhaps at most one-fifth o f the separated workers could be hired by the new private contractors in charge o f the cleaning, surveillance, and maintenance of government facilities. These contractors will be under close scrutiny to comply with labor market regulations. The rest o f the separated workers, on the other hand, will most likely end up inthe informal sector, andget noperquisites or benefits attached to theirjobs. Table3.1: Structureof employmentin2002 Private Private Public Formal Informal Total Agriculture 314 1,511 206,316 208,140 Industry 133 1,087 30,281 31,501 Construction 259 4,382 9,321 13,962 Transport 68 1,911 1,935 3,914 Commerce 67 1,434 38,190 39,690 Services 1,203 9,715 6,676 17,594 Education 861 773 276 1,910 Health 621 180 378 1,179 Public Administration * 21,916* 600 357 22,874 Total 25,442 21,593 293,730 340,764 m:Wodon, * Civilians Creppy and Tsimpo (2006a), basedon estimationusingthe 2002 HouseholdSurvey. and military 3.3 Skilled workers benefited from the wage increase more than their unskilled counterparts.There were large differences interms o fwage trends between skilled and less skilled workers. As shown in Figure 3.2, the average wages for skilled workers categories increased much more than those for less skilled workers categories. The cumulative average increase inreal terms for the skilled workers categories (categories A to N, see Box 3.1) reached 93 percent, while the corresponding value for unskilled workers (0 to Z) was only 38 percent (which is only about half the increase enjoyed by the skilled categorie~).~~ Takinginto account boththe number o fworkers andtheir wages, the total annual wage which could be identified in the 1997 census amounted to 4.2 billion CFAF. In 2005, the corresponding value was 6.1 billion CFAF. This a nominal 36The cumulative inflation from 1997 to 2005 was 93.5 percentwhile the average real GDP growth rate was -33 percent. 37The average nominal wage increasefor all categories was 143percent. - 31 - increase o f about half, but ifwe consider the highinflation rate in 1997, andfully account for this high inflation in our comparison o f the data for the two years, it implies that in real terms the expenditures for public servants increased only slightly. That is, the increase inpay was compensated inpart by a decrease inthe number o f civil servants. In total the salaries o f workers in levels 0 to Z amount to 1.4 billion CFAF in the 2005 sample o f 10,248 workers, o f which 3,439 are in the 0 to Z categories (this number for less skilled workers i s similar to the estimate advanced in policy discussions in the country). Box 3.1: AdministrativeCategories inthe Civil Service The categories A to Z corresponding to administrative positions as follows: ........ Level A: General Inspector (Inspecteur General) LevelB : General Director (Directeur General) LevelC: FirstRank Director (Directeur de Service Premiere classe) 0 LevelD: SecondRank Director (Directeur de Service Deuxitme classe) LevelE: ThirdRank Director (Directeur de Service Troisieme classe) 0 Level F: Division Chief (Emploi de Chef de Rtpartition) Levels G to H:Technical position, superior level (emploi de technicien suptrieur) Level I: Technical position, mid-level (emploi de technicien moyen) Levels J to N:Technical position, basic level (emploi de technicien de base) Levels 0 to Z: Manual worker such as guards, drivers, cleaning and maintenancepersonnel, l o w skill workers, etc. Source:2005 civil service census, MOF. 3.4 Despite the increase in the wage bill, the comparison of the data from the civil service censuses of 1997 and 2005 indicate a decrease in the civil service payroll. In 1997, a census o f the civil service allowed identifying a total o f 15,963 civil servants. Another census in 2005 established this number at around 12,000 (the 10,258 figure reported in Table 3.2 and throughout this chapter i s obtained after cleaning for missing data). This reduction was inpart due to increased control over the pay system, which allegedly allowing cleaning the payroll o f ghost workers. 3.5 The decrease of the payroll seems to have been caused by a reduction in skilled workers. Further analysis o f the 2005 census data suggests that the decrease in the number o f public servants can be attributed almost entirely to a reduction in the number o f skilled workers (those in the categories A to N) from 13,467 to 6,819. By contrast, the number o f less skilled workers in the categories 0 to Z decreased only slightly from 2,496 to 3,439. This would suggest that as a proportion o f the total numbers o f workers, unskilled workers may be too numerous today, which is one o f the arguments advanced for their retrenchment. - 32 - Table3.2: CivilServantsby CategoryandWage 3.6 The wage Numberof Mean increase for skilled Civil servants Wage workers relative to 1997 2005 1997 2005 their unskilled A 11 5 1,183,556.0 1,353,600.0 B 301 160 769,242.8 1,894,500.0 counterpart C 267 108 611,109.6 1,092,800.0 contributed to increase D 169 45 562,084.0 1,303,680.0 wage inequality. As E 534 81 526,504.8 1 106844,O mentioned in the F 660 412 458,459.1 1,007,930.0 G 1,174 98 412,042.2 1,239,869.0 previous section, the H 1,124 445 398,640.4 1,083,398.0 increase in wages I 1,372 728 374,405.6 697 134,l between 1997 and 2005 J 1,360 984 376,145.5 599,092.7 was larger for skilled K 1,720 396 359,327.1 544,000.0 L 889 562 363,205.7 656,301.8 workers than for less M 1,424 1,124 364,591.5 493,135.9 skilled workers. As a N 1,349 1,179 336,803.9 383,316.0 result, wage inequality 0 59 295 334,194.2 557,060.3 P 81 59 332,661.1 533,532.2 among public servants Q 162 282 331,179.0 504 102,l increased over time. R 301 71 329,011.6 286,377.5 Table 3.3 provides a S 212 241 324,010.3 725,616.6 range of inequality T 301 226 321,232.3 394,279.6 U 357 203 317,732.4 457,111.3 measures, all of which V 703 117 316,606.1 308,923.1 show a large increase in W 217 315,022.7 284,151.2 inequality. For example, X 229 767 447,432.6 the Gini index, which as Y 91 64 313,238.8 414,900.0 Z 1,113 897 311,037.4 258,525.8 the other measures of A-Z 15,963 10,258 382,480.0 594,838.8 inequality takes a value A-N* 13,467 6,819(-49%) 506,865.6 970,968.6(92%) between zero and one, 0-Z* 2,496 3,439(+39%) 322,356.9 424,355.5(32%) Source:Wodon, Creppy and Tsimpo (2006a) based on the 1997 and2005 civil censuses increased from 0,0986 in * Figuresfor the number of civil servantsare totals while those for meanwages are group 1997 to 0,3639 in 2005 average(the figures inparentheses are growthrates from 1997 to 2005). (see Box 3.2). Table 3.3: InequalityinPublicWages, 1997 and 2005 1997 2005 Relative mean deviation 0.0701 0.2654 Coefficient ofvariation 0.2775 0.9067 b Standarddeviation of logs 0.1825 0.5982 Ginicoefficient 0.0986 0.3639 Mehran measure 0.1275 0.4664 Pieschmeasure 0.0841 0.3127 Kakwani measure 0.0127 0.1193 Theil entropy measure 0.0262 0.2571 Theil meanlogdeviation 0.0207 0.2155 Source: Wodon, Creppy and Tsimpo (2006a) based on 1997 and 2005 censuses of civil servants. - 33 - Box 3.2: Wage Inequality:the GINI Coefficientandthe LorenzCurve A Gini index of zero implies that there is complete equality in wages, with all civil servants having exactly the same wage. A Gini index of one impliesthat there is complete inequality, i.e. one civil servant has all the wage income, and the others have no income at all. Graphically, the Gini index is relatedto the Lorenz curve representedin the figure below. To construct the Lorenz curve, the horizontal axis gives the cumulative share of the civil servants population rankedby increasinglevel ofwage. The interval0.0-0.1 correspondsto thebottomwage decile, while the interval 0.9- 1.0 correspondsto the top wage decile. The vertical axis representsthe share of total public servant wages enjoyedby the correspondingpercentageof the civil servant population. It canbe seen for example that the bottom 20 percent of civil servants have about 5 percent of the total wage income in the census sample. The Lorenz curves of wage inequality goes through the points (0,O) and (100,100). Perfect equality is representedby the diagonal. The Lorenz curve i s always below the diagonal. A Lorenz curve further away from the diagonalindicatesa higher level of wage inequality.A curve going through the points (O,O), (100,O) and (100,100) would represent perfect inequality, with one public servant havingall of the wage incomeinthe sample. The Gini coefficient can be shown to be equal to twice the area betweenthe diagonaland the Lorenz curve. The fact that the Lorenz curve for wage inequality in 2005 is further away fromthe diagonalthan the correspondingcurve for 1997confirmsthe increase ininequality (actually, becausethe two Lorenzcurvesdo not intersect,we would observean increase ininequalitymeasures, not only those shown intable 3.3). Another way to look at the changes inwages is to graphthe growth incidence curves for wages between 1997 and 2005 (Ravallion and Chen, 2003). The curve providesthe growth rates inreal wages at various levels of wages, from the lowest wages to the highest wages. The upward trend of the Lorenz curve implies that the increase in wages was muchlower at the bottomofthe distribution than inthe upperpart of the distribution, confirming that wage growthhas been stronger among better off civil servants. In other words, the proportional gains in wages were much higher for better paidworkers than for civil servantswho are poorer. 1 0.9 Lorenz Curve for Wages (1007 and 2005) 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 0.1 0.2 0.3- 0.4 0.5 0.6 0.7 0.8 0.9 1 -Year 1997 -Equity -Year2005 m: Wodon, Creppy and Tsimpo (2006a) - 3 4 - ng in a pubfic sec r retrenchment p gram is to improve cyt that is, to reduce the size of ~ o ~ ' s ~~ e~ n white ~ ~~ n ~~ ~i t es. When h ~ d l e do p t ~ ~ ~d ul wl ~ s,~ z i nwill yield pos~t~vercsults g since it allows a rcallucat~onof workers across sectors with a net gain in pro s ~ e c ~ ~ c adl lo~ ,~ n ~wilX~bei ad~risa~le the u p - ~ o cost ~enera~eds ~ n ~ if ~ t i cd hment program, or more Y, if shouldbe u n d e ~ ~The~ratio o f theoften used to evaluate Pub e . public sector wage bill as a percent of GDP for instance is reco~mendedto be less than 10 percent in ~ e ~ e l ocountries. A higher level might be ~ i n ~ ~ ~ d ~ c aoft ian eo ~ e r s t ~ f f epublic service or thc e ~ ~ to which the state is involved in the ~ d e n ~ e c u ~ o ~~y .o t ~~epr p ~ o ~ ~ a ~ e ent of the total public s to bill, can p r o ~ ~also usefd hints, As a rule ofthumb, it is rec d g rat e ~ c e c d ~ n ~ either denin^ in^ fiscal deficit or ~ u b ~ t isalaries ~ 25 percent. Above this thr~shold,the g o ~ e ~ e n ~ t ~ ~ andnother current: e ~ p e n d i withrcapital ~ ~ ~ ~ e ~ ~ e ~ thereby~ hampering~ e c o ~ o ~growth prospects, These ~ ~ d i c a only d i ~ ~ e ~ c t o ~ ~ewe as the ~ f f o r ~of ~the~~ i ol ~~~ ?~wage bill and advise on the o ~ p ~OF~ ~ n i ~ ~ e ~ ~ ~ n ~ fhc 3.8 ~ ~ ~ ~ ~ ~ ~~ b-sector cwage bill had uswollensout of control over the last l ~~ ~ ~ s a ~ decades due to the c o ~ ~ effect of an M n c o ~ ~ ~ oriseein the ~ ~ r n of civil s e ~ ~ a n ~ ~ ~ n e ~ ~ l d b e ~ and r n i l i and ~ ~ ~ ~ harp increase in n o ~ i salaries. ~ a ~ increase in the number of civil fy from the ~~~~~~~g the public a d ~ 1 ~ n ~ s tinathe~ o ~ ~ t ~ o s t ~ c ~recons~ctioneffort. The evidence from thc n ~ ~ c t c e ~ s ~ $ carried out between e $ 1997and 2005 ~nd~~atedthe wage bill as a percent o f the GDP grew steaditY from 3 p e r ~ e n ~ that nt in~ ~Thc~wage5 bill,as percent o f total expenditure, which tii`as a r o ~ ~ d .8~ o ~ e ~more r , 75 p ~ e ~~~~ s salaries payment since 1997. T u n s u s t a ~ ~ ~andl ecould cndanger the fragile m a c r o ~ c ~ n o n ~s itca b ~ ~In~ this context, the ~ b i ~ . ~ o v e ~ i sn~~n ce r~c a~~ ir~~~~~~ n ~ l ~ on external and ~ o ~ ~ s~ t i ~c ~in orderito dose~budget c n gaps* rm ~ ~ vvill obe ~ r ~ ~ ~ a r gbotht~ ~ i~ ~l and ~ ~ civil servants. As m ~ n t i o n ~ d ~ a ~ earlier, the laxity o f the r e c ~ i tprocedure ~ ~ ~ t in the m y weak ~ o o f r c t ~~r ~ ~ e n t ~, s ~ o ~ ~ ~ in the public semicc have led to labor r ~ d u ~ which come at~ the expense of ~ a ~ ~ ~ e ~ taxes or tower social e ~ p e ~ d ~ tsooners or Inter, ~ r e rget only workers in the 0-Z ofthem, It i s worth obvious that un in dant (see Figure 3.2 >.Woreoover^ ~ ~ d and "ghost"d workers are p r o b a ~hidden in at1 u ~ ~ ~ l ~ categories. The reason that this phise focuses only on u ~ s k ~ l ~ oerdk e r sstre ~ h r ~ e f oFirst, ~ d , since the ~ ~ ~ c o n ~ ~~n~reasethe number o f public s e ~ a n ~the o l l e din s , nun1 ofunskilled has nof "35 - been reduced significantly, while the number o f skilled has been halved, from 12,000 in 1997 to approximately 6,000 in 2005. Second, the government is planning to contract out non-public services that could be better executed by private firms. Such services, which correspond to activities exercised by workers in the categories 0-Z, include cleaning, maintenance and surveillance o f government facilities. It is generally advisable that the central government focuses its efforts on the core public sector activities that have high social value, such as implementing sound economic policies and delivering basic health and education. Finally downsizing the unskilled workers i s the least costly politically relatively to their skilled counterpart. The unstable political and social environment leaves little room to engage in a comprehensive retrenchment reform. The second phase o f the reform will target the military. The rationale behind military retrenchment i s to keep both their wage bill and their number at sustainable levels. Indeed, the increase (8- fold) inthe salary o f the militarythat occurred in2003 contributed to significantly explode the public sector wage bill. The government i s planning to retrench 1,600 militarypersonnel. 3.10 Compensationpackagesfor separatedpublic sector workers are not new.They have been used by governments all around the world, often with support from donor countries and multilateral organizations. Workers have sometimes, however, been compensated better than "just right" -- actually, far better than"just right." A survey o fthe cross-country experience with public sector retrenchment in developing countries and transition economies, estimated at U S D 2,400 the average spending perjob separation3*.However, the figure was as highas U S D 13,000 inthe civil service of Senegal, USD 16,000 inthe mining sector of Bolivia, andUSD 17,000 in the public enterprises o f India, despite the fact that all three countries had a per capita income o f less thanU S D 1,000 per year Chong and Rama, 2001). 3.11 Separation packages are often estimated using quantitative tools in order to avoid unfairness and overspending. When practical tools to design compensation packages are lacking, past mistakes translate into present overspending, as the excessive figures or formulas o f previous downsizing episodes become a reference for new public sector restructuring endeavors. Ingeneral, governments overspend on separation packages for the purpose ofpolicy credibility. Ifthe compensation offered were too low, workers would not be compelled to leave, and the political costs o f such a failure could be high. Given that the package that elicits indifference on the worker (that is, the "just right" amount) i s not directly observable, policymakers may prefer to err in the direction o f overcompensation. Another reason why governments may pay more than the `Ijust right" amount is that they usually face tight constraints on their current expenditures, but may all o f a sudden have access to abundant resources for public sector restructuring from donor countries and multilateral organizations. This combination o f a hard- budget constraint for salaries and a soft-budget constraint for separation packages creates an incentive to compensate better than "just right". Overspending in compensation packages i s questionable, both on economic and on fairness grounds. From an economic point o f view, every transfer entails a cost. Sooner or later, taxes will have to be raised, or other government expenditures will have to be cut, to make up for the compensation o f separated public sector workers. The larger this compensation, the heavier the burdenon the rest o f society. Inregard to fairness, it is important to keep inmind that public sector workers are usually not poor, at least compared to their fellow citizens working in the informal sector or living in rural areas. A 38Haltiwanger and Sin& (1999) - 36 - government committed to poverty alleviation could therefore find better uses for its resources thanoffering far better than"just right" compensation to public sector workers. 