Report No. 27135-UG The Republic of Uganda Public Expenditure Review 2003 Supporting Budget Reforms at the Central and Local Government Levels September 2003 Poverty Reduction and Economic Management 2 Country Department of Uganda Africa Region Document of the World Bank GOVERNMENTFISCALYEAR Fiscal 2002/03: July 1,2002,to June 30,2003 CURRENCY EQUIVALENTS Currencyunit: Uganda shilling (USh) Market mid-rate: US$l.OO = 1990.75 ABBREVIATIONS AND ACRONYMS AIDS Acquired Immune Deficiency Syndrome MDP MunicipalDevelopmentPlan BFP Budget framework paper M&E Monitoringand Evaluation BOU Bank of Uganda MFPED Ministry of Finance, Planning and Economic BTVET Business, technical, and vocationaleducation Development and training MISS ManagementInformation Systems CAPEP Capacity and PerformanceEnhancement MOES Ministry of Education and Sports Project MOH Ministry of Health CAO Chief administrativeofficer MOLG Ministryof Local Government ccs Commitmentcontrol system MOPS Ministry of Public Service CFAA Country Financial AccountabilityAssessment MTCS Medium-termcompetitiveness strategy CFO Chief Financial Officer MTEF Medium-termexpenditure framework CPI Consumerprice index MTTI Ministry of Tourism, Trade and Industry DANIDA Danish InternationalDevelopmentAgency MWLE Ministry of Water, Lands and Environment DDP District DevelopmentPlan M2 Broad Money DFID Department for InternationalDevelopment NAADS National AgricultureAdvisory Services DWD Directorate of WaterDepartment NARO National Research Organisation EFA Education for All NRM National ResistanceMovement EFMP Economic and FinancialManagementProject NGO Nongovernmentalorganization ESR Education SectorReview NIMES National IntegratedMonitoring and Evaluation EU European Union System FD Finance Department NPV Net present value FDS Fiscal Decentralization Strategy OAG Ofice of the Auditor General GDP Grossdomesticproduct OOB Output-orientedbudgeting HIPC Highly Indebted Poor CountriesInitiative OPD Out Patients Department HIV Human immunodeficiencyvirus PA Public administration HLGs Higher Local Governments PAC Public Accounts Committee HOD Head of department PAF Poverty Action Fund HSSP Health Sector StrategicPlan PAYE Pay As You Earn IA Internal Audit PBR Pupil-book ratio IDA InternationalDevelopmentAssociation PCR Pupil-classroomratio IFEM Inter bank foreign exchange market PEAP Poverty Eradication Action Plan IFMS IntegratedFinancial Management System PER Public Expenditure Review IGG Inspectorate General of Government PFAB Public Finance and Accountability Bill IMF InternationalMonetary Fund PEMCOM Public Exuenditure ManagementCommittee IPF IndicativePlanning Figures PHC Primary Health Care - IPPS IntegratedPayroll and Personnel System PMA Plan for Modernizationof Agriculture JLOS Justice, law, and order sector PMES Poverty Monitoring and Evaluation Strategy LG Local government PNFP Private, not-for-profit LGBFP Local governmentbudget framework paper PPET Post primary education and training LGDP Local GovernmentDevelopment Program PRSC Poverty Reduction Support Credit LGFAR LG Financial and Accounting Regulations PSR Poverty Status Report LGFC Local GovernmentFinance Commission PTR Pupil-teacherratio LGPAC Local GovernmentPublic Accounts REER Real effective exchange rate Committee ROM Results-orientedmanagement LLG Lower local government SFG School facility grant LM Line Ministries SIDA SwedishInternationalDevelopment Authority LPO Local Purchase Orders SWAP Sectonvide approach LTEF Long-termexpenditure framework SWGs Sector working groups MAAIF Ministryof Agriculture, Animal Industries and TFP Total factor productivity Fisheries TSU Technical SupportUnit MDGs Millennium Development Goals UBOS Uganda Bureau of Statistics ucs UgandaComputer Services UPPAP UgandaParticipatingPoverty Assessment UDHS UgandaDemographicandHealthSurvey Project UIA UgandaInvestmentAuthority URA UgandaRevenueAuthority UNCDF UnitedNation CapitalDevelopmentFund USh Ugandashilling(s) UPE UniversalPrimaryEducation UTB UgandaTourismBoard VAT Value addedtax Vice President: Callisto Madavo Country Director: Judy O'Connor Sector Manager: Frederic Kilby Task Team Leader: R. Sudharshan Canagarajah PREFACE This Public ExpenditureReview (PER) Report for Uganda is the outcome of the budget- related activities carried out during the budget process for fiscal year 2003/04. Specifically, this report captures the details o f the budget reforms designed and implemented at the central and local government levels, analytical work carried out on specific budget-related themes, and outcome o f the consultations between the various stakeholders during the budget process. It was task managed and written by Sudharshan Canagarajah with written inputs from Rita Babihuga and Alema Siddiky (AFTP2) and Lawrence Bategeka (consultant). This report reflects the outputs and outcome o f PER- related work that was facilitated by the PER working group, which constitutes representatives from the Ministry o f Finance, Planning and Economic Development (MFPED) and major budget support donor representatives. Iam especially thankful to the MFPED for organizing the multistakeholder PER workshop inKampala May 19-20, 2003, during which many o f the issues raised inthe report were discussed indetail, which helpedimmenselyinwriting this report. In addition, Ihave benefited from comments and suggestions from Florence Kuteesa (director, budget), Ishmael Magona (commissioner, budget policy), Martin Brownbridge (macroeconomic adviser), Peter Fairman (senior economic adviser), Fiona Davies (macroeconomist), Giuilio Federico (economist, budget policy), and John Rudman (budget policy adviser), all from MFPED. Ihave also benefited from useful comments from PER working group members from the donor community, specifically Tim Williams and Jonathan Beynon o f the U.K. Department for International Development (DFID). Ihave also significantly benefitedfrom inputs and comments from the Uganda country team members o f the World Bank, especially Francis Ato Brown, Peter Okwero, Harriet Nannyonj0, Robert Townsend, Satu Kahkonen, Denyse Morin, Kithinji Kiragu, Yitzhak Kamhi, John Oloya, Tony Thompson, Louise Fox, and Robert Blake. Ialso want to acknowledge with appreciation the detailed comments Ireceived from the peer reviewers-Anand Rajaram, Allister Moon, and Roland White-which improved the quality o f the report. The PER report decision meeting provided additional useful suggestions. It was attended by many Uganda country team members and from the government o f Uganda, the MFPED, and memberso f the PER working group. Iam also thankful to the DFID, which assisted in carrying out the special thematic study on local government budget and financial management issues as part of this PER report. Iam thankful to the overall guidance provided by Frederick Kilby (Sector Manager) and Judy O'Connor (Country Director). Finally, Iam grateful to Mellen Candage for her editorial support and Vikki Taaka, Sarah Babirye, Tanisha McGill, and Arlette Sourou for their assistance informatting this report. Table of Contents PREFACE ......................................................................................................................... I EXECUTIVE SUMMARY ....................................................................... , ....................... I A. INTRODUCTION .......................................................................................................... B. MACROECONOMIC ANDFISCALCHALLENGES........................................................ ..i 11 FiscalChallenges....................................... GrowthPerformance and Challenges ............,.,,,,,,,.,..,..,,,,,.,...,.... ........,...........,,......,,, *. ,,, ,...........................................................11 , 11 *. The Way Forward...................................................................................................... ... 111 C. ................................................................................................... 9 . . Issues Relatingto the MTEF ..................................................................................... BUDGET PROCESS 111 ... 111 The Way Forward....................................................................................................... iv D. BUDGET MainIssues............................................................. EXECUTION............................................................................................... ..................................................... v E. BUDGETEFFICIENCY ............................................................................................. vi1 The Way Forward....................................................................................................... v vi .. Issues Addressed.....................,, . The Way Forward .................................. .. .......,..,,,., ..,,,,.,,..,,,,,,.,,,,,,,,.,,.....,,,...,....,..............vi1.. .....,,.......................................................... ......ix F. LGBUDGETANDFINANCIALMANAGEMENT ............................................... ix Issues Addressed.....,,,.,,, ....,.,..... ,, .,,,,,,,...,,,..,,...ISSUES..................ix ,,,,,, , ,,,,,,, , , . . .. .. .................. G. CONCLUSION ........................................................................................................... xi .TheWayForward....................................................................................................... MainFindings......,.,......,......,......................................................................................ix xi 1. AN OVERVIEW OF UGANDA'S MACROECONOMIC AND FISCAL CHALLENGES ................................................................................................................. 1 A. ANOVERVIEWOFECONOMICPERFORMANCE .......................................................... 1 ExternalSector ............................................................... GrowthPerformance.................................................................................................... 1 ..................... ....................3 ExportPerformance....................... ......................................................... .. ..,..,,,.,,,....,,,,........,,............,............................4 ............... ... 4 FactorProductivityandUganda's Growth.................................................................. ExportCompetitiveness......................,, ... . . ,, MacroeconomicChallengesandOutlook............... .......... 5 B. OVERVIEWOF FISCALTRENDS ................................................................................. . ................................... .......6 7 DomesticRevenueTrends........................ GovernmentExpenditureTrends .....,.., ..... ........,...................................... ..................7 C. PUBLIC ADMINISTRATION ......................................................................................... 9 ,,, ,,,,,,,,, ,, ,... ...................................................8 D. AID, EXTERNAL DEBT,AND FISCAL SUSTAMABILITY ............................................ 10 E. UGANDA'S FISCALDEFICITCHALLENGES OVER THE MEDIUM TE....................... 12 F. MEDIUM TERMCHALLENGES ANDTHEWAYFORWA RD........................................16 2. UGANDA'S BUDGET PROCESSAND MTEF,2003/04 TO 2005/06 ...............19 A. INTRODUCTION ....................................................................................................... 19 B. PHASE 1OFTHEBUDGET: BUDGET CEILINGSANDBUDGET PREPARATION ....... ....20 I PARLIAMENT .................................................................................................................. c. PHASE2 OF THE BUDGET: PARTICIPATION AND BUDGET PRESENTATION TO Medium-Term Resource Envelope............................................................................ 