3.12 In the case of Guinea-Bissau, civil service reformthrough separationpackagesis an issue that has been around for a while. About ten years ago, in the preparation for its third structural adjustment program, the government o f Guinea-Bissau had considered suppressing all positions in occupational categories "0" to "z", which correspond to unskilled personnel working on services like cleaning, maintenance and surveillance. The objective was to subcontract these activities to private firms, to be selected through competitive bidding. According to the authorities, the bidding mechanism was to be designed so as to favor those firms that would hire the largest number o f separated government emp10yees~~. It was estimated that roughly one fifth o f the separated employees could be hired by the new private contractors. The implementation o f this program had to be postponed due to the rebellion o f part o f the military against the government, which led to the 1998 armed conflict. This rebellion was not prompted by, or related to, the structural adjustment program in preparation. Today, the retrenchment o f government employees is regaining prominence inthe economic policy agenda. The chapter thus focuses on the appropriate level o f compensation for those employees. As Guinea-Bissau counts among the poorest countries on earth, it is important to base any retrenchment program on a credible assessment o f the losses in earnings and benefits that government employees can expect as a result o f separation. It is therefore necessary to evaluate the loss in earnings and benefits o f a representative sample o f government employees in the event o fjob separation based on a carefid data analysis. ESTIMATING COSTmENEFIT OFDOWNSIZING CIVILSERVICE THE THE Social Cost/Benefit 3.13 This analysis seeks to evaluate separation benefits based on the estimated loss that the separated public sector worker would experience. The following discussion is based on the DOSE software to estimate compensation benefits as part o f the policy dialogue on a civil service reform in Guinea-Bissau based on the 2002 household survey and the 2005 civil census data. This work is an update by Chong, Rama, Creppy andWodon (2006) o f an earlier attempt to determine the "right" amount o f compensation for redundant civil servants (Chong and Rama, 2001; see Box 3.3 for a discussion the key findings o f the paper, and more details on the DOSE methodology inBox 3.4). 3.14 Predictedearnings under the 2002 data suggest that in the event of a retrenchment some of the discharged public servants are likely to earn less in the private sector4'.With predicted earnings based on the 2002 household survey if all workers were to work inthe private sector 58 percent o f them would experience a loss in earnings. Ifall public sector workers were to work in the formal private sector, 38 percent would suffer from an immediate wage loss according to the predictedwages. The fact that a smaller share o f "0 to Z" workers would suffer from a loss in 2005 than in 1997 is not surprising given that less skilled public sector workers suffered from a relative loss inwages betweenthe two years as compared to better skilled public sector workers. Inother words, it is likely that the wage advantage o f less skilled public sector 39Suchpracticeis potentiallyharmfulto privateinvestmentand notadvisableas it mayput furtherconstrainton investors. 40The survey data was scaled up usingthe inflationratebetween2002 and 2005. - 37 - workers versus private sector workers was reduced between 1997 and 2005, which inturnwould implythat the likelihood o f suffering from an immediate loss inwages may have decreased over time (this does not mean that the total loss o fpublic sector workers could not be large, giventhat this total loss also includes losses innon-wage benefits such as pensions, as well as a loss injob security). It would be unrealistic, however, to assume that all separated workers will get formal sector jobs. Especially if one were to focus on unskilled public sector workers, perhaps at most one-fifth o f the separated workers could be hired by the new private contractors in charge o f the cleaning, surveillance, and maintenance o f government facilities. These contractors will be under close scrutiny to comply with labor market regulations. The rest o f the separated workers, on the other hand, will most likely end up in the informal sector, and get no perquisites or benefits attached to theirjobs. - 38 - Box3.3: SummaryFindingsofthe CompensationBenefit Study basedon 1997 data Chong and R a m (2001) suggests that public sector employees would value the benefits associated with their jobs, including old-age income security, at roughly one-fifth o f their cash earnings. Some government employees would experience a very small loss in the event o f separation (direct approach), while roughly 85 percent of the employees would face a total loss in excess o f CFAF400,OOO (indirect approach). Based on the direct and indirect approaches, the table below reports the cost and fairness o f four packages A, B, C and D), for nine satisfactionrates ranging from 10 to 90 percent. The zero satisfactionrate i s uninteresting, because it would require no compensation. At the other end, the 100 percent satisfaction rate can only be reached at unreasonably highseparation costs, because the package has to be generous enough to match the highestloss inearnings and benefits throughout the sample, and this is often an outlier. The only difference between the direct and indirect approaches concerns the loss indicator used. The figures inbold inthe table indicate the cheapest package, and the fairest package, for each satisfaction rate. Inthe case o f Guinea-Bissau, packages A and B are quite similar. This i s because o f the very small variation inearnings across employees in categories "0" to "Z". If all workers have roughly the same salary in government, a package defined as a multiple o f this salary is not too different from a lump-sumpayment. Whereas salaries ingovernment are roughly the same for all employees incategories "0" to "Z", losses from separation are not. As the correlation coefficient between a constant and a variable is zero, packages A and B perform very poorly in terms o f fairness. However, they are cost-effective at low satisfaction rates. If downsizing can be carried out without fully compensating more than one-third o f government employees, "flat" packages o f this sort could keep downsizing expenditures under control. At the other end, "tailored" Compensation, as inpackage D, is the most cost-effective alternative if satisfaction rates need to be high. Packages C and D have a similar performance at low satisfaction rates. The average cost per separated worker i s slightly higher for package C, but the fairness coefficient can be higher as well. Inorder to attain high satisfactionrates, however, the number o f months o f salary per year o f service requiredby package C becomes too high. At a 70 percent satisfaction rate, package C costs roughly twice as much as package D, but the latter does much better in terms o f fairness. The gap between the two packages becomes dramatic at very high satisfaction rates. In fact, "tailored" compensation dominates all the alternatives, both on grounds o f cost and fairness, if a majority o f the workers need to be "bought off." The compensation formula that the government o f Guinea-Bissau was planning to use for its public sector downsizing operation, which is a special case o f package C, turns out to be quite onerous. This formula would lead to a satisfaction rate o f 78 percent, based on the direct approach, or 73 percent, based on the indirect approach. The average cost per satisfied worker would be roughly CFAF 2 million, and the fairness coefficient would be around 45 percent. Better outcomes could be attained spending less than half. A lump sum o f less than CFAF 1million would achieve the same satisfaction rate o f the formula considered in Guinea-Bissau. A similar result, at a similar cost, would be obtained offering less than seven years o f salary as compensation. Package D would be even less expensive, as it would cost roughly CFAF 880,000 per satisfied worker. The fairness coefficient would, inthis case, be close to 70 percent. Averagecompensation per Separatedemployee (thousands of CFAF) 1/ Direct Approach Indirect Approach % of employee satisfied Alternativepackages Alternativepackages A B C D A B C D 10 176.0 138.8 227.1 232.7 358.0 361.0 441.0 412.7 20 300.5 300.4 451.5 397.8 437.6 437.6 636.0 556.7 30 468.8 464.3 660.6 465.2 513.0 514.0 813.5 620.6 40 632.5 633.9 818.8 567.2 624.2 618.7 989.7 650.4 50 722.0 724.3 964.3 667.0 760.0 762.4 1,188.3 705.5 60 818.0 816.2 1,157.3 736.7 843.0 841.6 1,446.7 783.0 70 939.5 944.2 1,464.0 809.1 932.0 930.6 1,862.8 861 80 1,007.4 1,014.4 2,143.6 871.8 1,065.5 1,064.9 2,759.8 940.1 m:Chong 90 1,106.3 1,107.2 3,642.0 927.3 1,193.0 1,194.8 4,806.3 1,052.8 andRama (2001), Tables 5 and 6. 1/ Package A is a variable lump sum; package B is a variable multiple o f the salary ingovernment; package C is a variable multiple of the salary multiplied by the number o f years o f service; package D is a variable lump sum, plus half a month o f salary per year o f service, minus CFAF 50,000 per year o f completed education, plus CFAF 60,000 for female employees, plus CFAF 150,000 for employees who work out o f Bissau The variable component o f each package is increased untilthe percentage o f satisfied employees indicated in the first column o f the table is reached. The figures in bold indicate the cheapest and fairest package for each satisfaction rate. Box 3.4: Methodology of the DOSE Model Overview of the model.The DOSE is based on two approaches. The direct approach measures the total loss that the separated public sector worker would experience based on the present values o f the earnings loss the worker would experience in all the years since displacement until retirement, and the income loss over all the years between retirement and death. The indirect approach, in turn, measures the total loss based on the value o f the predicted change inearnings. Based on these two approaches, three alternative packages identified with the letters A to C are estimated. Package A i s a lump sum. Package B i s a multiple o f the employee's salary in government. In package C, the amount o f compensation is set inmonths o f salary per year o f service. Data requirement.The 2002 household surveyand the 2005 civil census data. Formulas 1. Predicted losses. Earnings functions for people who do not work in the public sector are estimated. These functions quantify the relationship between individual characteristics and labor earnings. With the annual earnings denoted as Ei, and characteristics such as gender, education, age or region, or residence as "X1", "Xz)', ...,"Xi', and a dummy variable Fi.that indicates whether the person works inthe private formal sector the earnings function can be written as follows: 2. Predictedearnings after separation are derived using the estimated coefficients (indicated by a caret) from (1). F*i is a bivariate dummy variable for the sector in which the separated worker i s expected to end up (F*i=l for the formal sector and F*i=Ofor the informal sector). This variable i s equal to one if the worker gets ajob inthe formal private sector, and equal to zero ifhe or she gets ajob inthe informal sector. ki = +bl Xli +,b2X2i +...+bkXki +bF F;) (2) 3. Estimated losses with the direct approach. The direct approach measures the loss ~4that the separated public I sector worker `Y would experience as: The first sum o f terms inthe right-hand side o f equation (3) is the earnings loss the worker would experience in all the years since displacement until retirement, measured in present value. In this sum, Wi represents salaries and tangible benefits inthe public sector, Ai is the number o f years before the worker reaches the legal retirement age, and `r' i s a subjective discount rate. The second sum o f terms is the income loss over all the years between retirement and death. Inthis sum, Pi i s the old-age pension the worker would be entitled to ifhe or she stayed inthe public sector untilretirement age, and Ni i s the expected number o f years before death. Equation (3) assumes no loss inold-age income ifthe worker gets a formal sector job after separation. 4. Estimatedlosses with the indirect approach. The indirect approach measures the loss LIias follows: Lf = Ai wj(l+vmaX)-kj c (5) t=l (I+.)` Inthis expression, 7- represents the highest "credible" value o f the predictedchange inearnings. 5. Comensation packages. Eachpackage i s denoted S with a superscript for the correspond letter A to C. Package A siA = eA (6) BA is a parameter allowed to vary to make compensation more attractive. Package B sB=eBT (7) SCii s set inmonths o f salary Wi per year o f service, Yi. Variable Yi couldbe one o f the individual characteristics Xi considered when estimating potential earnings out o fthe public sector inequation (1). Source:ChongandRam(2001); see also Chong, Ram,Creppy and Wodon (2006) for a discussion. - 40 - 3.15 The estimated losses under the direct approach tend to be larger than those under the indirect approach with the 2005 census data. First, it is important to clarify what the direct and indirect approaches really mean inthe DOSEmodel. The direct approach measures the total loss that the separated public sector worker would experience based on two elements: first, the present values o f the earnings loss the worker would experience inall the years since displacement until retirement, and finally the income loss over all the years between retirement and death. The indirect approach, in turn, measures the total loss based on the value o f the predicted change in earnings using a parameter (qmax), which is a proportion of the nominal wage. The direct approach i s often used when accurate information on wages and pensions are available. When such information is not reliable or not available, the indirect approach is recommended. On the basis o f the mandatory retirement age o f 65 years, the estimated losses under the direct approach are larger than under the indirect approach41.Also, the overall values for the estimates of the losses are lower in 2005 than 1997, if one takes into account inflation, which is consistent with the finding inthe previous section suggesting that not all workers would necessarily stand to lose injoining the private sector, with the resulting proportion o fworkers loosing being smaller in2005 than in 1997. 3.16 At the highestsatisfactionrate (70 percent),packagesB and C are less expensive than the lump sum package for both the direct and indirect approaches. Before comparing the packages, it is interesting to note that for a low satisfactionrate (30 per~ent)~', the cost o f the package is lower even innominal terms in2005 than in 1997 (see Table 3.4). This is because in 2005, a number o f public servants are estimated not to make a loss in moving to the private sector (assuming that 20 percent would move to the formal sector). It i s also worth noting that the estimates o f the costs increase faster in 2005 with the satisfaction rate o f 70 percent. This i s inpart because there i s more variability inthe wages o f workers in the 0 to Z categories in2005 than in 1997. The most important difference between 1997 and 2005 is that while in 199743,for the highsatisfactionrate o f 70 percent, package C was more expensive than package B, this is not the case in 2005: the cost o f packages B and C are similar, and cheaper than the lump sum package A. This is probably again because o f higher variability in wages among occupational categories 0 to Z in 2005, so that there is also a higher variability in losses, making a lump sum package expensive in order to compensate a larger number o f workers under high satisfaction rates. Table 3.4 reports the cost o f each o f the three packages, for three satisfactionrates, set at 30 percent, 50 percent, and 70 percentsu 3.17 On fairness ground, package C is likely to fare better than the other two regardless of the approach considered. In terms o f fairness, the obvious question is whether packages A to C can be compared? An appropriate concept o f fairness should probably factor inthe contribution that.civil servants have given to their employers, which is often measured by the number o f years o f service among others. Ifone agrees with the fact that workers who have been employed for more years -- and have perhaps given up some 4'When the retirement age is reduced to 55 years, the estimated losses from the two approaches are very close (as in Chong and Rama, 2001) 42The satisfactionrate i s the percentageof civil servants who would be satisfied by the amount of compensationreceived (if the package exceedsthe expected loss, it is assumedthat the employeewouldbe satisfied). 43 44The 0 and 100 satisfactionrates are notreportedbecausethe fist one i s uninteresting, andwouldrequireno compensation,whereas the 100percentsatisfactionratecan only bereachedat unreasonablyhighseparationcosts (i.e. the packagehastobe generousenoughto match the highestloss inearningsandbenefits throughout the sample, andthis is often an outlier). -41 - other opportunities along the way -- should benefit fiom a better treatment when being retrenched, then package C i s better than packages A and B in terms o f fairness. Moreover, Package C is also cheaper than the two other packages for the same satisfaction rate, at least for the higher rates o f satisfaction that can be considered. As to the choice o f the satisfaction rate to be considered, given that many civil servants are poor, and that they don't have much o f a say in the fact that they are being retrenched, one could argue that a high satisfaction rate, perhaps o f the order o f 70 percent, is to be achieved. Precisely because the wages o f civil servants inthe categories 0 to Z has not increased as much as that o f higher ranked civil servants since 1997, achieving a satisfaction rate o f 70 percent might not be too onerous for the state. For example, under package C the average cost per satisfied worker i s roughly CFAF1,640,000 (USD 3,100) and CFA2,370,000 (USD 4,400) for the direct and indirect approaches, respectively. If instead the average o f the two packages B and C i s considered under each approach the cost would be slightly lower than C alone (CFAF1,613,000 and CFAF2,230,000, respectively). The total cost is between CFAF5 to 6 billion, which represent 3 to 4 percent o f GDP (see Table 3.5). Table 3.4: PerformanceofAlternativePackages-Direct andIndirect Approaches, 2005 Satisf. rate Direct approachinthousand CFAF Indirectapproachinthousand CFAF A B C Average BC A B C Average BC 30 254 265 211 620 587 810 50 750 745 766 756 1,500 1,375 1,610 1,493 70 2,750 1,585 1,640 1613 3,350 2,090 2,370 2,230 Source:Chong, Rama, Creppyand Wodon (2006) Note: PackageA is a variablelump sum; package B is a variablemultipleof the salaryingovernment; packageC is avariablemultipleof the salary multipliedbythenumberofyears ofservice Table 3.5: Estimatesof Total SeveranceCost for the Government,2005 Total severancecost of civil workers 0-Z Direct approach Indirectapproach B C Aver.BC B C Aver.BC InbillionCFAF =Packageunitvaluex2,800) 4.6 6.6 4.5 5.9 6.6 6.2 Inpercentof2005 GDP 2.9 4.1 2.8 3.7 4.2 3.9 w:Estimationbased on Chong, Rama, Creppy and Wodon (2006) Short-Term Fiscal impact of Downsizing 3.18 The proposed downsizing of unskilled worker is likely to improve the quality of the corresponding services; however, the reform would have a relatively limited fiscal impact. According to the 2005 census data, in total the salaries o f workers in levels 0 to Z amount to 1.4 billion CFAF (this is the sum total o f wages for the 3,439 workers who are in the 0 to Z categories; this number o f workers in these categories i s similar to the estimate advanced inpolicy discussions in the country). The target level agreed upon under the SMP for the number o f workers to be retrenched in the occupational categories 0 to Z is 2,800. Such a level o f retrenchment would provide savings o f about CFAFl-15 billion. However, without these workers on payroll, the various ministries and other branches o f government would need to rely on services provided by private firms to provide a range o f services. It is indeed hoped that private firms will hire some o f the retrenchedworkers in order to provide the services that the workers were originally providing. It is difficult to estimate what the cost of such services would represent for the budget; however a number o f assumptions can - 42 - be made. It can be assumed that private firms would be more efficient inproviding services. For example, they might be able to do the samejob with 25 percent less workers. It can also be assumed that private firms would pay their staff at lower wage levels than those currently enjoyed in the public sector. Perhaps it could be reasonable to assume that formerly public workers in categories 0 to Z could face a 40 percent wage cut when joining the private sector. On the other hand, there would be a premium paid to the firms providing services in order for them to make a profit, and it could be assumed that this premium would represent 10 percent o f the wages o f the workers. Under these assumptions, the cost for the government o f hiring privately provided services would represent half o f the cost o f maintaining the workers who are providing the services todap. Thus, the retrenchment o f less skilled public workers from occupational categories 0 to Z could reduce the public sector wage bill by about CFAFl.15 billion, which would correspond to about 0.72 percent o f GDP. Yet the net reduction incosts for the government mightbe only halfthat amount, or 0.36 percent o f GDP. 3.19 Additional restrictive fiscal policies would be needed, including a reduction in the size of the armed forces and further downsizing aimed at workers in other categories. While the retrenchment policy would helpto reduce the public sector wage bill, it would by far not be enough to reduce the budgetary deficit andreachthe convergence criteria o f WAEMU. Indeed, it would enable the country to make only marginal progress in that direction. Additional restrictive fiscal policies would be needed, including a reduction inthe size o f the armed forces and a reduction inthe public servant pool among workers belonging to the occupational categories A to N. Long-Term Economic Impact of Downsizing 3.20 This section focuses mainly on the financial return since the operation i s part o f a broader government endeavor to reduce fiscal deficits. The previous computation o f the fiscal gain (0.36% o f GDP) was inherent to the gain in the first year after the reform is implemented. It was simplified with the purpose to show the short term saving in term o f wage bill cut the government will enjoy when the fixed cost o f the operation (the up front cost) is not taken into account. It, infact, largely underestimates the fiscal gain. Indeed, since the retrenchment provides a net gain inthe first year o f implementation and inall subsequent years, the total fiscal gain o f the retrenchment is much higher. More precisely, the net fiscal gain is the present value o f all h t u r e saving interms o f wage bill cut andpensions liabilities. The fiscal impact will therefore be the difference between this fiscal gain and the compensation packages. Since the expenditure and the benefits occur at different points in time (the up front cost occurs immediately while the saving occurs over many years), the retrenchment decision makingby the government requires the careful evaluation o f whether the present cost isjustified bythe benefits it is expected to confer. 45The exact cost i s 49.5 percentofthe original cost since 1*(1-0.25)*(1-0.40)*1.1=0.495 - 43 - Box 3.5: Long-TermFinancialReturnof Downsizing Ifwe ignore any other externalitieswhich may influence the financial retumand suppose that the redundantworker will not be rehired, thenthe presentvalue ofthe fiscal impactofthe downsizingoperationcanbe expressedas follows: Where : c is the upfront cost of the reform, which i s the compensation packages. The mean value of the severance pay per worker is 2 millions. The second terms in the right-handside of the equation is the fiscal gain in terms of budget cut andthe third is the gain interm of pensionliabilities. n denotesthe average lengthof serviceof workers inthe categories0-Z beforeretirement.This could be estimatedby the difference between the average age of workers and the legal retirement age. A recent study on the old pension systeminGuineaBissau46estimatedthe average age ofcivil servantworker to 42. Giventhe legalretirementage of 60, it followed that 18years will be agoodestimatorof n . Si refers to the fiscal gain ffom the reduction in the wage expenditure. In the first year, which i s year 0, So has already been estimated to 575 millions. It will be assumedhere that the saving will be constant over the subsequent years, so If j , s j =so. pj indicatesthe savinginterms ofpensionsliabilities, thatis, theold-agepensionthegovernmentwould havepaidif theworkershadnot left. The retirement benefitan eligible member would be entitledto receive couldvary between60 to 90 percent of the average salary earned during the last two years prior the retirementperiod, but the total pension liability for the government is around 71.3 percent of the wage bill. It can therefore be reasonablybe assumedthat the pension entitlement will be 71.3 percent of the current earning of the unskilled workers and stay at that level until death. a i s the average expectednumber of years workers would have beenreceivingthe pensionbenefit before death. This could normally be approximatedby the differencebetweenthe life expectancy of workers and the legalretirementage. However, life expectancy being very low and less than the legal retirement age, it can be supposed that only 20 percent4' of the workers would reachthe retirement years andreceivethe pensionbenefitover 5 years. r indicatesthe discountrate, which can be fixed at 4.95 percent. Under those assumptions, the above equationyield a fiscal benefit of 7.09 billions (4.47 percent of GDP) and a net financial positive retumof 2.08 billions (1.3 1percent of GDP). The financial break-even period, which Haltiwanger definedas the number of years requiredbeforethe scheme breaks even on financial costs and benefits,is between 11.5 years. This meansthat the cost ofretrenchmentwill be fully recoveredby 2018. This analyze could be extendedto the evaluationof the economic return to the retrenchment reform. In additionto the financial yield, some of the discharged workers will contribute to the economic activity when they are reallocatedto sector where they are moreefficient. A redundantworker, who is ableto find ajob inthe privatesector will always add positive value to the aggregateoutput since its productivity was zero inthe public sector. Hence, the assessmentof the economic return requiresknowing the number of redundantworkers. Inthe present case, this numbercan be estimated to one fourth of the number of workers in categories 0-Z to be retrenched. Indeed, as previously assumed, the new privately owned firm would be able to provide improved service delivery with only 75 percent of the unskilled retrenchedpublic workers, the rest (860 workers) can be consideredas redundant.The study also gives an indication of their earning once they leave the public sector. The present value of those future earnings, added to the financial impact,will give an assessment ofthe economic impactof the downsizing. 3.21 Box 3.5 shows that the present value o f the reduction in government spending (4.47 percent o f GDP) is higher than the up-front cost in terms o f severance pay, and that the downsizing operation will yield a positive financial return o f 1.31 percent GDP. As in making an investment decision, this result indicates that the retrenchment reform is worth 46Barry and Bamba(2006) 47This i s the dependencyratio of the old-age pensioners. - 44 - undertaking, even though the highlevel o f the financial break-evenperiod (11.5 years) may be indicative of a generous compensation packages (risk o f overcompensation). 3.22 This outcome can be challenged when important issues such as adverse selection and rehiring of redundant workers are o f serious concerns. It could be supposed that those effects would be minimized and would not offset the primary positive gains. In particular, the likelihood o f subsequent rehiring o f redundant workers is near zero since the government is contracting out to private firms. The economic impact o f the downswing operation can be very complex to evaluate since it may be affected by how workers are reallocate across sectors, the equilibrium level o f taxes and several other externalities. Overall, since the aim o f the retrenchment is to relocate workers with almost "zero productivity" in the private sector where they mightbe more productive, the reform should normally increase total output andthus improve aggregate welfare. ASSESSING POTENTIALPOVERTYIMPACT OFRETRENCHMENTWITHOUT SEVERANCE THE BENEFITS 3.23 The low skilled workers targeted for retrenchment are on average poorer than any other person. The Guinea-Bissau 2005 poverty assessment estimated the incidence o f poverty and extreme poverty (PO) at 65.7 and 21.6 percent, respectively. On average, households with members who are working inthe public sector as a group are somewhat less likely to be poorer than the population as a whole (headcount index o f poverty o f 56.2 percent, which remains very high, but i s below the estimate o f 65.7 percent at the national level). Yet those households with a member inthe public sector who belong to the least paid groups o f public sector workers are as likely, ifnot more likely to be poorer (headcount index o f 70.6 percent) thanthe national average (see Table 3.6). Povertv(USD 2) in% % PO P, P, Extreme PovertvP,(USD 1 inP, Pi7 Rural areas 70.3 28.5 14.7 25.7 6.8 2.7 Urbanareas 52.6 17.5 7.6 9.8 2.0 .60 Total 65.7 25.7 12.9 21.6 5.5 2.2 Lowestpaid civil servants II 70.6 27.3 13.1 21.8 3.9 1.1 All civil servants 56.2 19.9 8.9 11.8 2.7 0.9 All households 65.7 25.7 12.9 21.6 5.5 2.2 3.24 Retrenching low skilled workers without providing them a severance pay is likely to worsen their condition. From a policy point o f view, given that many workers targeted for retrenchment belong to the least qualified categories o f public servants, and that these workers would probably see their income reduced substantially after retrenchment, one could argue that it will be important to provide good enough separation packages to offset such losses. Ifno severance pay is offered to retrenched workers, incomes losses are likely to lead to a decrease in consumption levels and a worsening o f poverty. Combining the estimates o f the losses in earnings with data on the consumption per equivalent adult o f the households to whom the civil servants belong allows to providing some indication o f the impact on poverty that retrenchment would imply, under the assumption that the workers would not benefit from separation packages. The results inTable 3.6 suggest that the poverty incidence among low skilled workers will increase from 70.6 to 73.2 percent, but the increase - 45 - with the poverty gap and squared poverty gap is much larger, suggesting high negative impacts inthe absence o f severance packages. Retrenched public sector workers should also benefit from training in micro businesses to enhance their productivity and help them find employment inthe private sector. Table 3.7: Estimatesof Poverty among CivilServants Assuming no SeverancePackage Beforeretrenchment After retrenchment Normalized P" in0," Pl in P, 0% Normalized P, inoA P1inoA P,inoL Lowestpaidcivil servants 0.912 70.6 27.3 13. 83.1 73.2 32.9 17.6 All civil servants 1.167 56.2 19.9 8.9 14.5 56.9 21.4 10.2 All households 0.981 65.7 25.7 12. 97.9 65.8 25.9 13.0 -e : Wodon, Creppy and Tsimpo (2006b). CONCLUSIONAND POLICY CONSIDERATIONS 3.25 This chapter attempted to estimate the potential cost o f separation benefits for low skilled workers in Guinea-Bissau's public service. In the short-term, the average compensation benefit per satisfied worker is expected to be between CFAF1.6 and 2.4 million, which translates into a total program cost o f CFAF5 to CFAF6 billion (3 to 4 percent o f GDP). On the other hand the gross salary saving from the downsizing reform is expected to be CFA1.15 billion (0.72 percent), which points to high short-term fix costs. Inthe long- term, however, the present value o f the reduction in government spending (4.47 percent o f GDP) is higher than the up-front cost in terms o f severance pay. This implies that the downsizing operationwill yield a net positive financial return o f 1.31percent GDP. Based on these results, it is safe to conclude that the proposed compensation alternatives seem financially sound, because they trade the equivalent o f a few years in salary for a lifetime o f payments, inthe form o f salaries, benefits, and old-age pensions. However, a comprehensive economic analysis should also take into account the effects o f downsizing on the allocation o f labor across sectors. Moreover, further downsizing o f other categories would need to be implemented inthe future to consolidate the primary gains. 