25 Medium Term Expenditure Framework (MTEF)...................................................... 26 28 35 D CHALLENGESINTHE BUDGET .Assessment ofthe Budget Process ............................................................................ PROCESS .................................................................. 36 E. THEWAYFORWARD .............................................................................................. 37 3. BUDGET EXECUTION AND MONITORING, FISCAL YEAR 2002/03 ........39 A. BACKGROUND B . ........................................................................................................ 39 REVENUE PERFORMANCE ....................................................................................... 40 EXPENDITURE PERFORMANCE ................................................................................ 40 D. C. BUDGET CUTS ................................................................ 42 E. OVERRUNSAND BUDGET UNDERPROGRAMMING AND UNDERFWDGETING 42 COUNTERPART FUNDING.................................................................. ..................................................... F. INADEQUATE G PROBLEMSOF IRREGULARFLOWOFRESOURCES................................................... H PAYMENTOFARREARS .. 43 44 AND THE ccs.................................................................... 44 I. MANAGEMENT BUDGET MONITORING REPORTING................................................................. 47 OF BUDGET CUTS ........................................................................... 45 AND K. J. CHALLENGES FOR THE FUTURE .............................................................................. 48 4.L EFFICIENCY . CONCLUSION .......................................................................................................... 49 OF PUBLIC EXPENDITURES ................................................... 51 A. WATERAND SANITATION Background................................................................................................................ 52 SECTOR.......................................................................... 52 Allocative Efficiency ................................................................................................. 55 B.OperationalSECTOR..................................................................................................... Efficiency............................................................................................... 57 Background................................................................................................................ HEALTH 58 Allocative Efficiency ................................................................................................. 58 60 63 C.Background., SECTOR............................................................................................... Operational Efficiency............................................................................................... EDUCATION .............................................................................................................. 66 66 Allocative Efficiency ................................................................................................. 68 69 70 E..Operational Efficiency............................................................................................... D JUSTICE,LAW, CONCLUSION .......................................................................................................... ANDORDERSECTOR(JLOS)........................................................... 72 5. LOCAL GOVERNMENT BUDGET AND FINANCIAL MANAGEMENT ISSUES ............................................................................................................................. 73 A. INTRODUCTION ....................................................................................................... 73 Emerging Challenges................................................................................................. Theoretical Framework for Decentralization of Public Service Delivery and LGand Legal Framework inUganda........................................................................ 73 74 B.The Process o f Fiscal Decentralization: A Phased Approach ................................... 74 75 C. PLANNING ............................................................................................................... BUDGETING ............................................................................................................ 76 D. COMPOSITION OF EXPENDITURE REVENUES AND ................................................... 77 E. INTERNAL CONTROLS............................................................................................. 77 F. PLANNING FORPROCUREMENTAND TENDERING .................................................... 78 G. .................................................................................... 78 Payroll Management.................................................................................................. Commitment and Expenditure Control...................................................................... CASHFLOWMANAGEMENT 79 Accounting and Record-Keeping .............................................................................. 80 Deviations from Budget............................................................................................. 80 81 H.DebtManagementMEASUREMENT ...................................................................................................... 82 PERFORMANCE. AND VERTICAL REPORTING WITHIN THE LG ADMINISTRATION Reporting to Politicians ............................................................................................. ........................................................................................................... 82 83 External Reporting to the Central Government and Donors Reporting to the Beneficiaries and Civil Society Organizations ............................... ..................................... 83 84 Internal Audit............................................................................................................. Monitoring and Analyzing Performance................................................................... 84 Local Government Public Accounts Committee (LGPAC) ...................................... 85 External Audit............................................................................................................ 85 86 Parliament LGPAC and Other Control Measures ..................................................... Final Accounts ........................................................................................................... 86 Transparency and Accountability inFinancial Management.................................... 87 87 External Factors......................................................................................................... Institutional, Management, and Political Factors...................................................... 88 89 6. CONCLUSIONS AND RECOMMENDATIONS ................................................ 91 A. INTRODUCTION ....................................................................................................... 91 B. MACROECONOMIC c. PERFORMANCE......................................................................... ............................. 92 91 D. BUDGET BUDGETPROCESS, MEDIUM-TERMBUDGETFRAMEWORK, MTEF EXECUTION 93 E 94 F.. .............................................................................................. BUDGETEFFICIENCY .............................................................................................. LGBUDGETAND FINANCIAL MANAGEMENT ............................................................ 96 ANNEXES ........................................................................................................................ 98 ANNEX1: BUDGET PERFORMANCEANDOUTTURN2002/03 .......................................... ANNEX2: MEDIUM TERMEXPENDITURE FRAMEWORK 2002/03 & 2005/06 ...............102 99 ANNEX3: STATEMENTB Y THEWORLD BANK,ONBEHALF DEVELOPMENT PARTNERS, ONBUDGET PERFORMANCE, 2002/03 ....................................................... OF 105 ANNEX4: STATEMENTB Y THEWORLD BANK,ONBEHALF THEDEVELOPMENT PARTNERS,ON THEMTEF 2003/04 To 2005/06 .......................................................... OF 108 ANNEX5: UGANDA IMPLEMENTATION STATUS OF ACTIONSTO STRENGTHEN TRACKING OFPOVERTY-REDUCINGPUBLIC SPENDING ................................................................. 112 BIBLIOGRAPHY .......................................................................................................... 114 FIGURESAND TABLES FIGURES Figure 1.1 UgandaGrowth inGDP-Agriculture. ..................................... Figure 1.2 Noncoffee Export Earning............................................................................................. Industry. and Services 2 4 5 Figure 1.4 UgandaTrends inProject and Program Support (% of GDP) ..................................... Figure 1.3 UgandaRealEffective Exchange Rate and Terms of Trade Indexes....................... 12 Figure 1.5 Government Expenditure, Domestic Revenues, and Aid (% of GDP)......................... Figure 1.6FiscalDeficit, ExcludingGrants (% of GDP).............................................................. 13 Figure2.1 Sectoral Increasesfor 2003/04 with LGDP Allocations .............................................. 14 35 44 Figure4.1 Trends inInfant Mortality, by Sanitation Facility and Water Source.......................... Figure3.1 BudgetedandActual CashFlows for July 2002/June 2003......................................... 53 Figure4.2 New Outpatient Attendance inUgandan Government and Private, Not-for-Profit Health Units.......................................................................................................................... 64 TABLES Table 1.1 Key Macroeconomic Indicators....................................................................................... 3 Table 1.2 Public Administration Budget Trends, 1997/98 to 2002/03 ............................................ 9 11 Table 2.1 Budget Processfor the Year 2003/04 ............................................................................ Table 1.3 Official Development Assistance as a Share of GDP.................................................... Table 2.3 Resource Envelope 2003/04 to 2005/06 (USh Billion) ................................................. Table 2.2 MOPS Budget (MTEF) Proposal for Wage Bill, 2003/04 to 2005/06 (UShBillion)....21 24 Table 2.4 URA RevenuePerformance, July 2002/March 2003 (UShBillion) ............................. 27 Table 2.5 Sector Ceilings, IncludingDonor Projects (UShBillion) ............................................. 28 29 Table 2.7 Distribution of Budget Increasesfor 2003/04, by Sectors............................................. Table 2.6 Sectoral Shares of Expenditure over the MediumTerm..................................... 30 34 Table 4.1 Trends in Selected Mortality Indicators ........................................................................ Table 3.1 Average Budget Deviations........................................................................................... 47 53 Table 4.2 Water Coverage Patterns(2002).................................................................................... Table 4.3 Rural Clean Water Supply by Source, 1999/2001 (Number of Sources of Supply)......54 55 Table 4.6 PAF (Health Sector) Expenditures, Fiscal 2001/02 to 2005/06 (UShbillion) ..............59 Table 4.5 Summary o f HealthIndicators (2002) .......................................................... Table 4.4 Water and Sanitation Sector ResourceAllocation (2002) ................................... 