3.26 There are possible risks to downsizing. Many government employees have a very low productivity, but it is not at all obvious that they would be more productive out o f the public sector. Moving from government employment to unemployment, or to the informal sector, could actually reduce total output. There i s an important political caveat. If appropriate compensation i s needed to make the workers accept the prospect o f downsizing, it is very likely that the package will have to be negotiated with public sector trade unions. Inthat case, the final compensation formula may not be the cheapest one, or the fairest one, but rather a compromise shaped by the bargaining power o f all the players involved. From this perspective, the simulation approach proposed in this chapter should not be seen as an example o f "mindless social engineering," but rather as a tool to introduce some economic rationality in delicate political negotiations. Although the approach could certainly be refined, it represents an improvement compared to the ad hoc way in which these negotiations are usually carried out. - 46 - 4. DEBT SUSTAINABILITYANALYSIS4' As highlighted in the previous chapters, Guinea-Bissau's large public sector wage bill has limited the government ability to service its domestic and external debt. In the view of Bank and Fund staffs, Guinea- Bissau is a country in debt distress. However, assuming that HIPC Completion Point would be achieved by end- 2009, debt distress risk is expected to be reduced to moderate levels. Even though current levels of debt and debt service diverge by far from the indicative thresholds for debt sustainability in countries with poor institutions, reaching HIPC completionpoint would allow Guinea-Bissau to reduce external debt to sustainable levels under the baseline scenario. Stress tests point, however, to the vulnerability of this country to a strong exchange rate depreciation, a reduction in thegrowth rate, or a reduction in the grant content of new loans. PUBLICDEBTSITUATIONAS OFEND-2005 4.1 At end-2005, Guinea-Bissau's stock of external public and publicly guaranteed (PPG) externaldebt amountedto USD 1billion, ofwhich USD327 millionare arrears49. Multilateral debt amounts to 49 percent o f total PPG debt (14 percent i s owed to AfDB/AfDF, 29 percent is owed to IDA, 1.2 percent is owed to IMF) and bilateral and commercial amounts to 51 percent. The NPV of debt amounts to USD 745.8 million or 247.3 percent of GDP and Figure4.1: Guinea-Bissau: Stock of debt, 1999-2005 656.9 percent of exports (see Figure 4.1). -Percrnl ofGDP PCTC.nl Ofsxports E=lPcTMnt 0f1symYcs Despite the concessional nature o f most o f the PPG external debt, the debt burden 2wo indicators far exceed the policy dependent debt thresholds for poor I500 performers'O. loo0 4.2 The nominaldebt to GDP ratio 500 has decreased slightlyover the last five years from 353 in 2000 to 332 percent " in 2005. This positive development, however, i s not due to a significant increase in the current account balance or in foreign direct investment (FDI), which would reduce the financing needs of the country. Instead, this debt to GDP reduction was due mainly to a strong appreciation o f the CFAF (pegged to the Euro) with respect to the dollar, which translated into highgrowthrates of nominal GDP inUSdollar terms, despitelowrealGDP growthrates. 4.3 Nominalpublicdomestic debt amountedto 48 percentof GDP in 2005.Nearly90 percent of this debt corresponds to a requiredcapital contributionto join WAEMU for a total o f CFAF 70 billion to be paid over 25 years starting in2005. The remainingdomestic debt i s composed o f obligations to BCEAO linkedto statutory advances (no longer allowed) and to the recapitalization o f two banks after the civil war. Domestic arrears are not included inthis debt estimate as an audit will be necessary for a fairly accurate estimation of these obligations to bepossible. 48This DSA i s base on the low income country (LIC) framework and was prepared by a joint Bank-IMF team. The analysis focuses on external debt. Domestic public debt is mainly composed ofarrears, a settlementof which will be partof foreign financed programs. 49Bilateral debt data areprovidedby the authorities, while multilateral debt data are reconciledwith creditors. 50According to the World Bank Country Policy and Institutions Assessment (CPIA), Guinea Bissau has a poor quality of policies and institutions. The CPIA rating for 2005 is 2.7, in ascale from 1to 6, below the operational cutoff ratingof 3.25 for poor performers. - 47 - 4.4 Since the PRGF went off track, Guinea-Bissau has not benefitedfrom most of debt relief committed at decision point. At decision point in2000, creditors representing 80 percent o f Guinea-Bissau external debt pledged to provide HIPC relief amounting to USD 790 million (USD 417 million in NFV terms). Since 2001, Guinea-Bissau had to service a large share o f external debt infull. The IMF suspended the provision o f interimrelief andthe Paris Club declared nullandvoid any debt rescheduling agreements beyond end-2001. Many agreements signed with other multilateral and bilateral creditors have not been implemented because the country failed to remain current on debt service obligations." The Bank and the AfDB continued providing interim relief', but both multilateral banks will reach the statutory ceiling for the delivery o f interim relief in end-2007 and end-2006 re~pectively.'~ Only China andCuba cancelled all outstanding claims. 4.5 After the decision point, Guinea-Bissaucould not service its external debt and accumulated arrears to most of its external creditors. Since 2001, the country has not repaid any creditor that did not provide interim relief, with the exception o f the IMF. Debt service due has increased from about 10 percent o f export in 2001-02 to 55 percent on average in 2003-05. The stock o f arrears has increased to USD 139 million before decision point to USD 327 million at end-2005. Inthe following analysis, arrears are supposed to be cleared thanks to debt relief and concessional rescheduling before reaching the completion point. However, the HIPC process would be delayed if additional payments arrears arise to multilateral creditors or if Guinea Bissau could not agree on a repayment schedule o f post- cutoff date arrears with the Paris Club. MACROECONOMIC FRAMEWORK OFTHE DSA 4.6 The debt indicators o f the D S A are based on the assumptions that the economy will grow inreal term at the rate o f 5 percent inthe medium-termS4.Projections for the evolution o f external debt and external debt service are carried out in a baseline scenarios and several alternative scenarii: the baseline scenario assumes that the country stays on track with IMF programs and successfully reengages with the donor community; a historic scenario assumes key variables at their 10-year historical average; and differentbound test scenarii, simulating temporary shocks to growth, exports, revenues, the exchange rate, prices and financial flows. The macroeconomic assumptions underlying the baseline projection scenario up to 2009 are described in detail in Box 4.1. The baseline assumes that an international concerted assistance effort i s inplace by end-2006 to help finance urgently needed structural reforms in the public sector. Post-conflict assistance i s assumed to be provided by the Fund by end- 2006, followed, after two years, by a PRGF. Growth will be sustained by reforms in the electricity sector and investment in infrastructure which would loosen the bottleneck for private investment and agricultural activity. Growth averaging 5 percent will be sustained by 5'See Guinea-Bissau, Selected Issues and Statistical Appendix, November 2004, Box 10 for a comprehensive list of debt rescheduling agreementsreachedby Guinea-Bissauwith the creditors as part of the provision of HIPCrelief. 52Interim relief from IDA amounted to 100% of debt service falling due on debt disbursed before end-1999 until October 2003, the expected date of the completion point. Interimrelief was decreased to 90% of debt service thereafter. The AfDB extended 100% debt service reductionto end-2006. 53IDAcould extendinterimreliefto 50 percentof total HIPCreliefifthe countrymaintains satisfactoryprogressin(i)meetingthe floating completion point triggers including progress in the preparation of the PRSP and (ii)maintaining a sound macroeconomic program, as indicatedby performanceunder an IMF-supportedPRGF program. Equally, the AfDB could raise the interimrelief ceiling from 40 to 50 percent. 54This assumption i s consistentwith the authorities' PRSP forecasts. - 48 - exports heavily depending on energy-intensive services. The external current account and the overall budget deficits are projected initially to deteriorate as more resources are assigned to public investment and poverty reduction, but this would be offset by a strong export growth and fiscal consolidation inthe medium and long term. The primary fiscal deficit is initially expected to remain high as more resources are assigned to social sectors. In the medium term, however, stronger revenue performance and Civil and security service reforms would contain the public sector wage bill and improve the fiscal balance. Public external borrowing at concessional terms would be contained over time, while domestic debt and arrears would be progressively paid down. HIPC Completion Point is expected by end-2009. Total HIPC debt relief, as established at Decision Point in 2000, amounts to USD 413 million in net present value terms and would reduce debt service payments by about 7 percent o f GDP per year. DEBT SUSTAINABILITY 4.7 Guinea-Bissau is currently in a situation of debt distress with accumulation of external arrears. The sustainability o f the current external debt isjudged based on various indicative debt and debt service thresholds established for countries with poorly performing economic policies and institutions, and shown in Table 4.1. The simulations (presented in Tables 4.2 to 4.4 and in Figures 4.2 to 4.3) point to a country currently in severe debt distress, with arrears accruing to multilateral creditors, Paris Club creditors and other official and commercial creditors. Such arrears are projected to be settled by 2009 if the HIPC process remains on track. Assuming HIPC relief, indicators based on debt and debt service ratios in simulated scenarii point to mixed signal o f debt distress. The NPV-of-external debt to GDP ratio remains above the 30 percent threshold during most o f the simulation period in the baseline scenario as well as under the stress test scenarios. Inthe case o f total public debt, the debt to GDP ratio remains well above the threshold levels for the whole period and the ratio to government revenues surpasses the threshold for a significant period, reflecting the weight o f capital contribution commitments to WAEMU on top o f the external debt burden. On the other hand, the debt service ratios to exports and government revenues quickly fall below threshold levels in the baseline scenario right after Completion Point i s reached. Alternative scenarios and stress tests indicate that significant exchange rate depreciation, reduction in GDP growth, and a lower grant element o f new loans would substantially increase the risk o f debt distress. The most severe scenario i s posedbythe effects o f large (30 percent) exchange rate devaluation, which would cause the NPV-of- external debt to GDP ratio to significantly exceed the 30 percent threshold during the whole projection period. These simulations do not include the additional debt relief that could be obtained from the Multilateral Debt Relief Initiative (MDRI), which would amount, at Completion Point, to USD 77.3 million innet present value, or 21 percent o f GDP. Inthis case, the external debt service to exports ratio would be reduceby an additional 3 percentage points. - 49 - Table 4.1: SummaryofBaselineDebtSustainability Indicators(*) Indicative 2005 2016 2026 Average threshold (**) 2006-24 NPVo fdebt-to-GDP 30 247.3 36.1 27.5 68.8 NPVofdebt-to-exports 100 656.9 86.8 66.0 168.4 NPVofdebt-to-revenue 200 1482.3 171.8 130.8 456.3 Debtservice-to-exports 15 43.6 4.5 3.9 8.5 Debtservice-to-revenue 25 94 11.9 10.6 19.8 Note: Figuresinbolds are underthe indicative thresholds. (*)All debtindicatorsassumethe full delivery of HIPCreliefandreferto theNPV ofpublic andpublicly guaranteed externaldebt. (**)Thresholdoverwhich countrieswith similar evaluationsofpoliciesandinstitutionswould havean at least 25 percentchanceof having aprolongedincidentof debt distressinthe comingyear. Guinea-Bissaulies within the pottom] quintile of countriesrankedby the World Bank's Country Policy and InstitutionalAssessmentIndex (CPIA). Box4.1:MacroeconomicAssumptions under the BaselineScenario GDP growthrate increases from 3.5 percent in2005 to 5 percent by 2008, and is sustained over the following years. GDP deflator is assumedto grow at a rate slightly below CPI Mation, due to expected worsening inthe terms o ftrade. After 2009, both GDP deflator and CPI are assumedto increase at a rate o f 2 percent. The non-interest current account is assumedto deteriorate over the mediumtermreflecting, both, an increase inforeign-financed investmentprojects and lower official transfers inthe form o f budget support. Over the long run, said balance as a proportiono f GDP converges back to its 2005 level. Exports are assumedto grow at around 8 percent per year until2010, drivenby steady growth incashew at least inthe medium-term; export growth is assumes to decreaseto about 6 percent thereafter. Foreign direct investment to GDP ratio is assumedto remain constant inthe medium and long term at the level observed in2005 (1.3 percent). Newborrowing willbe contracted at high concessional terms, with the grant element innew disbursementsis assumedto remain slightly above the 50percent that was agreed onunderthe 2006 SMP. Government primary deficit i s assumedto decreasefrom 4.5 percent o f GDP in2006 to 1.7 percent by 2010, maintaining this level from then on. Total external loan financing is assumedto be 5 percent o f GDP per year on average over the next 20 years, with 57 percent grant content. CONCLUSION AND STAFFS APPRAISAL 4.8 The updated external debt sustainability analysis points to a country in debt distress, which would be substantially reduced if the HIPC Completion Point is achieved by end- 2009. Continued political stability as well as a concerted international assistance effort are key elements for the country to achieve Completion Point and the associated debt reduction to sustainable levels. Even within this scenario, the country's capacity to sustain this debt level may be compromised by significant shocks to the nominal exchange rate, the growth rate, and the degree o f loan concessionality. Inclosing, given that Guinea-Bissau will hit the one-third limit of IDA HIPC interim relief in 2007, consideration should be given to the possibility o f extending IDA interimassistance as allowed by the Board-approved `Provision ofInterimHIPC DebtReliefon IBRDLoans andIDA Credit'. - 50 - m re ! + c f 2 2 : ji j o o q m S S 4 W d.E I Table 4.3: SensitivityAnalyses for Key Indicators ofPublic and Publicly Guaranteed External Debt, 2006-26 (in percent) ProJeeHons 2006 2007 2008 2009 2010 2011 2016 2026 NPV of debt-to-GDP ratlo Basehe 239 230 224 145 69 40 36 25 A. Alternatlve Scenarios AI. Key variables at their historicalaverages in2007-26 1/ 239 226 217 136 63 35 31 24 A2 Newpublicsectorloans on lessfavorable t m in 2007-26 2/ 239 231 227 150 75 47 47 38 B. BoundTests BI.RealGDP growthat historicalaverageminus onestandarddeviationin 2007-08 239 252 270 175 84 49 43 30 B2. Exportvalue growthat historicalaverageminusone standarddeviationin2007-08 3/ 239 232 231 152 16 41 41 21 B3 US dollar GDP deflatorat historicalaverage minus one standarddeviationin 2007-08 239 236 236 153 73 43 38 26 E4 Netnon-debtcreatingflows at historicalaverageminus onestandarddeviationin2001-08 4/ 239 230 225 146 70 41 37 25 B5. CombinationofBLB4 usingone-halfstandarddeviationshocks 239 242 249 162 78 45 40 28 B6 One-time30 percentnominaldepreciationrelativeto the baselinein 2007 51 239 327 318 206 99 51 51 36 NPV ofdebt-to-exportsratlo Baaelhe 611 574 548 349 164 96 87 62 A. Alternatlve Scenarios Al. Key variables at their historicalaverages in2007-26 I/ 611 565 533 328 148 83 76 61 AZ. New publicsectorloanson less favorable tams in2007-26 2/ 611 578 556 360 178 112 114 94 B. BoundTests B1. RealGDP growthat historicalaverageminus one standarddeviationin2007-08 611 574 548 349 164 96 87 62 B2. Exportvalue growthat historicalaverageminus one standarddeviationin2007-08 3/ 611 677 768 497 245 152 137 92 B3. USdollarGDP deflatorat historkalaverage minus one standarddeviationin 2007-08 611 574 548 349 164 96 87 62 B4. Net non-debt creatingflows at historicalaverageminus onestandarddeviationin2007-08 4/ 611 575 551 351 166 98 89 63 B5. CombinationofBLB4 usingone-halfstandarddeviationshocks 611 606 613 391 184 108 98 70 B6. One-time30 percentnominaldepreciationrelativeto the baselinein 2007 51 611 574 548 349 164 96 87 62 Debt aerviee-to-exportsratlo B u e h e 32 29 25 22 4 4 4 A A. Alternative Scenarios AI. Key variables at their historicalaverages in2007-26 I! 32 28 25 21 4 4 4 3 A2 New publicsectorloanson lessfavorable tennsin2007-26 2/ 32 29 25 22 5 6 5 6 B.Bound Teata BI. RealGDP growthat historicalaverageminus one standarddeviationin2007-08 32 29 25 22 4 4 4 B2. Exportvalue growthat historicalaverageminus onestandarddeviationin 2007-08 3/ 32 33 34 30 6 6 7 B3. USdollarGDP deflatorat historicalaverage minusone standarddeviationin 2007-08 32 29 25 22 4 4 4 B4. Net non-debtcreating flows at historicalaverageminus one standarddeviationin2007-08 41 32 29 25 22 4 4 5 B5. CombinationofBI-B4 usingone-halfstandarddeviationshocks 32 30 28 24 5 5 5 B6. One-time30 percentnominaldepreciationrelativeto the baseline in2007 5/ 32 29 25 22 4 4 4 Memorandum item: Grant element assumedon residualfinancing (i.e., financing requiredabovebaseline) 61 55 55 55 55 55 55 55 -55 Source:Bank and FundStaffprojections and simulations. I/VariablesincluderealGDPgrowth,growthofGDPdeflator(inU.S.dollartm),non-interestcurrentacwuntinpercentofGDP,andnon-debtcreatingflows. 2/ Assumesthat the interestrate onnew borrowingisby 2 percentagepointshigherthan inthe baseline., whilegraceand maturityperiodsare the same as in the baseline. 3/ Exportsvalues are assumedto r m i n permanently at the lowerlevel,but the current account as a shareofGDP is assumedto retumto its baselinelevelafterthe shock (implicitly assumingan offsettingadjustmentin importlevels). 4/ Includes officialandprivatetransfersand FDI. 5/ Depreciationis defmedas percentagedecline in dollar/localcurrency rate,suchthat it never exceeds 100percent. 61Appliesto allstressscenariosexcept for A2 (less favorable financing)in whichthe t m on all new financing are as specifiedin footnote 2. - 52 - YI 1 m v, 1 rn w P m N w o VIN .. .. ..: x ' - N ' S . . . ' 4 O N W N N C N : m % Figure 4.2: Indicators of Public and Publicly Guaranteed ExternalDebt under Alternative Scenarii, 2006-2026 (inpercent) 350 - I NPV of debt-to-GDP ratio 300 Baseline 250 ---Historical scenario 200 Most extreme stress test v - Threshold 150 100 50 0 I 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 900 N P V of debt-to-exports ratio 800 - 700 ---Baseline Historical scenario 600 - Most extreme stress test -Threshold 500 400 300 200 100 O L ' ' 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 - 40 Debt service-to-exports ratio 35 - 30 - ---Baseline Historical scenario Most extreme stress test 25 - _I__ - Threshold 20 - 1 5 ---- ~ ~ ~ m . m . m 10 - 5 - 0 - 54 - Figure 4.3: Guinea-Bissau: Indicators of Public Debt Under Alternative Scenarii, 2006-2026 1/ 400 I NPV of Debt-to-GDPratio -Baseline 300 - / I_ - Most extreme stress test Threshold 250 200 150 100 50 0 1 """"" 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 - 1400 NPV of Debt-to-RevenueRatio2/ 1200 Baseline Most extremestress test 1000 -Threshold 800 600 400 200 0 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 so 45 Debt Service-to-RevenueRatio 2/ -Baseline 40 - Mostextremestress test Tbreshold 35 - -- -- 30 25 -- -- I_-- _I ___l I_ _I UI __I 20 15 i o 5 a 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 w: Bank-FundStaff projectionsandsimulations. I/Mostextremestresstestistestthatyieldshighestratioin2016. 2/ Revenueincludinggrants. - 55 - 5. CONCLUSIONTO THE PER 5.1 The purpose o f this PER update was to review progress in macroeconomic and fiscal management since the previous PER, and analyze public sector wage and debt sustainability issues, which are the main fiscal challenges currently facing Guinea-Bissau. The main messages o f the PER and key policy issues to be addressed in the near future are summarized below. MAINMESSAGES PER OFTHE 5.2 Public finance management has gradually improved since 2004, but the overall fiscal situation has remainedvery difficult. As analyzed inthe second chapter, public finance management has gradually improved since 2004 whit the implementation o f the cash flow management system. However, the overall fiscal situation has remained very difficult. Although the economy started recovering since 2004, the current wage burdenand difficulties to pay salaries are a serious threat to social and economic stability. 5.3 The proposedretrenchmentreform is a positive step towards improvingthe fiscal stance; however, it will take time before the fiscal situation is fixed. Compensation benefits would help mitigate losses from separation. However, the fiscal effect o f the reform is limited. Two types o f actions will be needed to improve the primaryfiscal surplus inthe medium-term: (i) primary spending by extending the reduce retrenchment reform to other segments o f the public administration and the military; and (ii) accelerate macroeconomic and structural reforms, which would allow tapping into HIPC debt relief andtherefore reducing the interest bill after completion point in2009. 5.4 In the meantime, it is vital that donors continue to support the government's reform agenda, and the authorities will need sizable external support in order to implement both the civil service reform and the SSR. The impact o f the public sector downsizing will probably take time to materialize. In addition, given the current macroeconomic difficulties, Guinea-Bissau is not expected to reach the HIPC completion point and therefore benefit from unconditional debt relief before 2009. Inthe meantime, it is important that the authorities ensure the normal functioning o f the public administration and the delivery o f basics social services while implementing the retrenchment. Any significant disruption such as the non payment o f salaries will contribute to stall the reform process. It i s therefore important that the donors continue to support the government's reform agenda. THEWAY FORWARD 5.5 In order to provide more leverage to ongoing fiscal reforms, the next generationof reforms would needto address issues relatedto public sector pensions and the transition from the general sales tax to the value added tax. As stressed throughout this report, Guinea-Bissau's main fiscal problem stems from the public sector wage bill, which diverts financial resources from social andproductive sectors. While the reforms proposed by the government represent a step forward, they will need to be supplemented by hrther expenditure rationing and revenue enhancing policies inthe near - 56 - future. Two key elements o f interest in that regard are the reform o f the public sector pension system andthe transition from the current IGV to the TVA. 5.6 Public sector pension issues. Pensions represent about 16 percent o f the total wage bill, and have grown in nominal terms by more than 8 percent between 1997 and 2005 (see Annex 4, Table A4.3). As with other low income countries, Guinea-Bissau's current pension system is based on the `pay-as-you-go' principle. It has the advantage o f being simple because no specific pension fund is required. Retirement benefits represent, therefore, a debt on the Government. The current system provides a safeguard mechanism against the misuse o f pre-established retirement funds as it has been the case in other countries. However, the government does not have a tool for analyzing the fiscal cost o f the benefit system. Since 2000, the Government had been planning to reform the State pension with the support o f donor-financed technical assistance. However, this objective was not achieved due to the political instability that the country has gone through in the past five years. Ina country with such a scarcity o f resources focusing on the efficiency and impact o f what is spent on pensions is as important as focusing on how to raise additional resources without generating additional poverty. For this reason it is important duringthe next PERexercise to work together with the government PERteam to develop the Pension Reform Option Simulation Toolkit (PROST)55 model for Guinea-Bissau, which canbe then transferred to the authorities. 5.7 The transition from the IGV to the TVA. As a member o f WAEMU since 1997, Guinea-Bissau is required to apply the sub-regional TEC, and conform to other regional fiscal and financial regulations and norms. The country is the only member o f WAEMU who has continued to apply the IGV with an average rate o f 15 percent. Adoption o f the VAT system, with a rate o f 17 percent, therefore, will be a major reform for the government in 2007. The main advantage o f the VAT is that it is levied on all imported goods and services and thereby contributes to broaden the tax base. For a country with a large retail segment as it is the case in Guinea-Bissau, the VAT would offer all the economic advantages o f a tax that includes the entire retail price within its scope. At the same time, the payment o f the tax is spread out and over a large number o f businesses instead o f being concentrated on particular groups such as wholesalers or retailers. Another advantage o f the VAT i s that it limits tax evasion and ensures revenue security for the government: if tax evasion takes place in the VAT, the government is expected to lose only the tax on the value added at that stage, whereby inthe IGV system all the taxes due on the product are lost to the government. On the other hand, the main disadvantages o f the VAT are that it is often regressive and inflationary. It i s claimed that the burden o f the VAT falls disproportionately on the poor since the latter is most likely to spend more o f their income than the rich. Duringthe next PER exercise, it would be useful to work together with the government PER team to: (i) analyze the potential fiscal impact and the poverty or distributive effect o f the TVA system; and (ii)explore mitigating factors for the poor after the TVA adoption. " ThePROSTisapensionmanagementtooldevelopedbythesocialprotectionteamoftheBank.Ithasbeenalreadyappliedinmany countries (e.g. China, Eritrea, Cape Verde etc.) - 5 7 - REFERENCES Alderman, H., S. Canagarajah, and S. Younger (1996) "A Comparison of Ghanaian Civil Servants' Earnings before and after Retrenchment." Journal of African Economies 4(2) :259-8 8. Assaad, R. (1999) "Matching Severance Payments with Worker Losses in the Egyptian Public Sector." WorldBank Economic Review 13(1):117-54. Haltiwanger, J. and M. Singh (1999) Cross-country Evidence on Public Sector Retrenchment" WorldBank Economic Review 1999 13: 23-66; doi:10.1093/wber/l3.1.23. http://wber.oxfordjournals.ordccri/reprint/l3/1/23 Barry,B-S.,andQ. Wodon (2006) "Public Sector Wage BillandEconomic Development inGuinea-Bissau". ForthcominginWorldBankWorker PaperSeries. Barth, R. and W. Hemphill (2000) "Financial Programming and Policy: The Case of Turkey", IMFInstitute Chong, Alberto, and IsmailRadwan (1999) "AnApplication o f Alternative Retrenchment Schemes to State-Owned Enterprises in Tanzania." World Bank, Development Research Group, Washington, D.C. Chong, Alberto, and Martin Rama. 2001. "Do Separation Packages Need to Be That Generous? Simulations for Government Employees in Guinea-Bissau," in Shantayana Devarajan, F. Halsey Rogers and Lyn Squire, editors, World Bank Economists 'Forum, World Bank, Washington, D.C. Chong, Alberto; MartinRama; EdwardCreppy; and Quentin Wodon (2006) " Simulating Separation Packages for Government Employees: and Update for Guinea-Bissau". Forthcoming inWorld BankWorking Paper Series. Fiszbein, Ariel. (1994) "An Opportunity Cost Approach to Redundancy Compensation: An Application to Sri Lanka." Estudios de Economia, Special Issue on Labor Economics inLessDevelopedCountries, pp. 115-26. Haltiwanger, John, and Manisha Singh. 1999. "Cross-Country Evidence on Public Sector Retrenchment." WorldBank Economic Review 13(1):23-66. IMF (2001) "Guinea-Bissau Statistical AppendixTountry Report No. 01/23, January 2001. Kikeri, Sunita. 1997. Privatization and Labor: What Happens to Workers When GovernmentsDivest. World Bank Technical Paper 396. World Bank, Washington, D.C. Manasan R.G. and R.G. Querubin (1986) "Revenue Performance o f National Government Taxes: 1975-1985" Philippine Institute for Development Studies, Staff PaperNo. 8701. http://dirp4.pids.gov.ph/ris/sp/pidssp8701.pdf - 58 - Rama, Martin. 1999. "Public Sector Downsizing: An Introduction." World Bank Economic Review 13(1):1-22. Rama, Martin, and Donna MacIsaac. 1999. "Earnings and Welfare after Separation: Central BankEmployees inEcuador." World Bank Economic Review 13(1):89-116. Wodon, Quentin (2006) "Public Sector Wage and Retrenchment in Guinea-Bissau". Forthcoming inWorld BankWorking Paper Series. Wodon, Q., E. Creppy, and C. Tsimpo (2006a) "Assessing the Potential Poverty Impact of Retrenchment Without Compensation Packages for Public Servants". Forthcoming in World BankWorking Paper Series. Wodon, Q., E. Creppy, and C. Tsimpo (2006b) "Trends in Public Employment and Wages inGuinea-Bissau". Forthcoming inWorld Bank Working Paper Series. World Bank (2004) "Enhanced Heavily Indebted Poor Countries (HIPC) Initiative: Provision o f Interim HIPC Debt Relief on IBRD Loans and IDA Credits. No. 29874. August 20,2004. - 5 9 - ANNEXES Annex 1: Fiscaland Economic Data Table Al. 1:CentralGovernmentOperations, 2001-06 (Inpercentof GDP) 2001 2002 2003 2004 2005 2006 Prel. Budget Est. Budget S M P Revenueandgrants 31.3 21.4 22.9 34.4 31.6 26.0 36.0 31.8 Revenue 16.8 15.3 15.2 17.2 18.4 17.6 19.5 20.7 Tax revenue 10.1 8.4 8.7 8.3 9.1 11.5 11.5 11.8 Nontaxrevenue 6.7 6.9 6.5 8.9 9.4 6.1 8.0 8.9 Grants 14.5 6.1 7.7 17.2 13.2 8.4 16.5 11.2 Budgetsupport 5.7 1.3 0.8 8.4 1.3 3.2 3.1 3.1 Projects 8.8 4.8 6.9 8.8 11.8 5.2 13.4 8.0 Expenditure 43.O 32.7 38.6 49.4 44.1 38.0 48.3 39.9 Current expenditure 28.2 23.8 25.5 29.6 26.8 27.6 25.4 25.3 Of which: wages andsalaries 7.5 7.4 10.0 11.3 12.3 13.4 11.3 11.1 Interest 8.4 4.8 5.7 6.2 3.6 4.1 3.1 3.1 Domestic interest 0.4 0.5 0.4 1.9 4.0 0.3 0.2 0.2 Externalinterest (scheduled) 8.0 4.3 5.3 4.2 2.9 3.8 2.9 2.9 Capitalexpenditureandnet lending 14.8 8.9 13.1 19.9 17.4 10.4 22.8 14.7 PublicInvestmentProgram 13.7 8.9 11.1 11.1 17.0 10.0 19.8 11.6 Domesticallyfinanced 1.7 0.5 0.0 0.3 1.3 0.7 2.2 1.3 Foreignfinanced 12.0 8.5 11.1 10.8 15.7 9.3 17.5 10.3 Other capitalexpenditure 1.2 0.0 1.9 8.8 0.4 0.4 3.1 3.1 Domesticallyfinanced (including WAEMU) 0.0 0.0 0.3 1.1 0.4 0.4 2.7 2.7 Foreignfinanced (includingPDRIU) 1.2 0.0 1.7 7.7 0.0 0.0 0.4 0.4 Domestically-financedprojects 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Other domestically-financedcapitalexpenditure 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Foreign-financedprojects 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Overallbalance, includinggrants(commitmentbasis) -11.7 -11.3 -15.7 -15.0 -12.5 -12.0 -12.3 -8.1 Overallbalance, excludinggrants (commitmentbasis) -26.2 -17.4 -23.4 -32.2 -25.7 -20.4 -28.8 -19.2 Overallbalance, includinggrants(cashbasis) -10.5 -7.3 -0.1 -8.9 -12.5 -10.2 -10.7 -6.5 Financing 10.5 7.3 0.1 8.9 12.5 10.2 10.7 6.5 Domestic financing -0.5 2.5 -4.0 -1.1 -1.6 -0.8 -1.1 -1.1 Foreignfinancing(net) 11.0 4.7 4.0 10.1 -9.8 11.0 1.o -0.9 of which: External arrears 1.9 10.2 8.6 10.5 0.0 7.0 5.5 5.5 of which: Debt relief 17.8 5.0 3.4 3.2 4.2 2.8 3.3 3.3 Financinggap (+= financingneeds) 0.0 0.0 0.0 0.0 23.9 0.0 10.8 8.5 Memorandumitem: Domesticprimary balance -4.7 -4.2 -4.1 -6.5 -6.0 -6.5 -5.1 -2.8 Source: MOF andIMFstaff estimatesandprojections. - 60 - Table Al. 2: CentralGovernmentOperations,2001-06 (InbillionsofCFAfrancs) 2001 2002 2003 2004 2005 2006 Prel. Budget Est. Budget SMP Revenueandgrants 45.6 30.4 31.4 49.1 50.2 41.3 62.1 54.9 Revenue 24.5 21.7 20.8 24.5 29.3 28.0 33.6 35.7 Tax revenue 14.8 11.9 11.9 11.8 14.4 18.3 19.8 20.3 Nontaxrevenue 9.7 9.8 8.9 12.7 14.9 9.6 13.7 15.3 Grants 21.1 8.7 10.6 24.5 20.9 13.4 28.5 19.2 Budgetsupport 8.3 1.9 1.1 12.0 2.1 5.1 5.4 5.4 Projectgrants 12.9 6.8 9.5 12.5 18.8 8.3 23.1 13.8 Total expenditure 62.8 46.4 52.9 70.5 70.1 60.3 83.2 68.8 Current expenditure 41.1 33.7 35.0 42.1 42.5 43.9 43.8 43.5 Wages andsalaries 11.0 10.5 13.6 16.2 19.5 21.3 19.5 19.2 Goods andservices 7.4 7.7 3.9 4.5 5.7 7.6 5.6 5.6 Transfers 7.2 7.4 5.1 5.0 6.4 6.1 10.0 10.0 Other currentexpenditures 3.3 1.2 4.5 7.7 4.6 2.4 3.5 3.5 Scheduledinterest 12.2 6.9 7.8 8.8 6.3 6.5 5.3 5.3 Domesticinterest 0.6 0.8 0.6 2.8 0.4 0.5 0.4 0.4 Externalinterest 11.6 6.1 7.3 6.0 5.9 6.1 5.0 5.0 Capitalexpenditure andnet lending 21.6 12.7 17.9 28.3 27.6 16.5 39.4 25.3 Public InvestmentProgram 20.0 12.7 15.3 15.8 27.0 15.9 34.1 20.0 Domesticallyfinanced 2.5 0.7 0.1 0.5 2.1 1.1 3.8 2.3 Foreignfinanced 17.5 12.0 15.2 15.3 24.9 14.8 30.2 17.7 Other capitalexpenditure 1.7 0.0 2.7 12.5 0.6 0.6 5.3 5.3 Domesticallyfinanced (incl.WAEMU) 0.0 0.0 0.4 1.5 0.6 0.6 4.6 4.6 Foreignfinanced (incl. PDRRI) 1.7 0.0 2.3 11.0 0.0 0.0 0.7 0.7 Overallbalance, incl. grants (commit. basis) -17.1 -16.1 -21.5 -21.4 -19.9 -19.0 -21.1 -13.9 Overallbalance, excl. grants (commit. basis) -38.3 -24.7 -32.1 -46.0 -40.8 -32.4 -49.7 -33.2 Net domestic arrears 1.9 1.4 14.8 1.5 0.0 -0.1 -0.5 -0.5 Accumulationcurrent year 1.9 5.3 17.7 3.9 0.0 4.2 0.0 0.0 Paymentspreviousyears 0.0 -3.9 -2.9 -2.4 0.0 -4.3 -0.5 -0.5 Externalinterestarrears currentyear 2.8 2.8 5.8 4.5 0.0 3.8 3.2 3.2 Float and statisticaldiscrepancies -2.8 1.5 0.8 2.7 0.0 -1.0 0.0 0.0 Overallbalance, including grants (cash basis) -15.3 -10.3 -0.1 -12.7 -19.9 -16.2 -18.5 -11.2 Financing 15.3 10.3 0.1 12.7 19.9 16.2 18.5 11.2 Domestic financing -0.7 3.6 -5.4 -1.6 -2.5 -1.3 -1.8 -1.9 Bankfinancing -3.9 3.2 -4.5 -1.2 -2.6 -1.5 -2.5 -2.5 Of which: IMFrepaymentshepurchases -0.5 -1.1 -2.8 -2.8 -1.7 -1.7 -2.2 -2.2 Nonbankfinancing 3.1 0.4 -1.0 -0.4 0.1 0.2 0.7 0.6 Foreignfinancing (net) 16.0 6.7 5.5 14.3 -15.6 17.5 1.7 -1.5 Disbursements 4.6 4.9 8.0 13.8 