56 62 Table 4.7 Variation in Per Capita Allocation of the PHC Conditional Grant to Districts, Table 4.8 Education Sector Targets for Quality Inputs................................................................. 2002/03 ................................................................................................................................. 62 Table 4.9 Education Sector ProgrammedExpenditure inthe MediumTerm (USh Billion).........67 Table 4.10 Actual and Projected Growth inPrimary 7 Leavers, 1997/2003................................. 68 70 EXECUTIVESUMMARY A. INTRODUCTION This 2003 Public Expenditure Review (PER) report addresses budgetprocess challenges at both the central and local government (LG) levels, highlightingthe progress made and identifying challenges for the future. The first chapter reviews country economic performance over the previous year and its implications for fiscal performance. In addition to tracking the various macroeconomic indicator trends, the chapter highlights areas o f challenge inthe economy and includes an expanded section on the challenges o f managing the recent fiscal deficit trends. Chapter 2 discusses the budget process, including issues concerning the preparation o f the budget, the roles o f the sector working groups (SWGs), and progress made inthe preparationo f sector budget framework papers (BFPs). The chapter also comments on the quality and depth o f participation o f various stakeholders inthe budgetprocess. Chapter 3 addresses budget execution challenges inUganda during the last budget cycle. Problems encountered in budget execution-non programmed allocation o f resources during the course o f the year, budget overruns and cuts, irregular flow of resources, arrears, under-and overprogramming elements o f the budget-are discussed. Inaddition, we comment on the progress made inthe area of budgetreporting and identify remaining challenges in improving budget monitoring and reporting. Chapter 4 deals with budget efficiency issues. The medium-term expenditure framework (MTEF), which guides government spending and especially the sectoral allocation o f budgetary resources, involves an ever-increasing need to better prioritize targets and ensure efficiency to achieve the priority targets. This chapter analyzes efficiency challenges as they relate to four sectors: namely health, education, water and sanitation, and the justice, law, and order sectors. Chapter 5 deals with the LG issues in improving budget processes: budget allocation, execution, and financial management. This chapter i s based on a study o f these issues in three districts and one municipality that was carried out in early 2003 as part o f the 2002/03 PER. The results show the heterogeneity of the LG budget and financial management performance and challenges and the need to benchmark and assist LGs accordingly to make progress in these areas inthe context o f supporting decentralization in Uganda. The results also underscore the extensive scope for learning from others within Uganda. Chapter 6 concludes with a summary o f the main findings of this year's PER and the challenges for the future. B. MACROECONOMIC CHALLENGES AND FISCAL GrowthPerformanceand Challenges The objective of maintaining macroeconomic stability was largely achieved in fiscal 2002/03, although there were periods o f rising inflation as a result o f the impact o f adverse weather conditions on agricultural production. Furthermore, uncertainty about donor aid inflows and higher international prices o f oil, arising from the war in Iraq, exerted depreciation pressures on the exchange rate. Thus, although inflation remained under control, inflationary pressures during 2002/03 were high. The upward pressureon food prices was caused by the impact o f adverse weather conditions on agricultural production. The depreciation of the Uganda shilling (USh) also contributed to the rise in inflation. At 5.9 percent, annual headline inflation was higher than the annual budgetary projection o f 1 percent. Real gross domestic product (GDP) grew by 4.8 percent, compared with 6 percent the previous year. The growth rate was also below the targeted 7 percent set for poverty eradication by 2017. The decline inthe rate o f economic growth was attributed to global economic growth slowdown, low prices o f coffee on the international market, and unfavorable weather conditions. Another domestic factor that explains the low growth i s the further delay inconstructionofthe Bujagali hydroelectricity project. The impact o f poor weather conditions was particularly felt inthe agriculture sector and compounded by poor international commodity prices; this sector declined interms o f its overall contribution to GDP. The agricultural sector grew by 3.2 percent, representing the lowest growth rate compared with other sectors. The service and industry sectors continued to grow strongly, at 6.8 and 7.5 percent o f real GDP, respectively. Growth in the industry sector was fueled particularly by the construction subsector, whose performance was strong partly because o f buildingnew schools and increased growth o f the manufacturing subsector-from 5.8 percent in2001/02 to 6.6 percent in2002/03. Fiscal Challenges Revenue performance inFY2002/03 was 99 percent o f approved budget estimates. As a share o f GDP, domestic revenue largely remained stagnant between 2001/02 and 2002/03: collections in 2002/03 amounted to 12.1 percent o f GDP compared with 12.3 percent o f GDP in 2001/02. Public expenditure as a share o f GDP, however, declined slightly in2002/03 to 23.3 percent compared with 25.3 percent in2001/02. This was in line with the government's policy of decreasing the fiscal deficit. The fiscal deficit, excluding grants, was 11.3 percent o f GDP in 2002/03 compared with 13 percent in 2001/02. Widening fiscal deficit and external debt sustainability emerged as crucial macroeconomic concerns during fiscal 2002/03. Because the fiscal deficit has largely beenfundedby increased donor inflows, inaddition to worsening Uganda's external debt position, the increased liquidity has put upward pressure on interest rates because of increased issuance o f treasury bills to absorb excess liquidity. Yet there are concerns that - .. 11 - sale o f foreign exchange to absorb excess liquidity could lead to "Dutch disease" effects, to the detriment o f exports performance. To maintain macroeconomic stability, the government intends to reduce the fiscal deficit over the medium term through a gradual fiscal consolidation, placing greater emphasis on higher fiscal revenue and controlling new nonconcessional borrowing. The ongoing Poverty Eradication Action Plan (PEAP) revision will aim at developing a medium-termstrategy for fiscal consolidation and debt sustainability. The Way Forward In the medium term, Uganda must focus on maintaining macroeconomic stability, improving growth performance, and sustaining this growth to fulfill the long-run objective o fpoverty eradication. There will be increased focus on bringingthe size o f the fiscal deficit under control, with particular concentration on strategies to increase domestic revenues and increase the effectiveness o f current public expenditures. The issues o f aid reliance, absorption, and effectiveness will also be addressed during the current PEAP revision process. C. BUDGET PROCESS The chapter on the budget process provides an update on the BFPs preparation process for 2003/04. It comments on the progress made inphases 1 and 2 o f the budget process. It clearly shows that phase 1has become more organized, and more time was given than last year for this process. The phase 2 external evaluation process also has become more systematic. The discussion shows how the parliament has become very actively involved inthe budgetprocess. Inaddition, there is a developing synergy betweenlocal processes and donor interaction on the budgetneeds to be harnessed inthe coming year. The SWGs have become more organized and entrenched inthe system, the preparation of BFPs is becoming an organized activity. In fact, the Ministry of Finance, Planning and Economic Development (MFPED) i s using the SWGs to revise the PEAP. Marginal improvements were observed inthe frequency with which SWGs met to deliberate issues of their sectors and the preparationo f BFPs. Therewere marginal improvements insome sectors interms o fthe quality and content of BFPs in areas such as intersectoral linkages, output-oriented budgeting (OOB) and results-oriented management (ROM), cross-sectoral issues such as gender, and prioritization o f activities. SWGs need to continue to strengthenthese areas of the BFPs. This year's PER meeting was attended by a wide range of stakeholders and addressed many issues concerning public expenditures, with a particular focus on the area o f budgetary efficiency in the sectors of education, health, water and sanitation, andjustice, law, and order sectors. Issues Relating to the MTEF At the government's request, the move from projects to budget support as a lending modality received increasing support this year from donors such as the Department for -111 9 . .- International Development (DFID), the European Union (EU), the World Bank, and the Ireland Aid . The process o f incorporating projects within sector ceilings begins in fiscal 2003/04, although the quality o f information is still imperfect. The integration o f the wage bill within the MTEF also remains weak. The 2003/04 budget maintains its poverty eradication focus on reducing the size o f the fiscal deficit and promoting strategic exports. Thus, the sectoral distribution o f budgetary resource increases should support implementation o f the PEAP and Millennium Development Goals (MDGs) targets through the Poverty Action Fund(PAF) modality. There are still challenges in terms of making SWGs effective organs through which the BFP process can be effectively carried out. Although SWGs have come a long way and are much more entrenched as a mechanism compared with any other Sub-Saharan African country, there i s still some room for further progress interms o f bothprocess and substance o f BFPs. The Way Forward Despitethe many areas o f the budgetprocess inwhich progress was made, it is clear that there are still areas where challenges remain. First, technical capacity constraints at the sector and LG levels limit the scope for making progress in BFPs-especially making them output oriented, linking them to ROM, prioritizing and costing activities, and so forth. SWGs should be given more time and assistance inthe preparation o f the BFPs if they are to become important policy planning instruments. In addition, this year the MFPED and the Ministry o f Gender conducted training workshops for LG and central government officials working on BFPs to address gender issues comprehensively intheir BFPs. Second, the determination of sector ceilings needs to become more participatory and transparent. This i s especially the case with regard to social sectors, where it i s clear that additional resources are needed. The national BFP and budget reflect big divergences, thus rendering the BFP an ineffective planning tool. This has led to substantial grievance among sectors and other stakeholders, including civil society and donors. Third, the proposed improvements inthe MTEF to capture all donor projects and wage bill issues are progressing slowly. This needs to be addressed in a systematic way. Fourth, budget monitoring processes have not improved greatly over the last few years. Budget performance reporting is hardly based on outputs because they are not clearly defined and monitored by sectors. Fifth, although the PER workshop has become an important forum through which broad and consultative budget issues are discussed, it still needs to be better synchronized with parliamentary budget processes to ensure greater impact. Sixth, civil society participation inthe budget process i s still weak as a result o f capacity constraints and needs to be looked into, because they have useful inputs for the budgetprocess through SWGs and the PER workshop. All these areas have been discussed widely, and the government is trying to streamline the technical assistance and other budget reforms to have the greatest impact. In this regard, the recent establishment o f the Public Expenditure Management Committee (PEMCOM) as a way o f coordinating budget reform i s very welcome. The challenge i s - iv - to ensure that these initiatives are systematically monitored and supportedto realize the intendedimpacts. D. BUDGETEXECUTION The issue of budget executioncovers revenue collection, expenditures released, and their corresponding outturns. Apart from a brief discussion of the revenue challenges, the chapter focuses mainly onthe expenditure managementissues. The chapter briefly looks at revenue areas that performedwell and those that did not and the possible reasons for the observed performance. In addition, the chapter looks at donor budget and project support releaseperformanceand how differences betweenthe government and donors on changes inexpenditureprioritieshadanimpact onthe time and size ofreleases. Expenditure performance is at the heart of budget management. Budget execution challenges related to timely flow of budgetary resources to cost centers, budget overruns and their management, effects of budget cuts on affected sectors, payment of arrears, inadequate counterpart funding of donor-supported development projects, the commitment control system, and other relatedbudget executionchallenges for the future are discussed insome detailinthis chapter. Inaddition,the chapter reviewsthe financial management aspects of budgets, such as capacity gaps, systems and processes of financial management,accountingand evaluation,and the link betweenbudget discipline and political involvement. Naturally, the progress on these issues in the medium term will have an important bearing on the extent to which PEAP and MDGs targets can be achieved. MainIssues As in the previous year, the fiscal year 2002/03 witnessed budget pressures emanating from security and public administration sectors, requiring additional supplementary resources to meet their new demands. In addition, the increase in domestic interest payments morethan doubledcomparedwith allocation, necessitating further cuts inother parts ofthe budgetto meetthese unforeseenexpenditure needs. Naturally,these were not well received by other budget stakeholders. The MFPED and line ministries did their best to minimize the impact of these disruptions, and there are some lessons for the future. Thus, as in the previous year, budget overruns in 2002/03 necessitatedbudget cuts, which caused disruption in programs of other government departments and line ministries. Defense frontloaded most of its expenditures inthe first half of the financial year, Expenditureon domestic interestpayments and on public expenditurewere higher than programmed. Although unforeseen expenditures could be managed, underprogramming-relatedproblems could not be adequately addressed in the course of the financial year. Inadequate counterpart funds resulted in underperformance of many donor-funded development projects. Counterpart funding problems were also related to the budget overrun problems in 2002/03. This problem of budget overruns seems to be perennial and shouldbe addressed. - v - Irregular flow of funds from the treasury characterized budget implementation during 2002/03. As a result of the cash budget and the frontloading of defense expenditure, some sectors did not get their portion on time. Insome cases, implementationof some activities was delayed. This delayed release of funds meant that activities were either partly implementedor not implantedat all. Thus, in some cases (as inroads andworks, agriculture),budgetaryobjectives couldnot be achieved. For example, inthe agriculture sector, activities for the particular season or crop could not be carried out because of delayedrelease of funds. The commitment control system (CCS), which was implemented systematically for central government nonwage and development expenditure, has not been fully extended to all categories o f expenditure at the central and LG levels. Because of incomplete implementationof the CCS, arrears have continued to build up. The govemment has committed itself to making progress and expects that the plans for the Integrated Financial Management System (IFMS) implementation in the next year will assist in arrestingthe problemof arrears buildup. There are a few lessons to learn from the budget cuts of 2002/03, particularly how the challenges that arose were managed. First, underprogramming should be avoided. Second, better prioritization of intrasectoral expenditure would minimize the adverse impact of budget cuts. The education sector tried to do this and minimized the adverse effects of budget cuts on implementationof its programs. Other sectors should do the same. Third, irregular budget flows should be avoided as much as possible. Fourth, counterpart funds shouldbe agreed with all parties concernedduring budget preparation to avoid underallocationproblems. Fifth, the CCS and IFMS implementationshould be vigorously pursued to avoid arrears buildup, which still plagues the budget system. Finally, the lessons learnedfrom managingbudget cuts should be pursuedby all sectors because some o f the budget cuts do not seem to be exceptions and need to be managed every year. The Way Forward Uganda should continue its progress in implementingthe budget as agreed. Given the substantialprogressmade inminimizingoverallbudget deviations, any effect on planned activities should be minimized. Public administration sector overruns should be decisively addressedsoon if the sectors are to have confidence inthe budget system. In this regard, the MFPED and the Ministry of Public Service have already agreed to undertake a study to address the problem in the coming year. In addition, budget executionissues at the LG levelinrecent years have becomemore important. This is an importantchallenge inthe context of improvingservice delivery. Budget execution and budget reporting are areas in which there has been limited progress, and further challengesthat needaddressingshouldbepartofthe budgetreforms inthe mediumterm. - vi - E. BUDGETEFFICIENCY Budget efficiency is especially important, not only because o f the need to ensure that public expenditures derive their correct value, but also withinthe context o f reduction o f the fiscal deficit. As the government aims to consolidate the size o f the fiscal deficit, emphasizing budget efficiency can enable it to achieve its development goals without subsequent disproportionate increases to its resource envelope. This report underscores the urgent need to spread the culture of ensuring public expenditure efficiency (both allocative and operational efficiency) across every sector o f the economy without exception, to have an impact on the overall resource envelope. The report focuses on the status o f efforts to increase efficiency within four sectors- water and sanitation, health, education, and justice, law, and order-highlighting the overall need for greater institutionalization o f the process, better use o f accurate methodologies for measuring value for money, and a greater degree o f translation o f the results o f analyses into policies and actions that ultimately increase value for money within the respective sectors. The report notes the differing levels o f attention to value for money depending on the sector, which in turn has implications for how the issue i s carried forward ineach sector. Issues Addressed The education sector provides a good model for other sectors inexpendituretracking and incorporating emerging issues from the studies into sector action plans in a biannual Education Sector Review (ESR) as well as reviewing progress on past reforms. This sector i s currently focused on improving the allocative efficiency o f expenditures by improving the sharing o f resources across subsectors without hurting existing programs, but also more generally, stressingincreased operational efficiency across all subsectors. More than five years into the Universal Primary Education (UPE) program, the sector i s increasing its focus on the postprimary segment to effectively absorb the UPE bulge. In addition to emphasizing efficiency issues in the postprimary segment, the sector has commissioned a draft strategy for postprimary education and training policy and costed framework. Its implementation will involve efficiency measures such as double shifting, especially where student concentration i s high and an improvement in the pupil-teacher ratio. A tracking study on teacher recruitment and payroll administration carried out in 2002 revealed significant payroll and financial management and made recommendations to the ESR. Further studies to analyze the school facilities grant program are planned for fiscal year 2003/04. Similarly, the health sector has made considerable progress in the last several years in tracking expenditures, although much work still needs to be done to track expenditures and improve efficiency at the district level. At the same time, much of the work continues to be carried out on an ad hoc basis, and there i s a needto better systematize it. There i s also a greater need to incorporate within the sector's plans issues raised by various studies and follow up on their effective implementation. - vii - In view of the lack of improvement in infant, child, and maternal mortality indicators, which are lower than the rate needed to achieve the MDGs, there i s a need to improve coordination among the health, education, and water and sanitation sectors to address this problem adequately. The health sector still needs to reconcile its funding requirementsfor implementing its long-term plans with the amount provided in the budget. As the sector increases the amount o f resources spent at the district level, there i s an even greater need to improve the tracking o fthese resources. There i s a debate within the health sector about the appropriate role o f the private sector. Evidence indicates that expanding this role would relieve an overburdened and underfunded national health care system, which would improve efficiency and the provision o f health services. The water and sanitation sector completed its first value-for-money study in fiscal year 2001/02. Although questions arise about the quality o f the results, this marked an important beginningin addressing efficiency in the sector. This work should be carried out annually and improved so that greater value for money i s realized, particularly at the level o f service delivery. Although the water and sanitation sector's share o f the budget has increased dramatically over the last three years, a substantial part of these resources continues to be spent at the center instead o f the rural areas, where the majority o f the poor live without access to clean water and adequate sanitation. There i s an urgent needwithin the sector to increase the allocative efficiency o f resources. At the district level, there are needs for better poverty targeting o f resources and addressing the regional differences that exist in the sector's provision