7.7 6.5 7.8 4.6 Projects 2.9 4.9 5.7 6.2 7.7 6.5 7.8 4.6 Programs 1.7 0.0 2.3 7.6 0.0 0.0 0.0 0.0 Amortization (scheduled) -17.4 -19.7 -19.0 -18.9 -29.9 -14.6 -11.4 -11.4 Treasurybills 0.0 0.0 0.0 0.0 0.0 10.0 -9.9 -9.9 Of which: regionally placed 9.0 -8.9 Externalarrears 2.8 14.4 11.8 14.9 0.0 11.1 9.5 9.5 Debtrelief 26.0 7.1 4.7 4.5 6.6 4.5 5.7 5.7 Gross financinggap (f = financingneeds) 0.0 0.0 0.0 0.0 38.0 0.0 18.5 14.7 Source: IMF - 6 1 - Table Al. 3: Selected Economic and FinancialIndicators, 2001-06 2001 2002 2003 2004 2005 2006 Prel. Est. Proj. (Annualpercentagechange, unlessotherwiseindicated) Nationalaccountsandprices RealGDP at marketprices 0.2 -7.1 -0.6 2.2 3.5 4.2 RealGDP per capita -2.7 -9.9 -3.6 -0.8 0.2 1.6 GDP deflator -5.4 4.6 -2.8 2.7 7.7 4.2 Consumerprice index (annual average) 3.3 3.3 1.6 0.8 3.4 1.o Consumerprice index (end-of-period) -1.9 2.5 0.7 2.2 1.0 2.5 Externalsector Exports, f.0.b. (based onU.S.dollar values) -19.4 7.1 16.1 21.8 32.9 7.8 Imports,f.0.b. (based onU.S.dollar values) 10.3 -9.8 0.2 17.2 43.4 10.0 Exportvolume 3.8 -0.5 7.3 15.8 9.5 7.7 Import volume -8.8 -9.4 -12.9 40.9 35.5 9.9 Terms oftrade (deterioration-) -22.6 2.4 -6.0 -3.9 14.4 0.1 Real effectiveexchangerate(depreciation-) 1.7 2.5 -2.3 0.6 0.4 ... Nominal exchange rate(CFAF' per U.S. dollar; average) 732.4 694.6 580.1 527.6 526.6 ... Governmentfinances Domesticrevenue(excludinggrants) -16.9 -1 1.6 -3.8 17.7 14.1 27.4 Total expenditure -7.9 -26.1 14.1 33.1 -14.4 14.0 Current expenditure -20.7 -18.0 3.8 20.4 4.1 -0.7 Capitalexpenditure 26.3 -36.5 20.2 3.6 0.4 26.0 Money and credit Creditto government(net) -0.2 0.3 -2.4 -7.4 -6.2 -5.7 Creditto the rest ofthe economy -0.1 -0.3 -1.8 -1.3 2.5 4.4 Broadmoney 9.5 24.2 -65.3 44.0 19.9 8.5 Velocity (GDPhroad money) 2.1 1.6 4.5 3.3 3.1 3.1 (InpercentofGDP, unless otherwiseindicated) Investmentsand savings Gross investment 14.4 9.6 12.6 13.2 14.6 17.2 Ofwhich: governmentinvestment 13.7 8.9 11.1 11.1 10.0 11.6 Gross domestic savings -19.9 -11.8 -1.6 -3.0 -2.9 -2.2 Ofwhich: governmentsavings -18.3 -12.1 -16.8 -30.1 -16.5 -13.2 Gross nationalsavings -7.7 -1.0 9.8 15.5 7.6 10.2 Governmentfinances Budgetaryrevenue 16.8 15.3 15.2 17.2 17.6 20.7 Total domesticprimary expenditure 21.5 19.4 19.9 23.7 24.2 23.5 Domestic primary balance -4.7 -4.2 -4.7 -6.5 -6.5 -2.8 Overallbalance (commitment basis) Includinggrants -11.7 -11.3 -15.7 -15.0 -12.0 -8.1 Overallbalance (commitment basis) Excludinggrants -26.2 -17.4 -23.4 -32.2 -20.4 -19.2 Externalcurrent account (includingofficial currenttransfers) -22.1 -10.7 -2.8 2.4 -7.0 -6.9 Excludingofficial currenttransfers -36.2 -18.7 -12.0 -7.4 -11.1 -13.0 Excludingofficial transfers other than fishing licenses -3 1.O -12.7 -6.2 -5.4 -5.4 -10.1 NPV of externaldebvexports of GNFS (inpercent) 824.0 815.9 635.7 477.1 359.8 320.1 Memorandumitems(inmillions of USdollarsunless otherwise indicated) Nominal GDP at marketprices (inbillions of CFA francs) 145.9 141.9 137.1 142.6 158.8 172.3 Nominal stock of debt (e.0.p; excl. arrears) 802.6 747.2 792.1 827.1 710.8 715.5 Nominal stock of arrears, e.0.p 137.5 169.0 206.3 244.2 272.7 290.4 Source: Guinea-BissauauthoritiesandIMFstaffestimatesandprojections. - 62 - TableAl. 4: BalanceofPayments,2001-06 (inbillionof CFA francs) 2001 2002 2003 2004 2005 2006 Prel. Est. Proj. Goods and services -50.0 -30.4 -19.4 -23.1 -27.7 -33.4 Goods -20.7 -11.8 -5.0 -3.8 -9.6 -12.4 Exports, f.0.b 36.6 37.2 36.1 40.0 53.1 60.2 Of which: cashew nuts 34.5 32.9 32.3 38.4 44.3 50.3 Imports, f.0.b -57.4 -49.1 -41.1 -43.8 -62.7 -72.5 Services (net) -29.3 -18.5 -14.4 -19.3 -18.1 -21.1 Credit 5.4 5.1 5.0 6.6 6.2 7.0 Debit -34.6 -23.6 -19.4 -25.9 -24.3 -28.1 Income (interest scheduled) -11.6 -6.1 -7.3 -6.0 -6.1 -5.0 Current transfers (net) 29.4 21.3 22.8 32.5 22.7 26.5 Official 2131 20.6 11.4 12.6 20.0 12.6 15.4 Ofwhich: balance ofpayments support grants 8.3 1.9 1.1 12.0 5.1 5.4 Ofwhich: fishinglicense fees 7.6 8.5 8.0 9.0 7.5 10.0 Private 8.8 9.9 10.2 12.5 10.2 11.0 Current account Includingofficial transfers -32.2 -15.2 -3.8 3.4 -11.0 -11.9 Current account Excludingofficial transfers 4/ -45.3 -18.1 -8.5 -7.6 -16.1 -17.3 Current account Excl. official transfers and interestpayments -33.6 -12.0 -1.2 -1.6 -10.1 -12.4 Capital and financial balance 4.4 4.3 -62.4 -8.4 2.0 -13.4 Capital account 9.1 27.2 9.4 15.4 10.3 16.0 Financial account -4.6 -22.9 -71.8 -23.9 -8.3 -29.4 Official medium- and long-term disbursements 4.6 4.9 8.0 13.8 6.5 4.6 Balance ofpayments support 1.7 0.0 2.3 7.6 0.0 0.0 Projects 2.9 4.9 5.7 6.2 6.5 4.6 Scheduled amortization -17.4 -19.7 -19.0 -18.9 -14.6 -11.4 Treasury Bills 0.0 0.0 0.0 0.0 10.0 -9.9 Commercial banks' net foreign assets -0.6 2.9 9.8 -0.1 -5.4 1.8 Private net foreign assetsand errors and omissions 8.8 -11.0 -70.6 -18.6 -4.9 -14.5 Overallbalance -27.8 -10.8 -66.2 -5.1 -9.1 -25.3 Financing 27.8 10.8 66.2 5.1 9.1 25.3 Net foreign assets (increase-) -3.7 -14.0 43.5 -18.9 -10.3 -7.8 Ofwhich; net IMF credits -0.5 -1.1 -2.8 -2.8 -1.7 -2.2 Purchasesand loans 0.0 0.0 0.0 0.0 0.0 0.0 Repurchasesand repayments -0.5 -1.1 -2.8 -2.8 -1.7 -2.2 Debt relief 25.9 7.7 5.1 4.5 4.5 5.7 Change indebt-service arrears (decrease -) 5.6 17.2 17.6 19.4 15.0 12.7 Financing gap 0.0 0.0 0.0 0.0 0.0 14.7 Memorandum items: Export volume growth (inpercent) 3.8 -0.5 7.3 15.8 9.5 7.7 Import volume growth (inpercent) -8.8 -9.4 -12.9 40.9 35.5 9.9 Scheduled debt service Inpercent ofexports andservice credits 69.1 60.9 64.0 53.6 34.8 24.3 Inpercent oftotal governmentrevenue 118.6 118.9 126.1 101.8 73.7 45.8 Current account balance (in percent o f GDP) Including official transfers -22.1 -10.7 -2.8 2.4 -7.0 -6.9 Excluding official transfers -36.2 -18.7 -12.0 -7.4 -11.1 -13.0 Stock o f external arrears, e.0.p 80.4 97.6 115.2 134.7 149.6 162.3 Source: IMFstaff - 63 - Table Al. 5: Monetary Survey, 2001-06 2001 2002 200311 2004 2005 2006 Proj. 21 (Inbillions ofCFA francs) Total assets 78.1 95.0 30.2 43.4 52.0 56.5 Net foreign assets 31.5 48.3 13.7 32.5 37.5 47.1 Centralbank 37.2 51.2 7.7 26.6 36.9 44.7 Depositmoneybanks -5.7 -3.0 6.0 5.9 0.5 2.4 Assets 2.8 5.5 6.0 6.5 3.8 2.6 Liabilities -8.5 -8.4 0.0 -0.5 -3.2 -0.2 Net domesticassets 46.6 46.8 16.4 10.9 14.6 9.4 Net domesticcredit 19.8 23.O 17.8 12.0 12.8 12.4 Net claims on governement 15.4 18.8 15.2 9.7 9.4 6.8 Net claims on centralgovernment 15.4 18.8 15.2 9.7 9.4 6.8 Centralbank 16.5 19.7 16.0 10.5 12.5 10.1 Claims 20.3 20.2 17.9 15.1 13.3 10.4 Advancesto the treasury 14.2 14.3 12.0 9.5 7.7 5.0 Statutorylimit 2.5 2.8 2.6 2.1 1.9 1.6 Use ofFundcredit 11.7 11.4 9.5 7.4 5.6 3.4 Other advances 0.0 0.3 0.0 Consolidatedloans 6.1 5.9 5.8 5.6 5.5 5.4 Deposits -3.7 -0.5 -1.9 -4.6 -0.7 -0.4 Depositmoneybanks -1.1 -0.9 -0.8 -0.7 -3.1 -3.3 Claims 0.0 0.3 0.5 1.o 0.1 0.0 Deposits -1.1 -1.2 -1.3 -1.7 -3.2 -3.3 Credit to the economy 4.4 4.2 2.7 2.3 3.3 5.6 Other items (net) 26.8 23.8 -1.4 -1.1 1.8 -3.0 Liabilities 70.1 87.0 30.2 43.4 52.0 56.5 Broadmoney 70.1 87.0 30.2 43.4 52.0 56.5 Local currency 70.1 87.0 30.2 43.4 52.0 56.5 Basemoney 53.1 71.1 21.3 32.6 39.9 43.3 Demanddeposits andquasi money 17.1 15.9 8.9 10.9 12.1 13.2 Demanddeposits 16.3 14.6 8.3 10.3 11.0 12.5 Quasimoney 0.8 1.3 0.6 0.6 1.1 0.7 (Annual change as percent ofbeginning-of-periodmoney stock, unlessotherwiseindicated) Net foreign assets 4.7 23.9 -39.7 62.4 11.4 18.4 Net domestic assets 6.7 0.3 -34.9 -18.4 8.5 -9.9 Domestic credit -0.8 4.5 -6.0 -19.3 1.8 -4.5 Creditto the governement -0.2 0.3 -2.4 -7.4 -6.2 -5.7 Creditto the economy(inpercent) -0.1 -0.3 -1.8 -1.3 2.5 4.4 Other items (net) 7.5 -4.3 -28.9 0.9 6.7 -9.3 Broadmoney 9.5 24.2 -65.3 44.0 19.9 8.5 Velocity (GDPlM2) 2.1 1.6 4.5 3.3 3.1 3.1 Source: IMF 1/ In2003, the BCEAO revisedits estimateofcurrencyincirculation, broadmoney, andnet foreign assets.The mission will check these figures. 2/ Basedongovernmentdraft budget for 2006 - 64 - Table Al. 6: Breakdownof GovernmentRevenue,2003-05 (inbillionof CFA) 2003 2004 2005 Budget Actual Exec. Budget Actual Exec. Budget Actual Exec. TOTAL REVENUE 68.9 31.4 46% 45.3 49.1 108% 33.2 41.3 125% Total budgetaryrevenue 32.0 20.8 65% 24.0 24.5 102% 27.6 28.0 101% 1. Fiscalrevenue 18.5 11.9 64% 10.4 11.7 113% 18.3 14.3 78% Direct Taxes 3.7 2.9 79% 2.6 2.8 109% 4.1 3.4 82% Incometax 3.6 2.5 69% 2.6 2.8 110% 4.1 3.3 82% Industrialcontribution 1.9 1.7 89% 1.3 1.6 119% 2.0 1.7 89% Predialcontribution 1.o 0.4 40% 0.8 0.6 82% 0.8 0.8 105% Urban 0.1 0.1 98% 0.2 0.1 23% 0.2 0.2 125% Rural 0.9 0.3 31% 0.6 0.6 105% 0.6 0.6 100% Professionaltax 0.7 0.4 58% 0.5 0.6 126% 1.3 0.8 58% Public administration 0.4 0.4 100% 0.2 0.2 150% 0.8 0.4 44% Other 0.3 0.4 108% 0.3 0.4 115% 0.5 0.4 78% Complementarytax 0.0 0.0 41% 0.0 0.0 983% 0.0 0.0 161% Other direct taxes 0.1 0.0 23% 0.0 0.0 73% 0.0 0.0 81% Capitatax 0.0 0.0 21% 0.0 0.0 70% 0.0 0.0 81% Proprietytax 0.0 0.0 24% 0.0 0.0 0.0 0.0 0% IndirectTaxes 14.8 9.0 61% 7.8 8.9 114% 14.2 10.9 77% Tax on internationaltransactions 8.6 3.8 44% 4.2 4.0 97% 6.4 4.9 76% Import taxes 3.2 2.3 72% 1.5 2.0 133% 3.8 2.8 74% Gasoline 0.4 0.5 21% 0.3 0.5 166% 0.9 0.9 110% Rice 0.5 0.5 07% 0.3 0.2 66% 0.7 0.4 62% Other imports 2.3 1.2 55% 1.o 1.4 140% 2.2 1.4 64% Exporttaxes 0.0 0.0 0.0 0.0 0.0 0.0 Extraordinarytax on cashew nuts 4.4 1.2 26% 1.7 1.7 100% 2.0 1.8 90% Customsservices 0.0 0.0 0% 0.0 0.0 0.0 0.0 Customsexemptions 0.9 0.0 0% 0.6 0.0 2% 0.0 0.0 Solidaritytax 0.1 0.0 0% 0.3 0.3 103% 0.7 0.3 41% Consumptiontax (IEC) 1.4 1.6 115% 1.o 1.4 137% 1.6 1.4 91% Importsgoods 1.2 1.2 98% 1.o 1.4 139% 1.6 1.4 91% Gasoline 0.5 0.6 111% 0.3 0.4 127% 0.5 0.5 84% Other goods 0.7 0.6 89% 0.7 0.9 144% 1.o 1.o 94% Localgoods/DGCI 0.2 0.4 250% 0.1 0.1 100% 0.0 0.0 Generalsalestax (IGV) 4.2 3.2 76% 2.5 3.2 128% 5.6 4.3 76% Importsgoods 3.0 2.7 87% 1.8 2.2 123% 4.3 2.9 69% Gasoline 0.6 0.5 96% 0.3 0.5 158% 0.9 0.9 104% Rice 0.5 0.6 13% 0.3 0.1 52% 0.6 0.5 71% Other goods 1.9 1.5 78% 1.2 1.6 131% 2.8 1.6 57% Local goods/DGCI 1.1 0.5 45% 0.7 0.9 139% 1.4 1.3 97% Freezingservices 0.2 0.1 39% 0.0 0.0 0.0 0.0 Combustible 0.1 0.1 56% 0.1 0.1 66% 0.1 0.1 91% Other 0.8 0.3 37% 0.5 0.8 159% 1.3 1.2 98% Tobacco 0.1 0.1 65% 0.0 0.0 0.0 0.0 Other indirect taxes 0.6 0.3 58% 0.1 0.3 264% 0.5 0.3 60% Stamp taxes 0.4 0.2 50% 0.1 0.3 264% 0.5 0.3 50% Democracytax 0.0 0.0 0% 0.0 0.0 0.0 0.0 Other 0.2 0.1 78% 0.0 0.0 0.0 0.1 - 65 - Table A1.6: Breakdownof GovernmentRevenue,2003-05 (inbillionof CFA) (continued) 2003 2004 2005 Budget Actual Exec. Budget Actual Exec. Budget Actual Exec. Solidarity tax 0.1 0.1 89% 0.0 0.0 0.0 0.1 Marine trade tax 0.0 0.0 0% 0.0 0.0 0.0 0.0 2. Non-fiscalrevenue 13.5 8.9 66% 13.6 12.8 94% 9.3 13.7 148% Taxes, fines and penalities 3.8 1.8 48% 5.3 4.7 88% 2.4 5.4 222% Taxes 2.6 2.6 100% 4.8 4.3 91% 1.7 3.2 183% Compensation for fishing licenses 2.2 1.7 77% 4.5 4.2 95% 1.7 3.2 185% Licenses 1.9 1.7 90% 4.2 4.2 100% 1.7 2.5 145% Transportation 0.1 0.0 0% 0.2 0.0 0% 0.0 0.1 Fishery fund 0.2 0.0 0% 0.2 0.0 0% 0.0 0.2 Petroliumactivity 0.0 0.0 0.0 0.0 0.0 0.0 Court o f Accounts services 0.0 0.0 0% 0.0 0.0 0.0 0.0 Police services 0.0 0.0 0.0 0.0 0.0 0.0 446% Passport services 0.1 0.0 25% 0.0 0.0 0% 0.0 0.3 Immigration services 0.0 0.0 3% 0.0 0.0 2% 0.0 0.0 Education services 0.0 0.0 7% 0.0 0.0 0.0 0.0 Port services 0.0 0.0 16% 0.0 0.0 3% 0.0 0.0 Other taxes 0.2 0.1 38% 0.2 0.0 21% 0.0 0.1 290% Militarytax 0.0 0.0 0.0 0.0 0.0 0.0 Taxa de selagem (aduan.) 0.0 0.0 25% 0.0 0.0 0.0 0.0 Civilian registration 0.0 0.0 53% 0.0 0.0 25% 0.0 0.0 131% Justice tax 0.0 0.0 4% 0.0 0.0 33% 0.0 0.0 Registration fees 0.0 0.0 17% 0.0 0.0 3% 0.0 0.0 131% Litigation Court fees 0.0 0.0 0% 0.0 0.0 0% 0.0 0.0 Civilian ID fees 0.0 0.0 124% 0.1 0.0 8% 0.0 0.0 Traffic tax 0.0 0.0 308% 0.1 0.0 6% 0.0 0.0 45% Notary fees 0.0 0.0 31% 0.0 0.0 0Yo 0.0 0.0 Secretariat fees 0.0 0.0 55% 0.0 0.0 32% 0.0 0.0 314% Agricultural veterinary services 0.0 0.0 42% 0.0 0.0 121% 0.0 0.0 211% Agricultural inspection services 0.0 0.0 309% 0.0 0.0 110% 0.0 0.0 122% Other 0.1 0.0 4% 0.0 0.0 119% 0.0 0.0 Fines and penalties 1.2 0.0 2% 0.5 0.4 84% 0.7 1.o 148% Illegal fishing 1.1 0.0 0% 0.5 0.4 86% 0.7 1.o 151% heaction to the fishing code 0.0 0.0 41% 0.0 0.0 163% 0.0 0.0 148% Other fines and penalties 0.1 0.0 34% 0.0 0.0 37% 0.0 0.0 35% Property income 1.8 0.0 0% 0.1 0.0 25% 0.1 0.2 312% Public and parastatal enterprises 0.5 0.0 0% 0.0 0.0 0.0 0.0 BCEAO 0.9 0.0 0% 0.0 0.0 0.0 0.0 Other income 0.4 0.0 0% 0.1 0.0 25% 0.1 0.2 247% Public sector 0.4 0.0 0% 0.1 0.0 25% 0.1 0.2 247% Public sector other -- generalservices 0.0 0.0 0% 0.0 0.0 0.0 0.0 Landincome 0.0 0.0 0% 0.0 0.0 0.0 0.0 Income from construction 0.0 0.0 0% 0.0 0.0 0.0 0.0 Other 0.0 0.0 0% 0.0 0.0 0.0 0.0 Transfers 7.3 5.8 80% 7.7 5.3 69% 5.8 7.9 138% Public administration 0.6 0.1 22% 0.4 0.4 91% 0.7 1.3 201% Autonomous funds 0.3 0.0 7% 0.1 0.0 32% 0.0 0.3 853% Tourism 0.0 0.0 32% 0.0 0.0 33% 0.0 0.0 73% Forest 0.2 0.0 5% 0.0 0.0 34% 0.0 0.0 281% - 66 - Table A1.6: Breakdown of Government Revenue, 2003-05 (in billion of CFA) (end) 2003 2004 2005 Budget Actual Exec. Budget Actual Exec. Budget Actual Exec. Road 0.1 0.0 7% 0.0 0.0 28% 0.0 0.3 Social security 0.3 0.1 41% 0.3 0.4 103% 0.6 1.o 164% Pensions 0.3 0.1 42% 0.3 0.4 104% 0.6 1.o 180% Pensions Sobrevivihcia -- Aposentaqao 0.0 0.0 0% 0.0 0.0 0.0 0.0 Pensionarrears - Sobrevivhcia 0.0 0.0 0% 0.0 0.0 0.0 0.0 0% Assistance for Tuberculosis 0.0 0.0 0% 0.0 0.0 0.0 0.0 0% Other assistance 0.0 0.0 0.0 0.0 0.0 0.0 Exterior 6.6 5.7 85% 7.3 4.9 68% 5.1 6.6 129% Consularservices 0.0 0.0 0% 0.0 0.0 0.0 0.0 EuropeanCommission 6.6 5.7 85% 4.8 3.8 81% 4.8 6.2 131% Financialcompensation 6.0 5.7 95% 4.1 3.8 94% 4.8 4.8 100% Other (scientificprograms) 0.7 0.0 0% 0.0 0.0 0.0 1.5 Other 0.0 0.0 2.5 1.1 0.3 0.3 105% Sales ofgoods andservices 0.7 0.5 68% 0.4 0.6 142% 0.9 1.3 144% Non-durablegoods 0.0 0.0 20% 0.0 0.0 160% 0.0 0.0 177% Services 0.7 0.5 71% 0.4 0.6 150% 0.9 0.6 71% Fishproducts 0.0 0.0 11% 0.0 0.0 22% 0.0 0.7 Domesticmarkets 0.0 0.0 11% 0.0 0.0 22% 0.0 0.7 Foreignmarkets 0.0 0.0 0.0 0.0 0.0 0.0 Sales of capitalgoods 0.0 0.0 0.0 0.0 0.0 0.0 Capitaltransfers 0.0 0.0 34% 0.0 0.1 259% 0.1 0.1 63% Public Sector 0.0 0.0 0% 0.0 0.0 0.1 0.0 0% Other sectors 0.0 0.0 0.0 0.0 0% 0.0 0.0 Returnedfunds 0.0 0.0 64% 0.0 0.1 0.0 0.1 554% Other currentrevenue 0.0 0.0 91% 0.0 0.1 575% 0.0 0.0 126% 3 Extraordinaryrevenue I 7.8 0.0 0% 1.9 0.1 5% 0.0 0.0 Privatizations 7.8 0.0 0% 0.1 0.0 0% 0.2 0.0 0% Public andparastatalenterprises 0.8 0.0 0% 0.0 0.0 0.2 0.0 0% Other 7.0 0.0 0% 0.1 0.0 0% 0.0 0.0 Other 0.0 0.0 1.7 0.1 6% 0.0 0.0 4. Extraordinarytransfers 29.1 10.6 36% 19.4 24.5 126% 5.6 13.3 240% Exterior 28.0 10.6 38% 19.4 24.5 126% 5.6 13.3 240% Foodassistance 1.6 0.0 0% 0.0 0.0 0.0 0.0 Projectsupport 20.9 9.5 45% 12.2 12.5 102% 0.5 8.3 BOP support 5.4 0.6 11% 0.0 6.7 2.1 3.5 164% Budget support 0.0 0.5 7.2 5.3 73% 3.0 1.6 55% Treasury special account 1.2 0.0 0% 0.0 0.0 0.0 0.0 Regionalintegrationsupport 1.2 0.0 0% 0.0 0.0 0.0 0.0 Source: Ministery of Finance - 67 - Table Al. 7: Breakdownof GovernmentRevenue,2003-05 (inpercentageof GDP) 2003 2004 2005 Budget Actual Budget Actual Budget Actual TOTAL REVENUE 50.2 22.9 31.7 34.4 22.3 27.8 Total budgetaryrevenue 23.3 15.2 16.8 17.2 18.6 18.8 1.Fiscalrevenue 13.5 8.7 7.3 8.2 12.3 9.6 Direct Taxes 2.7 2.1 1.8 2.0 2.8 2.3 Incometax 2.6 1.8 1.8 2.0 2.7 2.2 Industrialcontribution 1.4 1.2 0.9 1.1 1.3 1.2 Predialcontribution 0.7 0.3 0.5 0.4 0.5 0.5 Urban 0.1 0.1 0.2 0.0 0.1 0.1 Rural 0.6 0.2 0.4 0.4 0.4 0.4 Professionaltax 0.5 0.3 0.3 0.4 0.9 0.5 Public administration 0.3 0.3 0.1 0.2 0.5 0.2 Other 0.2 0.3 0.2 0.3 0.4 0.3 Complementarytax 0.0 0.0 0.0 0.0 0.0 0.0 Otherdirect taxes 0.0 0.0 0.0 0.0 0.0 0.0 Capitatax 0.0 0.0 0.0 0.0 0.0 0.0 Proprietytax 0.0 0.0 0.0 0.0 0.0 0.0 IndirectTaxes 10.8 6.6 5.5 6.2 9.6 7.3 Tax on internationaltransactions 6.3 2.8 2.9 2.8 4.3 3.3 Import taxes 2.3 1.7 1.1 1.4 2.5 1.9 Gasoline 0.3 0.4 0.2 0.3 0.6 0.6 Rice 0.4 0.4 0.2 0.1 0.5 0.3 Other imports 1.7 0.9 0.7 1.o 1.5 1.o Export taxes 0.0 0.0 0.0 0.0 0.0 0.0 Extraordinarytax on cashewnuts 3.2 0.8 1.2 1.2 1.4 1.2 Customs services 0.0 0.0 0.0 0.0 0.0 0.0 Customsexemptions 0.7 0.0 0.4 0.0 0.0 0.0 Solidaritytax 0.1 0.0 0.2 0.2 0.4 0.2 Consumption tax (IEC) 1.o 1.2 0.7 1.o 1.1 1.o Importsgoods 0.9 0.9 0.7 1.0 1.1 1.o Gasoline 0.4 0.4 0.2 0.3 0.4 0.3 Other goods 0.5 0.5 0.5 0.7 0.7 0.7 Local goosDGCI 0.1 0.3 0.0 0.0 0.0 0.0 Generalsales tax (IGV) 3.1 2.3 1.7 2.2 3.8 2.9 Importsgoods 2.2 1.9 1.3 1.6 2.9 2.0 Gasoline 0.4 0.4 0.2 0.3 0.6 0.6 Rice 0.4 0.4 0.2 0.1 0.4 0.3 Other goods 1.4 1.1 0.9 1.1 1.9 1.1 LocalgoosDGC1 0.8 0.3 0.5 0.7 0.9 0.9 Cervejae Refrigerante-ProduqPo Local 0.1 0.0 0.0 0.0 0.0 0.0 Combustible 0.0 0.1 0.1 0.1 0.1 0.1 Other 0.6 0.2 0.4 0.6 0.8 0.8 Tobacco 0.1 0.1 0.0 0.0 0.0 0.0 Other indirecttaxes 0.4 0.2 0.1 0.2 0.4 0.2 Stamptaxes 0.3 0.2 0.1 0.2 0.4 0.2 Democracytax 0.0 0.0 0.0 0.0 0.0 0.0 - 68 Table A1.7: Breakdownof GovernmentRevenue,2003-05 (in percentageof GDP) (continued) 2003 2004 2005 Budget Actual Budget Actual Budget Actual Other 0.1 0.1 0.0 0.0 0.0 0.0 Solidarity tax 0.1 0.1 0.0 0.0 0.0 0.0 Marinetrade tax 0.0 0.0 0.0 0.0 0.0 0.0 2. Non-fiscal revenue 9.9 6.5 9.5 9.0 6.2 9.2 Taxes, fines and penalities 2.7 1.3 3.7 3.3 1.6 3.6 Taxes 1.9 1.9 3.3 3.0 1.2 2.2 Compensation for fishing licenses 1.6 1.2 3.1 3.0 1.2 2.2 Licenses 1.4 1.2 3.0 3.0 1.2 1.7 Transportation 0.1 0.0 0.2 0.0 0.0 0.1 Fishery find 0.2 0.0 0.1 0.0 0.0 0.2 Petroliumactivity 0.0 0.0 0.0 0.0 0.0 0.0 Court o f Accounts services 0.0 0.0 0.0 0.0 0.0 0.0 Police services 0.0 0.0 0.0 0.0 0.0 0.0 Passport services 0.0 0.0 0.0 0.0 0.0 0.2 Immigration services 0.0 0.0 0.0 0.0 0.0 0.0 Education services 0.0 0.0 0.0 0.0 0.0 0.0 Port services 0.0 0.0 0.0 0.0 0.0 0.0 Other taxes 0.2 0.1 0.2 0.0 0.0 0.0 Militarytax 0.0 0.0 0.0 0.0 0.0 0.0 Taxa de selagem (aduan.) 