o f services. The value-for-money study highlighted issues of lack of capacity at the district level, nonadherence to procurement and financial management regulations, outright corruption, and weak capacity o f the private sector to which the contracts are awarded. The top- down approach to management in the sector also should be addressed to improve the impact o f the sector at the lowest levels. The justice, law, and order sector (JLOS), with its unique institutional structure, addressed value for money by focusing on financial management within its institutions. The study identifies many weaknesses in the system, concluding that a high level of fiduciary risk exists within the JLOS institutions. The challenge ahead for the sector i s how to implement the recommendations made by the study to reduce the level o f fiduciary risk and increase the efficiency o f expenditures. Tightening the loopholes in financial management is pertinent not only inthe JLOS but the other sectors as well, because improving this area will improve each sector's capacity to manage its resources. - viii - The Way Forward The sectors are at different stages o f addressing efficiency-education and health have a relatively longer history, and water and sanitation's and JLOS's processes are muchmore recent-yet there i s a need for every sector to increase the intensity o f the focus on efficiency. The process needs to become better institutionalized so that ensuring efficiency fits within the sector's overall objectives rather than being an isolated process, gains from which cannot be adequately capitalized upon across the sector. In addition, sectors that still do not have adequate measurement indicators need to develop them as a first step to pursuingefficiency objectives. Last, the value-for-money studies must not be an end in themselves-their results should be used to inform subsequent sectoral reform policies to correct the gaps inthe efficient use o f resources. F. LGBUDGET AND FINANCIALMANAGEMENT ISSUES Issues Addressed The fifth chapter of the PER 2003 describes Uganda's decentralization process with the aim o f pointing out areas infinancial management that need improvement. It looks at the internal and external factors affecting LG financial and budget management through a study that was carried out inthree districts and one municipality. The chapter assesses progress o f decentralization from a theoretical framework of public service delivery. It briefly looks at the legal framework for decentralization. Fiscal decentralization, which i s being implemented in a phased approach, i s a pivotal issue. Fiscal decentralization entails planning, budgeting, budget implementation (revenue and expenditure issues), financial management, internal controls, procurement and tendering, and manpower management (payroll issues). It also includes reporting (vertical, downward, and horizontal), cash flow issues, deviations from the budget (commitment control), debt management, and monitoring o f performance (internal and external audits and preparation o f statutory reports). The chapter further addresses the matter of financial accountability and efficiency inresource use. Main Findings The Local Government Act o f 1997 provides the legal framework for decentralizationin Uganda. The act empowers different layers o f LG to exercise political, administrative, and financial management for development in the respective areas. However, decentralization i s partial in the sense that some critical elements o f decentralization are missing (for example, resource allocation autonomy and strong local government institutions). Furthermore, most LGs suffer capacity constraints in several critical areas, and decentralization i s threatening to weaken rather than sharpen efficiency-promoting incentives by encouraging upward accountability o f LG rather than downward accountability to communities. - ix - The process of fiscal decentralization is phased, Fiscal decentralization has so far been possible for the recurrent budget and not for the development budget because o f technicalities involved, including reporting requirementso f some donors. All LGs are requiredto have a three-year development plan. The planning units of LGs develop the plans with inputs from the sectors and lower local governments. Involvement o f the citizens in planning i s limitedto the yearly budget conferences. Civil society plays hardly any role in the LG planning process. Weaknesses identified at the planning level include lack o f primary data; inconsistencies between the District Development Plans (DDPs), BFP, budgets, and work plans interms o f inputs, activities, outputs, and objectives; unrealistic plans due to lack o f cost-constrained planning; and finalizations o f the DDPs without due consideration o f the realism o f the plans, Regarding budgeting, the BFP is generally not well understood by district-level stakeholders, and its preparation i s seen largely as a technical exercise. Stakeholders therefore do not use the BFP as a tool for sector prioritization and decisionmaking. At the stage o fbudget implementation, there is little effort onrevenuecollection (withheavy reliance on transfers from central government making up about 90 percent of LG budgets). Political interference partly explains poor revenue collection inLGs. Apart from the central govemment regulations (LGFAR), tender board regulations, and so forth, LGs have hardly any written procedures or standards for internal control inuse o f public financial resources. Mechanisms for preventing conflict o f interest are not adequate. The procurement bodies requiredby law to be established are inplace. Most LGs have elaborate procurement plans and a list o f prequalified contractors, and they hold regular meetings. However, there are inadequate quarterly returns o f purchases to control successive purchases, weak transparency, failure to publish minutes o f tender board meetings, failure to update the list o f contractors, and poor follow-up on winningbids. Delayed access to the central government payroll (MOPS) by newly recruited staff is a big challenge. Accessing the payroll could take as long as a year. Deviations from the budget are common. Heads of departments have lost control o f their votes because vote books are located in the finance departments of LGs and not with heads o f departments. Most LGs have arrears equivalent to 2040 percent o f their own yearly revenues. However, LGs hardly have bank loans. Payment o f creditors i s hardly a priority to most LGs. LGs report mainly to central government. Reporting to beneficiaries (communities) is very weak. LGsprepare many reports (inthe range o f 10to 20 reports per year), many o f which are required for further disbursement of financial resources from central government to LGs. Often, analysis and follow-up on the reports i s o f limited value compared with the time and resources invested in their preparation. Although internal audit in LGs has improved, internal audit units face critical challenges, including lack o f facilitation, limited funds for field visits, lack o f computers and secretarial support, and - x - budget cuts. Internal audit units are not independent, and their powers are thus compromised. LGPACs review the internal audit reports and the reports from the OAG, but like the internal audit units, LGPACs are poorly facilitated. LGs hardly contribute to funding o f LGPACs, yet they are expected to do so. Worse still, LG councils hardly consider LGPACs' reports. Thus, there are many loopholes for financial abuse in LGs, with inadequate ways o fclosing them. The Way Forward Stakeholders in the development process should assist LGs in addressing the challenges that are inhibiting social service delivery at the LG level. Civil society and communities should be assisted to participate in the LGs' planning phase; measures should be developed to reduce or prevent conflict o f interest, especially inprocurement; and further measures should be implemented to improve financial management, transparency, and accountability at the LG level. Uganda's LGs face enormous budget and financial management challenges, which could be improved if systematic assistance i s provided. It i s clear that there are efficiency gains to be made inimproving budget processes and financial management. The results o f this year's PER LG study clearly illustrate the heterogeneity o f the LG performance inthese areas and the need to benchmark and assist LGs accordingly inthe context o f supporting the decentralization process in Uganda. It i s clear that LGs can learn from others within Uganda. G. CONCLUSION This year's PER has attempted to address the challenges in the budget process, budget execution, and budget efficiency at the central government level, and to a greater extent at LG levels. This PER has identified critical issues surrounding the budget process- progress as well as remaining challenges-and it has also analyzed the factors underlying all o f these. The credibility o f the budget process will depend on how progress made in upstream processes translates into downstream activities, thus this PER has focused on these issues at each stage o f the budget process. This has been discussed also in the context o f changing aid modalities and the implications for future effective use o f public resources. The underlying message i s that until these issues are clearly understood and addressed, budgets cannot make a useful impact on the growth and poverty reduction challenges that Uganda faces. - xi - 1. AN OVERVIEW OFUGANDA'S MACROECONOMIC AND FISCALCHALLENGES A. ANOVERVIEW OFECONOMIC PERFORMANCE 1.1 Uganda has continued to maintain macroeconomic stability despite some signs o f economic slowdown over the recent past. Uganda's growth rates, which were higher than 6 percent for more than a decade, have started to stagnate. The annual average headline inflation rates, which were much lower than 5 percent, have started to increase because o f increases in fuel prices and unfavorable weather conditions affecting food prices. Interest rates have risen, mainly because o f the government's efforts to mop up excess liquidity through treasury bill sales, leading to lower private sector credit growth. Despite these signs o f economic slowdown, Uganda has continued to pursue the implementation o f its Poverty Eradication Action Plan (PEAP) for achieving rapid and sustainable economic growth, which i s necessary for private sector growth and poverty reduction. In addition, the government i s committed to stimulating economic growth through increasing export diversity, encouraging private sector activity, and supporting rational and effective budgetpolicies. Growth Performance 1.2 Whereas Uganda's economic growth was largely robust over the last decade, it has been weaker than anticipated during fiscal year 2002/03. The 6.0 percent growth rate in 2001/02 moderated at 4.8 percent in 2002/03, which is lower than the programmed growth o f 7 percent for the year. This deterioration inthe growth rate i s mainly attributed to the impact o f prolonged dry weather conditions on the overall performance o f agricultural output, lower growth in food crops, higher fuel prices, and delays in the construction o fthe Bujagali hydroelectricity project. 1.3 The industry sector is the only sector o f the economy that grew steadily inthe last two years, from a growth rate o f 6.1 percent in 2000/01 to 7.5 percent in 2002/03 (exceeding average GDP growth), and the service sector has had real growth rates above 6 percent since 1998/99 (see figure 1.1). The driving force behind the growth o f this sector i s the construction subsector, whose growth more than quadrupled between 2000/01 and 2002/03. Growth rates o f both the services and agriculture sectors have slowed since 2001/02, with the greatest declines beingobserved inagriculture. 1.4 Nonetheless, at 41.2 percent o f total GDP, the service sector still contributedthe largest share o f GDP in 2002/03. This was followed by the agricultural sector, which contributed 39.4 percent o f total GDP. In 1991/92, the structure o f the economy was different, with the agriculture sector providing the largest share of total GDP, 51.1 - 1 - percent, and services providing 35.7 percent and industry, 13.2 percent, compared with 19.4 percent in 2002/03. There has been a slight shift away from agriculture to services and construction. This has partly to do with the decline inagricultural commodity prices, particularly coffee, duringthe last decade. Figure 1.1 Uganda Growth in GDP-Agriculture, Industry, and Services 25.0 I 20 0 I 5 0 --tGDPmarketprices +Agriculture 10 0 Industry 5.0 - ,*-Services 0.0 1993194 1995196 1997198 1999iOO 2001/02 -5 0 Source: Ministry o f Finance,PlanningandEconomicDevelopment(MFPED) 1.5 Inflation: From 2001/02 to 2002/03, the agricultural sector's growth declined most in comparison with all the other sectors o f the economy, mainly because o f unfavorable weather conditions. Consequently, food prices increased, bringing the 12- month inflation rate to 7.6 percent for February 2003, which also contributed to the rise in headline inflation. Increased food prices, and a steep rise in fuel prices during the latter part o f 2002, led to a rise in headline inflation to 10.9 percent in April 2003, the highest annual headline inflation rate since May 1997. Thus, achievement o f the annual budgetary projections o f 1.O percent for headline inflation and 3.5 percent for underlying inflation were not achieved. Annual headline inflation for the fiscal year averaged 5.7 percent. 1.6 Underlyinginflation, which excludes food prices, also rose to 3.3 percent inApril 2003 from 2.4 percent in April 2002. The increase was mainly attributed to increases in prices o f imported commodities, especially fuel products, which put upward pressure on the prices o f some locally manufactured goods and transport fares. The average annual rate o f underlyinginflation for fiscal year 2002/03 was 2.4 percent, and at the end o f the period, inflation stood at 10.2 percent. 1.7 Exchange rate: The Ugandan shilling has depreciated gradually against the U.S. dollar in nominal terms since December 2001. The pace o f depreciation increased at the beginning o f January, partly as a result o f uncertainty about the country's relationship with donors. The underperformance o f budget support and a steep rise in petroleum prices during the war in Iraq mainly explain the depreciation o f the exchange rate. The weakness o f the shilling was further compounded by the weakness o f the dollar against several other international currencies, which resulted in a cumulative depreciation o f 8 - 2 - percent in 2002/03. Consequently, the Bank o f Uganda intervened in the interbank foreign exchange market (IFEM) to reduce volatility in the foreign exchange market, which has since helped to narrow the spreads and minimize the rate o f depreciation. There was also a rise inthe real effective exchange rate (REER) duringthe last quarter o f 2003 mainly as a result o f the rise in domestic prices relative to Uganda's major trading partners. Because the nominal exchange rate against the dollar depreciated by 11 percent during 2002/03, and the dollar also depreciated against other currencies, the nominal effective exchange rate depreciated by more than 11percent in2002/03. 1.8 Monetary sector: The main challenge to monetary policy in 2002/03 was to effectively manage the excess liquidity created by donor-funded government expenditures ina manner that minimized instability inthe domestic and foreign exchange money markets. In line with the government's tighter liquidity policy, interest rates increased steadily, raising concerns about squeezing o f credit to the private sector. The monthly average rate on the 91-day, 182-dayY 273-day, and 364-day treasury bills increased from 6.55, 10.92, 14.42, and 15.51 percent in September 2002 to 10.95, 16.68, 17.95, and 19.13 percent, respectively, in December 2002. However, the 273 and 364 day treasury bill rates declinedto 13.97 and 13.79 by March 2003. 1.9 The government's tighter base money program did not have a substantial impact on broad money (M2) growth during fiscal year 2002/03. The 12-month expansion o f M 2 remained high. The commercial bank credit to the private sector has also increased steadily since the start of the financial year. However, real interest rates also increased. Table1.1.KeyMacroeconomicIndicators 1999/00 2000/01 2001/02 2002/03 2003/04 2004/05 projection projection Annual average inflation(%) . . 5.9 4.5 -2.0 5.7 5.9 3.3 GDP growth (market price) 5.3 5.5 5.3 4.8 6.2 6.7 Inpercentof GDP at marketprice Gross domestic investment 19.8 20.0 21.2 21.2 22.2 23.9 Gross domestic savings 6.5 5.5 7.8 6.9 8.6 8.7 Currentacct. balance(excl. -13.3 -14.5 -13.4 13.1 -13.2 -13.1 grants) Currentacct. balance(incl. -7.0 -6.8 -5.6 -5.5 -4.4 -4.9 grants) Reservesinmonthso f imports 6.1 6.1 6.2 6.4 6.0 6.6 Source; UgandaBureauof Statistics(UBOS), Bank of Uganda(BOU), and World Bank Live DataBase, External Sector 1.10 Uganda's current account deficit, as a share o f GDP excluding official transfers, widened substantially to 13.4 percent in 2001/02 and reached 13.1 percent in this fiscal year. The increase in the current account deficit i s attributed to an increase in the trade deficit: total imports increased by 8.0 percent while exports remained generally stagnant. There was an unexpected decline inthe surplus on the capital and financial accounts. - 3 - 1.11 The large import bill was financed mainly by donor inflows and to a lesser extent foreign direct investment.Private transfers, especially workers' remittances and transfers to nongovernmental organizations (NGOs), have remained strong in 2002/03. The overall stock o f gross reserves has remained sufficient and was equivalent to 6.1 months o f imports o f goods and services by the end o f 2002/03. ExportPerformance 1.12 After four consecutive years of deterioration in terms of trade, exports earnings are slowly picking up due to a rebound in the coffee sector. Coffee exports receipts fell by 53.9 percent between 1999/00 and 2001/02, but in 2002/03 they increased by 25 percent, from US$86 million in2001/02 to US$107.5 million. 1.13 Revenuefrom non-traditional exports has increased significantly in recent years, from US$335 million in 2000/01 to US$440 million in 2002/03. Revenue from fish products increased substantially, from US$50 million in 2001/02 to US$92 million in 2002/03, and i s expected to increase further inthe mediumterm, mainly as a result o f the considerable international demand for Ugandan fish. Tobacco exports are increasing steadily: the revenue from tobacco exports increased by US$48 million during fiscal year 2002/03 and i s expected to increase further inlater years. Better performance o f the non- traditional exports contributed to employment, income generation, and general private sector development. Export earnings from nontraditional exports will continue to dominate the overall export earnings o fthe country inthe mediumterm. Figure 1.2Noncoffee Export Earning I I I I 1207 I - I I 1999100 2000101 2001102 2002103 2003104 2004105 2005106 -Tobacco +Fishproducts .- .A. .-Flowers +Cotton +Tea Source: The World Bank Live Database Export Competitiveness 1.14 During 1992/93 to 2001/02, the REER appreciated by 32 percent (figure 1.3). However, terms of trade steadily declined after 1994/95. Uganda experienced a decline interms oftrade inthe late 1990s owing to a sharp decline ininternational coffee prices. This contributed to the appreciationo fthe REER, which suggests that the deterioration o f the terms o f trade put pressure on the REER to appreciate and made Uganda lose competitiveness in the international markets. However, using the consumer price index - 4 - (CPI) to calculate the REER may underestimate declines in competitiveness o f Ugandan export products because the CPI has risen at a much slower rate than the investment deflator-that is, it is. not representative o f all prices in the economy that affect competitiveness. Furthermore, other factors that affect competitiveness are not captured by the REER, suchas quality or standards. Figure 1.3UgandaRealEffectiveExchangeRateand Termsof TradeIndexes 200 150 .- .+-. I 100 * 7 50 0 ~ Note: TOT = Terms of trade. Source: The World Bank Live Database Factor Productivity and Uganda's Growth 1.15 Duringthe 1990s, Uganda achieved an extraordinary growth record, with growth rates averaging 6.7 percent between 1986 and 1999. Inaddition to economic growth and structural transformation, Uganda was also able to maintain a high rate o f poverty reduction between 1986 and 2000. More recently, however, growth has slowed and somewhat stagnated, raising concerns about whether this growth performance can be sustained inthe future. A number o f current studies have beencommissioned to analyze determinants o f long-term growth in Uganda, particularly using growth accounting methodologies to decompose real GDP growth.' 1.16 Uganda's total factor productivity (TFP) has been estimated to have been about 1 percent during the postconflict period 1986-92, and 3 percent during 1992-97 as a result o f the restoration o f macroeconomic stability, the return to economic fundamentals, and the removal of distortions. Between 1997 and 2002, TFP was 0.5 percent, reflecting a declining overall real GDP situationq2It has beentheorized that Uganda's economy i s at the point where gains from recovery have nearly been "exhausted" and to overcome the current slowdown in TFP and real GDP growth, there i s a need for deeper structural and institutional change, as well as sustained human and physical capital accumulation. 1. Some o f the recentreferenceson Ugandatotal factor productivitymeasures, as well as theoretical explanationsfor movementsintotal factor productivity,includeBevanand others (2003), Dunn(2002, Keefer(2000), Bethelemyand Soderling(2001), andthe OECD's DevelopmentCentre's study on EmergingApica (2001) . 2. TFP resultsadapted from Dunn(2002) - 5 - 1.17 One explanation advanced for the decrease in TFP i s the size o f public administration (PA). The role and size o f the public sector in Uganda's economy will need to become increasingly focused. There is a need for the public sector to focus on deepening the institutional reforms that have already begun, especially those that contribute to private sector growth. This includes the medium-term competitiveness strategy (MTCS) for promoting export growth and diversification, skill development, and reforms that will result in the deepeningo f financial markets. As the public sector takes on an increasingly focused role, there will also be a needto streamline its expenditures by reducing nonpriority items inthe budget, especially the size o f PA. The government has been rightly focusing on private sector growth as the engine for future growth, but this focus needs to be intensified. The issue should be reviewedinmore detail inthe context o fthe research agenda for the upcoming PEAP revision. MacroeconomicChallenges and Outlook 1-18 As Uganda moves toward attaining the MillenniumDevelopment Goals (MDGs), it faces a number of macroeconomic challenges interms of keepinginflation rates under control, achieving external competitiveness for exports, ensuring balance of payments and fiscal stability, sustaining buoyant economic growth, and eradicating poverty, Furthermore, the results o f the latest population census inNovember 2002 indicated that real income per capita has fallen as a result o f Uganda's highpopulation growth rate, one o fthe highest inthe world. 