0.0 0.0 0.0 0.0 0.0 0.0 Civilian registration 0.0 0.0 0.0 0.0 0.0 0.0 Justice tax 0.0 0.0 0.0 0.0 0.0 0.0 Registration fees 0.0 0.0 0.0 0.0 0.0 0.0 Litigation Court fees 0.0 0.0 0.0 0.0 0.0 0.0 CivilianIDfees 0.0 0.0 0.0 0.0 0.0 0.0 Traffic tax 0.0 0.0 0.0 0.0 0.0 0.0 Notary fees 0.0 0.0 0.0 0.0 0.0 0.0 Secretariat fees 0.0 0.0 0.0 0.0 0.0 0.0 Agricultural veterinary services 0.0 0.0 0.0 0.0 0.0 0.0 Agricultural inspection services 0.0 0.0 0.0 0.0 0.0 0.0 Other 0.0 0.0 0.0 0.0 0.0 0.0 Fines and penalities 0.9 0.0 0.4 0.3 0.5 0.7 Illegal fishing 0.8 0.0 0.4 0.3 0.4 0.7 Infractionto the fishing code 0.0 0.0 0.0 0.0 0.0 0.0 Other fines andpenalities 0.0 0.0 0.0 0.0 0.0 0.0 Property income 1.3 0.0 0.1 0.0 0.0 0.1 Public andparastatal enterprises 0.4 0.0 0.0 0.0 0.0 0.0 BCEAO 0.6 0.0 0.0 0.0 0.0 0.0 Other income 0.3 0.0 0.1 0.0 0.0 0.1 Public sector services 0.3 0.0 0.1 0.0 0.0 0.1 Public sector other -- general 0.0 0.0 0.0 0.0 0.0 0.0 Landincome 0.0 0.0 0.0 0.0 0.0 0.0 Income from construction 0.0 0.0 0.0 0.0 0.0 0.0 Other 0.0 0.0 0.0 0.0 0.0 0.0 Transferts 5.3 4.2 5.4 3.7 3.9 5.3 - 69 - Table A1.7: Breakdownof Government Revenue, 2003-05 (inpercentage of GDP)) (end) 2003 2004 2005 Budget Actual Budget Actual Budget Actual Public administration 0.5 0.1 0.3 0.3 0.5 0.9 Autonomous funds 0.2 0.0 0.1 0.0 0.0 0.2 Tourism 0.0 0.0 0.0 0.0 0.0 0.0 Forest 0.2 0.0 0.0 0.0 0.0 0.0 Road 0.1 0.0 0.0 0.0 0.0 0.2 Social security 0.2 0.1 0.2 0.3 0.4 0.7 Pensions 0.2 0.1 0.2 0.2 0.4 0.7 Pensions Sobrevivtncia -- Aposentaqao 0.0 0.0 0.0 0.0 0.0 0.0 Pension arrears - Sobrevivtncia 0.0 0.0 0.0 0.0 0.0 0.0 Assistance for Tuberculosis 0.0 0.0 0.0 0.0 0.0 0.0 Other assistance 0.0 0.0 0.0 0.0 0.0 0.0 Exterior 4.8 4.1 5.1 3.5 3.4 4.4 Consular services 0.0 0.0 0.0 0.0 0.0 0.0 European Commission 4.8 4.1 3.3 2.7 3.2 4.2 Financial compensation 4.4 4.1 2.9 2.7 3.2 3.2 Other (scientific programs) 0.5 0.0 0.0 0.0 0.0 1.o Other 0.0 0.0 1.8 0.8 0.2 0.2 Sales o f goods and services 0.5 0.4 0.3 0.4 0.6 0.9 Non-durable goods 0.0 0.0 0.0 0.0 0.0 0.0 Services 0.5 0.4 0.3 0.4 0.6 0.4 Fish products 0.0 0.0 0.0 0.0 0.0 0.4 Domestic markets 0.0 0.0 0.0 0.0 0.0 0.4 Foreignmarkets 0.0 0.0 0.0 0.0 0.0 0.0 Sales o f capital goods 0.0 0.0 0.0 0.0 0.0 0.0 Capital transfers 0.0 0.0 0.0 0.0 0.1 0.0 Public Sector 0.0 0.0 0.0 0.0 0.1 0.0 Other sectors 0.0 0.0 0.0 0.0 0.0 0.0 Returned funds 0.0 0.0 0.0 0.0 0.0 0.0 Other current revenue 0.0 0.0 0.0 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 3. Extraordinaryrevenue 5.7 0.0 1.3 0.1 0.0 0.0 Privatizations 5.7 0.0 0.1 0.0 0.1 0.0 Public and parastatal enterprises 0.5 0.0 0.0 0.0 0.1 0.0 Other 5.1 0.0 0.1 0.0 0.0 0.0 Other 0.0 0.0 1.2 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 4. Extraordinary transfers 21.2 7.7 13.6 17.2 3.7 9.0 Exterior 20.4 7.7 13.6 17.2 3.7 9.0 Food assistance 1.2 0.0 0.0 0.0 0.0 0.0 Project support 15.2 6.9 8.6 8.8 0.3 5.5 BOP support 4.0 0.5 0.0 4.7 1.4 2.3 Budget support 0.0 0.4 5.1 3.7 2.0 1.1 Treasury special account 0.9 0.0 0.0 0.0 0.0 0.0 Regional integration support 0.9 0.0 0.0 0.0 0.0 0.0 Source: Ministery o f Finance - 70 - TableAl. 8: Breakdownof GovernmentExpenditure,2003-05 (inbillionof CFA) 2003 2004 2005 2006 Budget4 Actual Exec. Budget Actual Exec. Budget SMP Actual Exec. Budget Total expenditure 44.1 52.9 120% 50.2 69.8 139% 62.5 40.4 60.4 149% 83.2 Current expenditure 38.6 35.0 91% 28.9 41.5 143% 37.4 38.8 43.9 113% 43.8 Personnelexpenditure 15.7 13.6 87% 15.5 16.2 104% 21.4 23.0 21.3 93% 21.1 Wage and salaries 9.1 6.3 70% 9.5 10.1 107% 17.5 18.6 18.9 102% 16.6 Bonuses 2.8 4.6 165% 2.5 2.5 100% 0.5 0.8 0.8 108% 0.6 Socialsecurity 3.9 2.7 70% 3.6 3.6 100% 3.4 3.6 1.6 44% 3.9 Goods andservices 9.2 3.9 43% 6.6 4.5 69% 5.8 6.6 7.6 115% 7.2 Durablegoods 1.3 0.4 28% 0.4 0.4 95% 0.7 0.7 0.4 55% 0.8 Nondurablegoods 3.2 3.2 101% 3.4 3.0 89% 2.6 2.6 3.0 115% 3.5 Services 4.7 0.4 7% 2.8 1.1 41% 2.5 3.3 4.2 127% 2.9 Transfers 5.5 5.1 92% 3.9 5.0 128% 4.4 4.2 6.1 146% 6.8 Non financialpublic firms 0.2 0.0 28% 0.1 0.1 100% 0.1 0.0 0.0 100% 0.2 Public administration 2.3 2.3 99% 1.5 1.5 100% 1.9 2.1 3.1 148% 2.6 Nonprofit privateinstitutions 0.2 0.1 67% 0.1 0.1 100% 0.1 0.1 0.2 106% 0.1 Financialinstitutions 0.8 0.6 72% 0.0 0.6 ... 0.0 0.0 0.0 ... 0.0 Institutionsfor security 0.0 0.0 ... 0.0 0.5 ... 0.0 0.0 0.9 ... 0.0 Families(scholarships, transport) 0.8 0.8 98% 0.7 0.7 100% 0.7 0.5 OS 100% 2.6 Foreign(embassies andinternational institutions) 1.3 1.3 100% 1.6 1.6 100% 1.6 1.5 1.5 100% 1.2 Other current expenditure 5.2 4.5 87% 1.8 7.0 385% 4.6 3.1 2.4 75% 3.4 Provisionalallocation 0.0 0.0 ... 0.5 0.5 100% 0.0 0.0 0.0 ... 0.0 Restitutionsofrevenue 1.8 1.1 62% 1.2 1.2 100% 1.1 1.o 0.7 68% 1.9 Commonexpenditure 3.4 3.4 100% 0.1 5.3 4417% 3.4 2.1 1.7 79% 1.5 Interest 1.6 7.8 478% 0.6 8.8 1458% 0.8 1.3 6.5 492% 3.6 Guarantieddebt 0.0 0.0 ... 0.6 0.1 9% 0.8 1.3 0.0 0?'h 3.6 InternalDebt 0.0 0.6 1595% ... 2.7 ... 0.4 0.3 0.5 131% 0.3 Commercialbanks ... ... ... ... ... ... 0.0 0.0 0.0 ... 0.0 EX-BCG ... .,* ... ... ... ... 0.1 0.1 0.0 0% 0.1 EX-BCN ... ... ... ... ... ... 0.2 0.1 0.0 0% 0.1 Statutary advances ... ... ... ... .., ... 0.1 0.1 0.0 0% 0.1 Externaldebt 1.6 7.3 452% ... 6.0 ... 0.4 1.o 6.1 619% 3.2 0.w: Guarantieddebt ... ... ... ... ... ... 0.0 0.0 0.5 ... 0.0 Arrears ... ... ... ... ... ... 0.4 1.o 0.2 15% 0.0 World Bank ... ... ... ... *.. ... 0.1 0.7 0.1 15% 1.6 FMI ... .., ... ... ... ... 0.0 0.1 0.0 0% 1.3 Other 1.4 0.0 0% 0.5 0.0 0% 0.5 0.6 0.0 0% 1.7 Capitalexpenditure 5.4 17.9 329% 21.3 28.3 133% 25.0 1.6 16.5 1031% 39.4 Domesticallyfinancedpublic invest. 1.7 2.7 155% 1.2 2.0 164% 1.6 1.6 1.7 106% 8.4 Ofwhich: Rural development ... .,. ... ... ... ... 0.0 0.0 0.1 ... 0.0 Infrastructure ... ... ... ... ... ... 0.0 0.0 0.4 ... 0.0 Health ... , * . ... ... ... ..* 0.0 0.0 0.7 ... ... 0.0 Education ... ... ... ... ... ... 0.0 0.0 0.3 0.0 Capacitybuilding ... ... ... ... ... ... 0.0 0.0 0.2 ... 0.0 Externally financedpublic invest. 3.7 15.2 411% 20.1 26.3 131% 23.4 0.0 14.8 ... 30.9 Source: MOF /1 No budget was preparedfor 2003, therefore, allocationsare basedon the 2002 budget. - 71 - Table Al. 9: Breakdownof GovernmentExpenditure,2003-05 (inpercentageof GDP) 2003 2004 2005 2006 Budget/l Actual Budget Actual Budget SMP Actual Budget Currentexpenditure 28.2 25.5 20.3 29.1 25.2 26.1 29.5 27.7 Personnelexpenditure 11.5 9.9 10.9 11.3 14.4 15.5 14.3 13.4 Wage and salaries 6.6 4.6 6.6 7.1 11.8 12.5 12.7 10.5 Bonuses 2.0 3.3 1.7 1.7 0.3 0.5 0.6 0.4 Socialsecurity 2.8 2.0 2.5 2.5 2.3 2.4 1.1 2.5 Goodsandservices 6.7 2.9 4.6 3.2 3.9 4.5 5.1 4.6 Durablegoods 0.9 0.3 0.3 0.3 0.5 0.5 0.3 0.5 Nondurable goods 2.3 2.4 2.4 2.1 1.8 1.7 2.0 2.2 Services 3.4 0.3 2.0 0.8 1.7 2.2 2.8 1.8 Transfers 4.0 3.7 2.8 3.5 3.0 2.8 4.1 4.3 Nonfinancialpublic firms 0.1 0.0 0.1 0.1 0.1 0.0 0.0 0.1 Publicadministration 1.7 1.7 1.o 1.o 1.2 1.4 2.1 1.7 Nonprofit privateinstitutions 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 Financialinstitutions 0.6 0.4 0.0 0.4 0.0 0.0 0.0 0.0 Institutionsfor security 0.0 0.0 0.0 0.4 0.0 0.0 0.6 0.0 Families(scholarships, transport subsidies) 0.6 0.6 0.5 0.5 0.5 0.3 0.3 1.7 Foreign(embassies andinternationalinstitutions) 0.9 0.9 1.1 1.1 1.1 1.0 1.o 0.8 Other current expenditure 3.8 3.3 1.3 4.9 3.1 2.1 1.6 2.1 Provisionalallocation 0.0 0.0 0.4 0.4 0.0 0.0 0.0 0.0 Restitutionsofrevenue 1.3 0.8 0.8 0.8 0.8 0.7 0.5 1.2 Commonexpenditure 2.5 2.5 0.1 3.7 2.3 1.4 1.1 0.9 Interest 1.2 5.7 0.4 6.1 0.5 0.9 4.4 2.3 Guarantieddebt 0.0 0.0 0.4 0.0 0.5 0.9 0.0 2.3 InternalDebt ....* . ... ... ... 0.3 0.2 0.3 0.2 Commercialbanks ... ... *.. 0.0 0.0 0.0 0.0 EX-BCG . * * ... ... ... 0.1 0.1 0.0 0.1 EX-BCN *.. ... ... ... 0.1 0.1 0.0 0.1 Statutory advances ... ... ... ... 0.1 0.0 0.0 0.0 Externaldebt ... ... ... ... 0.3 0.7 4.1 2.0 0.w: Guarantieddebt ... ... ,.. ... 0.0 0.0 0.3 0.0 Arrears ... ... ... ... 0.3 0.7 0.1 0.0 World Bank ... ... ... ... 0.1 0.5 0.1 1.o FMI ... ... ... ... 0.0 0.1 0.0 0.8 Other (recurrent costs on capitalexpenditure) 1.o 0.0 0.4 0.0 0.4 0.4 0.0 1.1 Capitalexpenditure 4.0 13.1 14.9 19.8 16.8 1.1 11.1 24.9 Domesticallyfinanced public investment 1.3 2.0 0.9 1.4 1.1 1.1 1.1 5.3 Ofwhich: Ruraldevelopment ... ... ... ... 0.0 0.0 0.0 0.0 Infrastructure , * * ... ... ... 0.0 0.0 0.3 0.0 Health .,. . * * ... ..* 0.0 0.0 0.5 0.0 Education ... ... ... ... 0.0 0.0 0.2 0.0 Capacitybuilding/economicmanagement ... ... ... ... 0.0 0.0 0.1 0.0 Externallyfinancedpublic investment 2.7 11.1 14.1 18.4 15.8 0.0 10.0 19.6 Source:MOF /1 No budgetwas preparedfor 2003, therefore, allocationsare basedon the 2002 budget. - 72 - TableAl. 10: FunctionalClassificationof CurrentprimaryExpenditure,2005-06 2005 2006 Budget Actual Exec. Budget (inbillionofCFA; figures in%are yearly execution rates) Socio-educative sectors 8.2 8.3 101% 9.9 0.w: Health 2.6 2.6 100% 3.0 Education 5.5 5.5 100% 5.3 Economic sectors 10 5.3 53% 11.1 0.w: Agriculture 0.5 0.5 100% 0.5 Fisheries 1.6 0.2 14% 0.2 Infrastructure 0.5 0.5 100% 0.7 Industry and energy 0.1 0.1 110% 0.2 Institutional sectors 4.6 3.8 83% 3.7 0.w: Executive /1 1.o 1.0 100% 1.1 Parliament ... 1.o ... 1.5 Judiciary 3.6 1.8 49% 0.4 Security secrtor 10.4 10.4 100% 11.4 Other 3.5 9.6 274% 4.1 Total 36.7 37.4 102% 40.2 (inpercentage ofGDP) Socio-educative sectors 5.5 5.6 6.3 0.w: Health 1.7 1.7 1.9 Education 3.7 3.7 3.3 Economic sectors 6.7 3.6 7.0 0.w: Agriculture 0.3 0.3 0.3 Fisheries 1.1 0.2 0.1 Infrastructure 0.3 0.3 0.5 Industryandenergy 0.0 0.1 0.1 Institutional sectors 3.1 2.6 2.3 0.w:Executive/1 0.7 0.7 0.7 Parliament ... 0.7 1.o Judiciary 2.4 1.2 0.2 Security secrtor 7.0 7.0 7.2 Other 2.4 6.5 2.6 Total 24.7 25.1 25.5 Memo item: GDP inbillion of CFA 148.7 157.9 Source: M O F /1 Presidency and Prime Ministry - 73 - Table Al. 11: Public Investment by Sector, 2003-08 Project Areas 2003 2004 2005 2006 2007 2008 Actual Programmed (inbillionofCFA francs) RuralDevelopment 0.2 0.6 0.1 ... ... ... Hydraulic Resources 2.5 0.5 0.3 ... ... ... Industry 0.0 2.9 3.0 ... ... ... Energy 0.0 0.0 0.0 ... ... ... Public Works /1 1.3 2.5 0.1 13.6 17.1 15.7 Transports 0.0 2.3 3.5 ... ... ... Health 2.0 2.7 1.4 3.0 7.3 8.9 Education 1.7 2.7 1.o 5.1 7.9 10.6 Social Affairs 0.6 0.5 0.7 0.5 2.4 2.5 Economic Management 6.6 1.8 1.9 12.6 6.5 6.2 Regional Development 0.1 0.0 0.0 0.3 0.6 0.4 Other 2.8 16.9 11.0 4.2 0.0 0.0 Total 17.9 28.3 16.5 39.3 41.8 44.3 (inpercentage ofGDP) Rural Development 0.2 0.4 0.1 ... ... ... Hydraulic Resources 1.8 0.4 0.2 ... ... ... Industry 0.0 2.0 2.0 ... .., ... Energy 0.0 0.0 0.0 ... ... ... Public Works /1 0.9 1.7 0.0 8.6 10.2 8.9 Transports 0.0 1.6 2.3 ... ... ... Health 1.5 1.9 0.9 1.9 4.3 5.0 Education 1.3 1.9 0.7 3.2 4.7 6.0 Social Affairs 0.4 0.4 0.5 0.3 1.4 1.4 Economic Management 4.8 1.2 1.3 8.0 3.9 3.5 Regional Development 0.1 0.0 0.0 0.2 0.3 0.2 Other 2.1 11.9 7.4 2.7 0.0 0.0 Total 13.0 19.8 11.1 24.9 24.9 25.0 Meno Items: Nominal GDP inbilliono f CFA 137.1 142.6 (48.7 157.9 (67.4 177.4 Source: General Direction o f Planning, Ministryo f Economy. - 74 - Table Al. 12: PIPProgrammedandExecutedin2003 (inbillionCFA francs unless otherwiseindicated) 2003 Programmed/1 Execution Overall ProjectAreas Ext. Grant Ext. Loan Dom. Fin Total Ext. Grant Ext. Loan Dom. Fin Total Execution Rate RuralDevelopment 0.5 0.0 0.1 0.6 0.2 0.0 0.0 0.2 39.0% Fishing 0.7 0.4 0.0 1.1 0.2 0.1 0.0 0.3 26.0% Mining 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0% Hydraulic Resources 0.4 0.3 0.1 0.7 1.2 1.3 0.0 2.5 344.6% Industry 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0% Energy 0.1 0.2 0.3 0.6 0.0 0.0 0.0 0.0 0.0% PublicWorks 7.0 1.8 1.4 10.1 0.7 0.5 0.0 1.3 12.4% Transports 0 0 0 0.0 0.0 0.0 0.0 0.0 0.0% Tourism 0.3 0.0 0.1 0.4 0.0 0.1 0.0 0.1 19.0% Health 0.6 0.6 0.1 1.3 1.2 0.8 0.0 2.0 159.4% Education 0.6 0.5 0.1 1.1 0.2 1.5 0.0 1.7 151.2% Social Affairs 0.6 0.0 0.1 0.6 0.6 0.0 0.0 0.6 95.7% EconomicManagement 4.9 2.1 0.2 1.2 4.9 1.4 0.2 6.6 91.0% RegionalDevelopment 2.1 0.0 0.1 2.2 0.1 0.0 0.0 0.1 0.0% Information 0.2 0.0 0.0 0.2 0.1 0.0 0.0 0.1 23.4% Other 0.0 0.0 0.0 0.0 0.0 0.0 2.4 2.4 Total 17.8 5.9 2.5 26.2 9.4 5.8 2.7 17.9 68.1% Source: GeneralDirection ofPlanning, Ministry of Economy. /1 Nobudget was preparedfor 2003, therefore, allocationsare basedon the 2002 budget. - 75 - Table Al. 13: PIPProgrammedandExecutedin2004 (inbillionCFA francs unless otherwiseindicated) 2004 Programmed Execution Overall ProjectAreas Ext. Grant Ext. Loan Dom. Fin Total Ext. Grant Ext. Loan Dom. Fin Total Execution Ratein% Rural Development 7.3 3.8 0.2 11.3 0.3 0.3 0.0 0.6 5.2% Fishing 0.4 1.2 0.0 1.7 0.0 0.0 0.0 0.0 0.2% Mining 0.7 0.0 0.2 0.9 0.0 0.0 0.0 0.0 0.0% Hydraulic Resources 2.7 0.0 0.0 2.7 0.3 0.2 0.0 0.5 19.2% Industry 0.0 0.0 0.2 0.2 0.0 2.9 0.0 2.9 1232.3% Energy 0.2 1.4 0.0 1.6 0.0 0.0 0.0 0.0 0.0% Public Works 36.1 8.2 0.3 44.6 2.3 0.0 0.2 2.5 5.5% Transports 2.8 0.0 0.1 2.9 1.7 0.4 0.1 2.3 80.0% Tourism 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0% Health 1.6 0.0 0.0 1.6 1.5 1.2 0.0 2.7 170.9% Education 1.o 0.4 0.0 1.4 0.3 2.5 0.0 2.7 198.9% Social Affairs 1.3 0.3 0.0 1.6 0.5 0.0 0.0 0.5 33.7% EconomicManagement 2.1 3.9 0.1 6.1 1.o 0.7 0.1 1.8 29.4% RegionalDevelopment 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0% Information 1.3 0.0 0.2 1.6 0.0 0.0 0.0 0.0 0.0% Other 0.0 0.0 0.0 0.0 5.2 5 1.5 11.7 Total 57.3 19.3 1.4 78.0 13.0 13.2 2.0 28.3 36.2% Source: GeneralDirection ofPlanning,Ministry of Economy. - 76 - Table Al. 14: PIPProgrammedandExecutedin2005 (inbillionCFAfrancsunless otherwiseindicated) 2005 Programmed Exucution Overall ProjectAreas Ext.Grant Ext. Loan Dom. Fin Total Ext. Grant Ext.Loan Dom.Fin Total Execution RuralDevelopment 8.1 4.4 0.1 12.6 0.1 0.0 0.0 0.1 0.8% Fishing 0.5 1.3 0.0 1.8 0.0 0.0 0.0 0.1 3.0% Mining 0.5 0.0 0.1 0.6 0.0 0.0 0.0 0.0 0.0% HydraulicResources 3.0 0.0 0.0 3.0 0.2 0.0 0.0 0.3 8.3% Industry 0.0 0.0 0.2 0.2 0.1 2.9 0.0 3.0 1844.3% Energy 0.1 2.3 0.0 2.4 0.0 0.0 0.0 0.0 0.0% PublicWorks 25.6 8.8 0.0 34.4 0.0 0.0 0.1 0.1 0.2% Transports 1.7 0.0 0.1 1.8 3.O 0.1 0.3 3.5 194.2% Tourism 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0% Health 1.5 0.0 0.0 1.5 0.2 1.2 0.0 1.4 91.9% Education 0.8 0.5 0.0 1.2 0.1 0.6 0.3 1.0 83.9% Social Affairs 1.7 0.5 0.0 2.2 0.7 0.0 0.0 0.7 33.9% EconomicManagement 1.1 3.8 0.1 4.9 0.3 1.3 0.2 1.9 38.9% RegionalDevelopment 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0% Information 1.2 0.0 0.2 1.4 0.0 0.0 0.0 0.0 0.0% Other 0.0 0.0 0.0 0.0 1.9 1.9 0.7 4.5 Total 45.8 21.5 0.8 68.2 6.7 8.1 1.7 16.5 24.2% Source: GeneralDirectionofPlanning,Ministry ofEconomy. - 77 - I co b I 5 3 g5 x x x x X X X e g x x m -c 5 x x x x x x x x x x x x X Q P 13 5 P U '28 P .r .E PE sc iEs 'i e .E Be, - I , , I I , 3 22 I f Annex 3: Analysis of Tax Revenue Productivity Tax Buoyancy and Elasticity - Definition and Methodology Tax Buoyancy. The buoyancy o f a tax i s defined as the increase inthe revenue collected compared with the relative increase in the GDP or the GNI. The change in revenue includes any effect o f changes in the tax system, including discretionary changes in the tax structure (Barth and Hemphill, 200056).The estimated coefficient o f tax buoyancy (a)can be derived econometrically by regressing the logarithm o f the actual tax receipts (T) on the logarithm o fthe GDP or GNI(Y)inthe following form: IogT, =a, p, logY, + (1) Tax Elasticity. The elasticity o f a tax is defined as the relative change in revenue from tax compared with the relative change inthe tax base (proxied with the GDP or the GNI), holding the structure o f the tax (tax rates, brackets, coverage, exemptions, and deductions) (Barth and Hemphill, 2000). Tax elasticity can also be estimated using the dummy variable (D) technique, which is an econometric method (Manasan and Querubi, 198657)o f simultaneously adjusting for discretionaryrevenue effects inequation (1). logT, =a,+p,logY,+y,D,+p,D,logY, (2) The estimated coefficient b, is the tax elasticity net o f discretionary tax policies5*. A tax system is elastic if it has an elasticity value greater than 1, suggesting that tax revenues are increasing at a higher rate that the tax base without discretionary changes in tax policy (i.e. adoption o f new taxes or increases in tax rates). A tax elasticity greater thanone maybedesirable inatax system andshouldbe encouraged incountriesinwhich government expenditure tend to increase more rapidly than GDP. The tax system i s likely to be elastic with respect to the tax base when taxes are levied on growing economic sectors and when tax rates are progressive. The tax system will be inelastic if taxes are specific rather than ad valorem, and/or taxes are not collected promptly. This last point is especially important in the case o f high inflation, as an unduly long lag between the assessment and collection o f taxes would erode the real value o f tax revenue (the so- called Tanzi effect). An elastic tax system i s also useful from the point o f view o f economic growth, which generally calls for sustained rise in spending on social and economic infrastructure maintenance. Ifthese expenditure increases are not accompanied by rising revenues, they can leadto undue reliance on deficit financing, either external or domestic. '`Barth, ''Manasan R.andW. Hemphill(2000) "Financial ProgrammingandPolicy:The Case of Turkey", IMF Institute R.G.and R.G. Querubin (1986) "Revenue Performanceof National Government Taxes: 1975-1985" Philippine Institute for DevelopmentStudies, Staff PaperNo. 