1.19 The economy i s still largely dependent on agriculture and exports o f basic commodities. Therefore, the performance o f the economy i s extremely vulnerable to weather and commodity price shocks. The overall GDPgrowth in the medium term i s expected to increase at an average rate o f 5.6 percent (over 2003/04 to 2005/06), assuming normal weather conditions and higher coffee production. However, this growth still remains below the PEAP target level o f 7 percent, which may jeopardize the long- runPEAP objective of reducing Uganda's poverty to less than 10 percent by 2017. In order to achieve high GDP growth, implementation o f the Plan for Modernization o f Agriculture (PMA) i s needed. The P M A will boost the food crop sector by increasing diversification. Rolling out the P M A to all subcounties by 2006/07, with its comprehensive approach to building farmers' productivity, i s expected to achieve an additional 4 percent annual growth in the agriculture sector. The main challenge for Uganda i s to sustain poverty reduction by maintaining robust economic growth. This will requirehigher rates o f private investment and the more effective use o f donor support in programs that would enhance productivity. 1.20 Uganda's supply constraints for production and trade expansion include poor physical infrastructure (roads, railways), unreliable public utilities (power, water), low levels o f education and skills, weak export institutional framework, market access problems, limited access to trade finance and market information, and cumbersome customs procedures. It i s critical that Uganda's exports, particularly nontraditional exports, do not meet new nontariff barriers. The government's short-term policy goals- such as removal and rationalization o f trade barriers and tariff reduction-are taken up in - 6 - the context o f the annual budgetary cycle on the basis o f the priorities identified in the medium-termexpenditure framework (MTEF). 1.21 Uganda's vulnerability to price shocks and declines in world prices for coffee in the last few years has become an issue meriting serious concern. To minimize adverse effects o f external shocks, diversification o f the country's export base i s very important. One o f the main concerns and challenges for Uganda i s loss o f export sector competitiveness arising from a real exchange rate appreciation that has eroded the competitiveness o f the private sector by increasing the price o f exports in terms o f local currency, thereby creating disincentives for export producers. Therefore, investments in nontraditional exports industries will be discouraged if the real exchange rate i s a disincentive to their production. To achieve a more sustainable external position and lessen the economy's vulnerability to external shocks, the government needs to focus on gradual fiscal consolidation and measures to expand and diversify the export base. To achieve higher growth and promote exports, the government needs to ensure a favorable investmentclimate for both domestic and foreign investors. Macroeconomic stability and better access to productive assets (especially land and credit) would augur well for the implementation o f the MTCS for private sector development and the P M A to promote rural development. The government may need to reconsider how the Strategic Export Program budgetary allocations should be used, with a view to improving the investment climate for all investors rather than being selective, which renders the playing field for investors uneven. B. OVERVIEWOFFISCALTRENDS 1.22 The Ugandan central government's fiscal deficit before grants, which had increased to 13 percent in2001/02 narrowed to 11.3 percent o f GDP inthis current fiscal year. This rate will be maintained in the next fiscal year before a gradual fiscal consolidation. The deficit has been financed mostly by donor aid inflows, o f which 4.7 percent o f GDP (2001/02) was net lending, while grants to the government (including Heavily Indebted Poor Countries [HIPC] Initiative assistance) accounted for 7.1 percent o f GDP in 2001/02 and increased to 7.3 percent o f GDP in 2002/03. Because o f the effect o f a larger than expected depreciation o f the shilling this fiscal year, donor budget support was about USh40 billion higher than programmed. Net domestic bank financing was 0.2 percent, and commercial banks increased their holdings o f treasury bills issued for sterilization purposes by the BOU. The widening deficit has led to large increases in liquidity created by the government's domestic fiscal operations. Domestic Revenue Trends 1.23 The performance o f revenue collection in Uganda has been poorer than anticipated. Domestic revenues increased from 10.8 percent o f GDP in 2000/01 to 12.3 percent o f GDP in2001/02 but have remained at 12.1 percent o f GDP for 2002/03, which i s 1 percent less than the approved budget estimates. The shortfall i s attributed to the underperformance o f non-Uganda Revenue Authority (URA) revenues. There has been very little change in the structural composition o f tax revenues between 2001/02 and 2002/03. The value added tax (VAT) still forms the largest share of tax revenues, - 7 - followed by excise taxes, income taxes, and, last, import taxes. Revenue collections from the VAT were above target mainly because of VAT on imports and other goods. Excise duty has not performed so well as a result of lower consumer demand for domestically produced goods. Income tax was the best performing tax in2002/03, while import duty i s expectedto register a shortfall as a result of a fail in demand for imported goods caused by external shocks andproblems with efficiency inthe administration ofthis tax. 1.24 Nontax revenue collections registered a shortfall for 2002/03 despite the government's 2001/02 introduction o f measures to enhance collections through, an 11- point reform program to improve the performance o f nontax budget revenue against targets. Most o f the reform measures focused on improving accountability o f funds previously collected as appropriation in aid by various government institutions. The nontax revenue shortfall i s partly due to the failure o f B O U dividends to materialize, as well as lower materialization o f line ministries and embassy remittances. The achievement o f revenue targets i s important because a shortfall in revenue collection will force the government to impose further expenditure cuts late in the fiscal year to avoid excessive budgetdeficits. 1.25 There is great potential for increasing local government revenue. On average, this type o f revenue currently accounts for about 10 percent o f the total funds available to the local governments. The sources o f revenue are graduated tax, market dues, trade licenses and fees, rates, rents, property tax, royalties, stamp duties, and registration fees. Graduated tax i s one of the leading sources o f revenue for local governments. The central government has undertaken studies to identify the constraints for local revenue mobilization. It has also embarked on a strategy for financing local government, which will establish the roles o f local revenues and conditional, unconditional, and equalization grants. GovernmentExpenditureTrends 1.26 Overall government expenditure has grown rapidly, from 22.4 percent o f GDP in 2000/01 to 25.3 percent in 2001/02, although it moderated at 23.3 percent in the current fiscal year (2002/03). Recurrent expenditures are expected to decrease slightly, by 0.3 percent in 2002/03, while wages and salaries will decline by 0.1 percent. Interest payments are expected to increase by 0.1 percent. Domestic interest payments were higherthanprogrammed inthe first three quarters o fthe year (2002/03) because o fhigher issuance o ftreasury bills than had beenplanned. 1.27 The government's strategy o f reducing the fiscal deficit is to cut spending. This calls for an examination o f spending areas with a view to identifying where cutting expenditures could be possible. The government i s using domestic revenue to finance expenditure areas or programs that do not get external assistance, such as security and PA. Cutting government expenditure amid demand for increased funding for certain areas (for example, education, health, and roads and works) to achieve MDGs within a limited timeframe is a major challenge that Uganda will have to grapple with in the medium term. - 8 - 1.28 The rapid growth of government expenditure is partly due to high growth in expenditure on PA. The PA challenges merit close observation, with the aim of suggestinghowgovernment spendingcouldbe cut to reducethe size ofthe fiscal deficit. C. PUBLICADMINISTRATION 1.29 Uganda's PA sector is very big, probably bigger than is necessary. It includes institutions such as State House, Office of the Prime Minister, foreign affairs, missions abroad, public service, local governments, mass mobilization, Office of the President, statutory and other commissions, the URA, and the Ministry of Finance, Planning and Economic Development (MFPED). Others are specified offices, such as presidential advisers and public service pensions. Existence of each of these institutions could be justified, and the PA sector should be adequately funded, subject to resource availability, Expenditure on PA in2002/03 was about 17.6 percent of total government expenditures. This proportion, however, is programmed to come down to about 15.9 percent in 2003/04. Nonetheless, such a proportion of expenditure on PA i s probably high comparedwith what obtains elsewhere inthe world, 1.30 Giventhe size of PA and, therefore, its adverse effect onthe generalperformance of the economy, efficient expenditure shouldbe given highpriority. PA as a sector has consistently failed to fit in the budget by running budget overruns. For example, in 2002/03, the Office of the Prime Minister had a budget overrun of 49 percent, the MFPEDhad a budget overrunof 42 percent, andthe ElectoralCommissionhada budget overrun of about 8 percent. Altogether, PA had a budget overrun of about 2 percent. Political involvement, especially political appointments, establishment o f commissions, and creation o f additional administrative units such as local governments, all of which have significant budgetaryimplications, explainthe increase inthe PA budget. Without political involvement and with adequate legal safeguards that ensure good public expenditure management, budget overruns can be managed. Giventhat budget overruns are the order o f the day, it i s certain that there i s a lot of political involvement underminingthe credibility of the budget process and imposingundesiredconstraints on budgetexecution. 1.31 The governmenthas realizedthat expenditure on PA i s adversely affecting overall budget performance, and it is trying to resist any further budgetary allocations to the sector. Inthe 2003/04MTEF, PA is programmedto receive USh365 billion-almost the same nominallevelas inthe previous fiscal year. Table1.2PublicAdministrationBudget Trends,1997/98to2002/03 Source: Ministry of Finance, Planning and Economic Development (MFPED). - 9 - 1.32 Persistent budget overruns by PA, as can be seen in table 1.2, testify to government budget indiscipline. Over the past six years, P A has claimed a big part o f supplementary expenditure. The policy o f a hard budget constraint has meant that increased spending by the PA sector i s at the expense o f spending by other sectors. Spending of other sectors has to be cut to accommodate the PA expenditure budget overruns. This has not augured well for the realization of PEAP objectives. The nonprotectedareas bear the burdeno fthe budgetcuts. 