8701. http://dirp4.pids.gov.ph/ris/sp/pidssp87O1 .pdf 58(b,+1,)i s also usedto measure the change in the level ofrevenueyield due to the discretionary changeafter adjustingfor a change inthe elasticityofthe tax system. - 81 - EstimationResults In the case of Guinea-Bissau the tax buoyancy and elasticity of total, direct and indirect tax revenue i s estimated usingequation 1and 2 above. As a reminder, total tax revenues stand for the sumofdirect andindirecttax revenues. Tax Buoyancy. With regard to buoyancy, the results indicate an estimated coefficient less than one for total tax revenues (bo o.l in Equation 1.1), and a coefficient equal and greater than one I for direct and indirect tax revenues, respectively (bo 1and 1.6 in Equations 1.3 and 1.2 I respectively). These results suggestthat while direct and indirect taxes" have grown more rapidly than the GNI, total tax revenues have increased less rapidly than the national income. This seems to suggest that discretionary tax policies may have had a deleterious impact on total tax revenues despite the positive effect on direct and indirect taxes. The results mayjustify the low tax-to-GDP ratio that the country experienced over the past decade (on average less than 12percent). Buoyancy o f total tax revenue: lOgT, =-0.24+0.l1logY, (0.25) (0.03)' Adj. R2 = 0.42 Prob > F = 0.00 Root MSE = 0.15 Obs.26 Buoyancy of indirect tax revenue: logT, = -7.71 1.6110gY, + (1.98) (1.97)' Adj. R2 = 0.72 Prob > F = 0.00 Root MSE = 1.17 Obs.26 Buoyancy o f direct tax revenue: lOgT, =-1.84+ 0.9910gY, (0.36)' (0.03)' Adj. R2 = 0.98 Prob >F = 0.00 Root MSE = 1.15 Obs.22 Tax Elasticity. The elasticity analysis is performed using the dummy variable technique described in Equation 2 above to `clean' the tax series of discretionary tax policy effects. A binomial dummy variable -- with a value 1for the years during which a discretionary tax reform was carried out, and a value. 0 otherwise --was introduced in the equation. The discretionary policies considered are the following6': 0 Business profit tax: starting with the 1997 budget public agencies have been required to withhold 3 percent o f their payments to contractors for supply of good and services. This withhold constitutes an advance on the businessprofit tax due. 0 Import duties: customs tariff introduced in 1980 (Decree 39/80) rate ranging from 0 to 60 percent levied on cost-insurance-fret (c.i.0 value o f imported goods; customs tax ''Indirecttaxesare essentially composed of the consumptiontax (IGV) andthe sales taxes (IEC). IMF(2001) "Guinea-Bissau StatisticalAppendix"Country ReportNo. 01/23, January 2001. - 82 - exemptions established by ministerial order in 1995; and port service charges introduced in 1983(Decree 36/A-83) 6-10 percentadvalorem onfiscal value ofimports. Consumption tax: introduced in 1983 (Decree 36/A-83), rates ranging from 5 to 85 percent and. 0 Extraordinary tax on cashew: reduced to 10 and 6 percent of the FOB value in 1997 and 2002, respectively(was 34 percent in 1989 and 12.5 1996). The results indicate that when the series of the actual tax collected is adjusted for discretionary tax policiesmentionedabove, the total tax revenue remains inelastic ( A 0.1 inEquation2.1). I Elasticity o f total tax revenue: lOgT, =0.09 O.O81OgY,-1.310, 0.120, .logy, + + (0.27) (0.03)' (0.57)' (0.06)' Adj. R2 = 0.51 Prob > F = 0.00 Root MSE =0.14 Obs.26 Elasticity o f indirect tax revenue: IOgT, =-5.57+1.4110gY, -8.750, +0.840,.10gY, (2.17) (0.21)' (4.53)" (0.45)" Adj. R 2 = 0.45 Prob > F = 0.00 Root MSE = 1.13 Obs.26 Elasticity o f direct tax revenue: logT, =-2.5 1 1.05110gY, + +1.040, -0.1OD,.logy (0.56)' (0.05)' (0.75) (0.73) (2.3) Adj. R2 = 0.98 Prob > F = 0.00 Root MSE = 1.15 Obs.22 This result suggests that total tax revenues are increasing at a lower rate that the tax base without the discretionary changes intax policy. By contrast the indirect and direct tax revenues are elastic The tax system i s likely to be elastic with respect to the tax base when taxes are levied on growing economic sectors and when tax rates are progressive. The tax system will be inelastic if taxes are specific rather than ad valorem, andor taxes are not collected promptly. This last point i s especially important inthe case o f highinflation, as an unduly long lagbetweenthe assessment and collection of taxes would erode the real value of tax revenue (the so-called Tanzi effect). An elastic tax systemis also useful fromthe point o f view o f economic growth, which generally calls for sustained rise in spending on social and economic infrastructure maintenance. If these expenditure increasesare not accompanied by risingrevenues, they can leadto undue reliance on deficit financing, either external or domestic. - 83 - Annex 4: Public Sector Wages Table A4.1: PublicSector Wages and Categoriesof Skills Numberof Mean Standard Minimum Maximum observations Wage Deviation Wage Wage A 11 820 789,2 351 924,8 453 216,l 1 127637,O B 301 533 465,s 254 693,8 418 245,s 1 619 841,O C 267 423 801,O 18 303,s 404 590,s 489 743,7 D 169 389 802,l 28 285,3 370 818,8 448 856,4 E 534 365 128,l 138 002,2 337 047,2 1 623 017,O F 660 317 938,8 29 991,6 303 275,s 414 944,6 G 1174 285 748,9 28 338,l 283 143,3 956 490,2 H 1124 276 454,9 24 464,2 262 652,9 865 389,3 I 1372 259 648,l 14 691,O 249 371,6 301 952,O J 1360 260 854,8 17 509,7 246 491,l 354 283,3 K 1720 249 191,3 14210,l 242 894,4 329 448,9 L 889 251 881,l 15 612,4 239 671,4 323 516,7 M 1424 252 842,l 11 877,l 236 432,8 283 719,4 N 1349 233 571,6 4 305,7 232 478,O 267 355,l 0 59 231 761,8 0,o 231 761,8 231 761,8 P 81 230 698,6 70,s 230 687,4 231 139,O Q 162 229 670,7 611,7 229 613,l 237 367,O R 301 228 167,7 45,8 228 165,O 228 959,l S 212 224 699,3 1683,l 224 568,3 249 044,6 T 301 222 772,8 91,l 222 762,2 223 556,3 U 357 220 345,6 1 268,7 220 271,O 244 217,9 V 703 219 564,s 307,s 219 539,2 227 293,l X 229 218466,4 23,7 218 464,8 218 823,O Y 91 217229,3 127,7 217 203,7 217 904,3 2 1113 215 702,7 1249,l 215 599,9 239 546,8 Total 15 963 265 247,7 73 619,O 215 599,9 1 623 017,O Source: Staff estimatesbased on 1997 census of civil servants. - 84 - Table A4.2: Public Sector Wages by Categoryof Skills andEmployment,2005 Numberof Mean Standard Minimum Maximum Observations Wage Deviation Wage Wage A 5 1 353 600,O 1 285 397,O 316 800,O 3 600 000,O B 160 1 894 500,O 1 066 927,O 230 400,O 5 414 400,O C 108 1092 800,O 353 710,4 288 000,O 2 880 000,O D 45 1 303 680,O 677 103,s 316 800,O 2 880 000,O E 81 1 106844,O 473 274,8 230 400,O 3 254 400,O F 412 1 007 930,O 392 381,2 259 200,O 5 414 400,O G 98 1 239 869,O 1 092 941,O 230 400,O 5 414 400,O H 445 1 083 398,O 728 582,4 230 400,O 4 867 200,O I 728 697 134,l 280 060,9 230 400,O 3 600 000,O J 984 599 092,7 314 500,2 230 400,O 6 307 200,O K 396 544 000,o 413 431,7 230 400,O 4 867 200,O L 562 656 301,8 444 364,3 230 400,O 3 600 000,O M 1124 493 135,9 355 281,4 230 400,O 4 867 200,O N 1179 383 316,O 205 059,4 230 400,O 3 254 400,O 0 295 557 060,3 236 087,s 259 200,O 2 793 600,O P 59 533 532,2 851 921,5 230 400,O 4 867 200,O Q 282 504 102,l 367 040,9 230 400,O 3 600 000,O R 71 286 3773 111 317,3 230 400,O 950 400,O S 241 725 616,6 796 559,6 230 400,O 5 414 400,O T 226 394 279,6 379 763,8 230 400,O 3 600 000,O U 203 457 111,3 442 798,9 230 400,O 3 254 400,O V 117 308 923,l 214 476,O 230 400,O 1 324 800,O W 217 284 151,2 289 337,7 230 400,O 2 880 000,O X 767 447 432,6 209 100,3 230 400,O 2 102 400,O Y 64 414 900,O 509 684,7 230 400,O 2 880 000,O Z 897 258 525,8 140 849,3 230 400,O 2 793 600,O m:Staffestimatesbasedon Total 10 258 594 838,8 539 317,9 230 400,O 8 092 800,O 2005 census of civil servants. Table A4.3: Componentsof the Government's Wage Billin1997and2005 1997 budget 1997 in2005 2005 Percentage Budget CFAF budget increase Total public wage bill 6.259 12.111 21.378 76.5% Wages for regular personnel 2.249 4.352 12.846 195.2% Wages for other personnel 0.513 0.993 0.385 -61-2% Benefits 0.378 0.731 3.920 435.9% Other wage costs 0.029 0.056 0.348 520.2% Occasional benefits 1.467 2.839 0.471 -83.4% Social security andpensions 1.624 3.142 3.407 8.4% m: Wodon, Creppy and Tsimpo (2006) - 85 - Box A4.1: Wage and Poverty A number o f statistics based onthe growth incidence curve are reported inthe table below. The "rate o f pro-poor growth" is computed as the mean growthrate inwages for the poor, or equivalently the percentage change inthe so-called Watts index o f poverty. The poor are definedina flexible way as those who are inthe bottomdecile or wages, the bottom 15 percent, the bottomtwo deciles, etc. It canbe clearly seen that the rate of growth inwages has beenhighest at the top part of the distribution. Growthrates Growthrate inmean 5,68 Growthrate at median 2,42 Meanpercentile growth rate 4,35 Corresponding percentile Rate o f pro-poor growth 10,o -3,74 15,O -3,80 20,o -3,69 25,O -3,08 30,O -2,68 40,O -2,05 50,O -1,33 60,O -0,40 70,O 0,39 80,O 1,21 90,O 2,25 m:Wodon, Creppy and Tsimpo (2006a). Growth IncidenceCurve All, 1997-2005 -Growth ......Growth IncidenceCurve -.-. rateinmean Growth rate inmedian .................................................................. -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-. / .. .......................... .-.-.-.-.-.-.-.-.-.-._._._._._._._. 5 1 0 -4- I O 2 40 50 60 70 80 90 Percentileofcivil servantsrankedby wage level in 1997(in2005 monetary unit) a: Staff estimatesbasedon the 1997and2005 censuses of civil servants m:Wodon, Creppy and Tsimpo (2006a) - 86 - Box A4.2: Determinantsof the ChangeinWages In order to analyze the determinants of changes in wages, a panel has been created with those civil servants who are both inthe 1997 and 2005 databases. The panel contains 6,845 workers. The categories A to Nhave been regrouped into four levels, as suggestedby De Weerd (2001): Level 1(categories A to C), Level 2 (D to F), Level 3 (G to J), and Level 4 (K to N). The categories 0 to Z have been kept together. The table below show the econometric results o f the regression o f the change inwage between 1997 and 2005 for each worker as a function o f a range o f characteristics including sex (male versus female), marital status (married, single, divorced, separated, or widow), the age in2005 and the square o f the age, the education level o f the worker (no education, primary education, secondary education, and superior education), and finally the occupational category (Level 1corresponding to A to C, Level 2 or D to F, Level 3 or G to J, Level 4 or K to N, and finally the categories 0 to Z). The results suggest that male workers benefited from a higher wage increase (gain o f about CFAF37,500). The impact o f marital status and age is not statistically significant. There are also no statistically significant differences inwage gains between workers with no education and workers with a primary education (most o f those workers belong to categories 0 to Z). Workers with a secondary education obtained an increase inwage o f 28,000 CFAF more than workers with no education, and for the workers with a superior education, the gain was about CFAF 221,000. Incomparison to workers incategories 0 to Z, and controlling for the other variables in the regression, workers incategories A to C gained an additional 471,000 CFAF. The corresponding value for categories D to F was even higher, at 633,500 CFAF. The gains were lower, but still large, for categories Gto J (287,000 CFAF) and categories K to N(82,000 CFAF). Determinant of the change in wages between 1997 and 2005 Coef. T- Stat Sex Male 37 499,97 [5.18]*** Female Marital Status Married 18 386,60 [1.421 Single 18 081,46 [1.331 Other (divorced, separated, widow) Age in2005 166,18 [0.04] Square o f the age in2005 -16,96 [0.42] Education No education Primary education -12 096,83 [1.79]* Secondary education 27 888,19 [15.171*** [2.68]*** Superior education 220 931,99 Occupational category Level 1(A to C) 470 822,68 [13.01]*** Level 2 (D to F) 633 584,85 [21-16] *** Level 3 (Gto J) 287 158,98 [29.01]*** Level 4 (K to N) 81 578,OO [10.20]*** Categories 0 to Z Constant -78 891,61 [0.82] Observations 6 845,OO R-squared 0.37 Robust t statistics inbrackets, *** significant *significant at 10%; ** significant at 5%; at 1% Source: Wodon. Creuovand Tsimoo 12006a) - 87 - FigureA4.1: Numberof Civil Servant per Category 1,800 1,600 1,400 1,200 1,000 800 600 400 200 0 FigureA4.2: Average Salary (CFAF per year) per Category of Civil Servants 2,000,000.q 1,800,000.0-' 1997-Mean 1,600,000.0-' D2005-Mean 1,400.000.0-` A B C D E F G H I J K L M N O P Q R S T U V W X Y Z A I I - 88 - Annex 5: CompensationPackageAnalysis TableA5.1: CompensationPackageSummary Statistics Householdsurvey Public Householdsurvey Public Public 1997 sector 2002 sector sector database database database 1996-97 2005 2005 Individual characteristics All Private All Private All obs. Bottom workers sector workers sector obs. Male (yes = 1) 0.73 0.71 0.61 0.56 0.49 0.76 0.73 Schooling(years) 6.57 6.38 5.89 4.20 3.17 8.94 3.98 Age (years) 31.95 32.27 38.94 35.22 33.65 44.67 47.98 Seniorityinjob (years) n.a. n.a. 11.67 27.41 26.03 28.73 36.98 Married(yes = 1) 0.92 0.91 0.76 0.69 0.67 0.64 0.62 Numberofdependents 5.78 5.65 3.59 na na na na Bissau(yes = 1) 0.71 0.68 0.65 0.54 0.48 0.59 0.61 Public sector (yes = 1) 0.22 0.00 1.oo 0.22 0.00 1.oo 1.oo Formalprivatesector 0.11 0.12 0.00 0.00 (yes = 1) 0.18 0.21 0.00 Earnings(thousandsof 372.0 354.5 169.9 414.8 CFAF per year) 418.8 394.5 596.2 No. of observations 488 344 3048 1620 1281 10261 3493 Source: Staffestimates basedof 1997and2005 civil census data. n.a.Not available Nofe: Earnings figures are measured as of 1997. In the household survey data, they include earnings in main and secondary occupation. In the public sector data, they include bonuses and allowances. The public sector data base includes employees in categories"0" to "Z" only. - 89 - -3E r4 3E 0 b 0 \o 0 ln 0 d 0 m 0 rn 2 0 2 v0 0 3 0 P 0 T 0 Y 0 t \ . I m o\ I -L 0 0 0 0 0 0 0 0 0 ~ 0 $ Q ) W b ( D V ) d r n N (%) seaAoldw3 palelnunmy 0 *- 0 3 0 0 -? 3 0 0 0, H 0 0 00 0 0 \o 0 * 0 0 0 hl 0 0 0 0 0 0 0 0 0 0 0 0 0 0 \ 0 0 ~ \ 0 v l b c r r h l ~ 4 U - (0 R - $0 - \ - 4 `0 0 0 0 0 0 0 0 0 0 0 0 9 9 9 9 9 9 9 9 9 9 9 8 8 Z i ? 8 8 5 $ 3 E i Z 0 v I o\ v, I 4 `0 0 0 0 0 0 0 0 0 0 0 0 9 9 9 9 9 9 9 9 9 9 9 8 8 Z i ? 8 8 3 $ 3 E i Z 0 7 Table A5.4: Poverty Measures in the Civil Service by Grade, 2005 Normalized consumption PO PI P2 Category A 2.63 1 0.200 0.077 0.030 B 3.682 0.019 0.008 0.003 C 2.124 0.037 0.013 0.005 D 2.534 0.044 0.017 0.007 E 2.151 0.037 0.014 0.006 F 1.959 0.019 0.005 0.002 G 2.410 0.082 0.043 0.023 H 2.106 0.027 0.009 0.004 I 1.355 0.022 0.009 0.004 J 1.164 0.028 0.009 0.003 K 1.057 0.773 0.134 0.026 L 1.276 0.489 0.137 0.040 M 0.959 0.723 0.236 0.078 N 0.745 0.846 0.319 0.122 ND 1.599 0.510 0.221 0.106 0 1.083 0.308 0.091 0.032 P 1.037 0.780 0.377 0.185 Q 0.980 0.628 0.250 0.107 R 0.557 0.915 0.458 0.232 S 1.410 0.498 0.243 0.124 T 0.766 0.779 0.411 0.222 U 0.888 0.709 0.369 0.197 V 0.600 0.880 0.474 0.258 W 0.552 0.954 0.521 0.286 X 0.870 0.621 0.228 0.097 Y 0.806 0.844 0.434 0.227 2 0.503 0.961 0.520 0.284 Total 1.156 0.512 0.207 0.093 m:Staff estimates based on the 2005 civil census. The categories are defined as follows: Level A: General Inspector (Inspecteur General); Level B : General Director (Directeur General); Level C: First Rank Director (Directeur de Service Premihe classe); Level D: Second Rank Director (Directeur de Service Deuxieme classe); Level E:Third RankDirector (Directeur de Service Troisibme classe); LevelF: Division Chief(Emploi de Chef de Rbpartition); Levels G to H: Technical position, superior level (emploi de technicien sup6rieur); Level I:Technical position, mid-level (emploi de technicien moyen); Levels J to N: Technical position, basic level (emploi de technicien de base); Levels 0 to Z: Manual worker such as guards, drivers, cleaning andmaintenance personnel, low skill workers, etc. - 96 -