1.33 The cost o f the PA sector could be attributed to certain key areas, which include the size o f the public service, spending on the political system, and spending on the administrative system. Public service reform has hitherto not addressed the matter of the size o f the public service (excluding education and health). Political institutions include parliament, the Movement Secretariat, and a number o f constitutional and nonconstitutional commissions. The country has about 67 cabinet ministers, about 310 members of parliament, 35 advisers and special assistants, and about 90 district commissioners and their assistants. On the administrative front, there are about 970 local governments (districts, municipalities, and subcounties) with implications for the size o f public service. 1.34 In addition to the constitutional commissions, there are a growing number of nonconstitutional commissions. Emerging commissions have introduced new challenges inplanning and budget execution. Some commissions are given semiautonomous status from the ministries within which they are located. An example o f this i s the Amnesty Commission located in the Ministry o f Internal Affairs, which has autonomous status with regardto budgetpreparation and execution for the commission. 1.35 The mandates o f different institutions are sometimes conflicting, which leads to duplication o f activities. For example, besides the MFPED, there i s a National Planning Authority in the president's office. Similarly, presidential advisers work alongside cabinet ministers. The Uganda InvestmentAuthority, the Ministry o f Tourism, Trade and Industry, and the Uganda Tourism Board all handle related or similar activities. Duplication has an added cost in terms o f operational costs. The issue o f the size o f government, which i s a main cause o f duplication o f activities, i s a matter that requires immediate attention. 1.36 However, duplication sometimes arises from uncoordinated donor activities, which generate different units with the same or similar mandates. Some donors have supported establishment o f certain units whose activities would have been handled by relevant line ministries. For example, some have expressed the view that it i s better to support capacity development inthe Ministry o f Tourism, Trade and Industryrather than supporting the establishment o f the Uganda Investment Authority, which has a similar mandate. D. AID, EXTERNAL DEBT,AND FISCAL SUSTAINABILITY 1.37 Uganda's external debt sustainability indicators have deteriorated since reaching the completionpoint o fthe enhanced HIPC framework inMay 2000. However, only one - 10- of the indicators, net presentvalue (NPV) of debt to export ratio, has exceeded the HIPC guidelines. The flow of debt service indicators remains well within the HIPC boundaries. The deterioration of indicators was primarily caused by a sharp decline in international coffee prices that substantially lowered Uganda's export earnings. The NPV of debt to exports ratio was at 209 percent in 2002/03, which compares unfavorably with a threshold of 150 percent. Debt sustainability is vulnerable to a substantial and prolonged slowdown in economic growth. It i s also vulnerable to the government contracting excessive amounts of newborrowing, even on concessionalterms. 1.38 The International Development Association (IDA) is Uganda's largest creditor, and to reduce the debt burden over time without reducing donor support, it would be necessary to increase the grant element of IDA flows by making them a mixture of pure grants and concessional loans. Under the Debt Sustainability Analysis (DSA), two scenarios of IDA lending and grant financing were examined. Uganda being a debt- vulnerable country, up to 20 or 40 percent of IDA allocation will be provided as pure grants under IDA 13 arrangements. Increasing the grant element o f disbursementsby 20 percent will reduce the peak of the NPV of debt to export ratio, allowing this value to reach the 150 percent target by 2012/13. Assuming that 40 percent of IDA allocations will be providedon a pure grant basis during the period 2002/03 to 2004/05, the NPV of debt to export ratio will fall more rapidly than projected, falling below the 200 percent mark in 2003/04, to reach the 150 percent target two years earlier, in 2010/11. Hypothetically, iffuture IDA arrangements were also to include the pure grant element, the NPV ofdebt to export ratio would fall belowthe 150percenttarget by 2008/09. 1.39 Uganda is heavily dependent on donor assistance, and it will remain dependent on donor support for its poverty reduction efforts in the medium term. Overall official assistance (both grants and loans) has increased. Grant financing has increased from 5.0 percent of GDP in 1998/99to 12.2percent o f GDP in2002/03. This increase is mostly in project grants. Loans financing slightly decreased from 4.1 percent of GDP in 1998-99 to 3.7 percent in 1999/00 and increased to 5.7 percent of GDP in 2001/02. Import support loans increased significantly, from 1.1 percent of GDP in 1998/99 to 3.4 percent of GDP in 2001/02. However, project loans have been on a declining trend since 1998/99. 1998199 1999100 2000101 2001102 2002103 Total Official 9.1 9.5 11.7 12.8 12.2 Grants 5.O 5.8 7.6 7.0 6.9 Import support 2.0 2.7 3.7 3.5 3.9 Project grants 3.0 3.1 4.0 3.5 3.O Loans 4.1 3.7 4.1 5.7 5.3 Import support 1.1 0.7 1.5 3.4 3.3 Project loans 3.0 3.1 2.6 2.3 2.0 1.40 Trends in project and program support have changed substantially over the last few years. Compared with the situation that prevailed in the past two years, project - 11- support has declined while program support has increased. Program support provides the flexibility to fully integrate recurrent and development spendingunder the MTEF. Figure 1.4 Uganda TrendsinProject andProgram Support (99 of GDP) x . . ....* * 1998199 1 I999100 2000101 2001102 2002103 - .x .- Project support --+- Programsupport 1 ~ Source: The World Bank Live Database 1.41 Some believe the overall resource envelope to support poverty reduction in Uganda i s still far short of the requirementsfor achieving the MDGs. It should be noted that budget support finances a greater share o f social services or nontradables than does project support. Hence, the objective o f reducing the fiscal deficit will face intense challenges in the medium term from those sectors requesting additional resources to address priority needs. E. UGANDA'S FISCAL DEFICIT CHALLENGESOVERTHEMEDIUM TERM 1.42 In the last two years, the issue o f Uganda's fiscal deficit has moved to the forefront o f the economic policy debate because o f a sharp rise inthe deficit that occurred between 1998/99 and 2001/02. The MFPED, su ported by the Bank o f Uganda, has argued strongly in papers and public presentations P that the fiscal deficit has grown too large, to the extent that it threatens both macroeconomic stability and the sustainability o f public finances. Consequently, the MFPED plans to gradually scale back the fiscal deficit over the mediumterm, starting in2002/03. 1.43 Beginning in the second half of the 1990s, Uganda enjoyed an increase in inflows o f budget support, including debt relief. Gross aid inflows increased by about 5 percentage points o f GDP, from about 8 percent in the late 1990s to 13 percent in the current fiscal year. This helpedto finance a steep rise in total government expenditures (including donor-funded projects that remained relatively flat as a percentage o f GDP) from about 17 percent o f GDP in 1997/98 to more than 25 percent in 2001/02. Unfortunately, the domestic revenuemobilization effort virtually stagnated inthis period: the increase between 1997198and 2001/02 was only 1.7 percent o fGDP, so that domestic 3. See GovernmentofUganda(2003a, chapter 3) and Brownbridge (2002) for a comprehensivediscussion ofthis issue. - 12- revenues reached only 12.3 percent o f GDP in 2001/02, reflecting (among other things) serious problems intax administration, including corruption (see figure 1.5). Figure 1.5GovernmentExpenditure,DomesticRevenues,andAid (99 of GDP) 30% 25% 20% tGovtExpenditure 15% -t- Revenue Domestic - - I 10% m -.-.--.L__-- ,- I 5% ~ 0% 1995196 1996197 1997198 1998199 1999100 2000101 2001102 2002103 Source: MFPED 1.44 Consequently, with a steep rise in government spending not matched by domestic revenues, the overall fiscal deficit, before grants, ballooned from only 6.3 percent o f GDP in 1997/98 to 13.1 percent of GDP in 2001/02 (see figure 6). Although the rise inthe fiscal deficit after grants was not quite as steep -from 1.1 percent o f GDP to 5.7 percent- it is the fiscal deficit before grants that is most relevant from the standpoint o fthe impact o f the fiscal stance on the macroeconomy. 1.45 In the 2002/03 fiscal year, the MFPED began a process of fiscal consolidation designed to bring the fiscal deficit back over the medium term to what is regarded as more sustainable long-run level, which i s probably inthe region o f 6-7 percent o f GDP. The projected outtum for 2002/03 is an overall deficit before grants o f 11.1 percent of GDP . 1.46 The concems of the MFPED and the BOUwith the rising fiscal deficit are (a) the impact on relative prices inthe domestic economy, particularly the real exchange rate and the cost of investmentgoods; (b) the impact on domestic financial markets, especially the very shallow money market; and (c) the vulnerability o f a government budget that relies on donors for half o f its hnding to any significant cutback indonor aid, and the knock-on effect this would have on the macroeconomy. - 13 - Figure 1.6FiscalDeficit, Excluding Grants (% of GDP) 14% 12?? lP? 80/0 6% 4% 2?? P ? 1595196 1!%57 1937/98 1 W 1W/@l W O 1 X13oOvoZ 2032103 Source: Ministry o f Finance,Planningand EconomicDevelopment(MFPED). 1.47 Economic theory predicts that an increase in a fiscal deficit, mainly funded by donor aid, would have effects analogous to that o f the "Dutch disease"-that is, there will be an appreciation o f the real exchange rate caused by the expansion in demand for nontraded goods. This has occurred inUganda over the last five years, with the prices o f nontraded goods rising at a much faster rate than traded goods' prices despite the 35 percent fall in the external terms o f trade that occurred during this period. The MFPED also presented evidence to the Public Expenditure Review (PER) in May 2003 to show that the unit costs of public construction rose particularly fast, by 70 percent between 1997/98 and 2002/03, and in doing so dragged up the cost o f construction for the private sector and hence the cost o f investment. The conclusion drawn by the government i s that, during a period when the economy required a real exchange rate depreciation to maintain external competitiveness in the face o f a sharp fall in the terms o f trade, the rising fiscal deficit had driven the real exchange rate in the opposite direction. Some, however, argue that the answer may depend on how the government can better reallocate its resources without the adverse effect o f public construction costs. It i s also clear that some o f these rising public construction costs have been due to poor procurement practices, which needto be addressed anyway. 1.48 The second problem arose becausethe increase ingovernment spendingunder the government budget, which was funded mainly by the expansion in budget support, ledto a creation o f high-powered (base) money by the BOU. (High-powered money i s created when the government spends its deposits in the BOU). The magnitude o f the increase was enormous. In 1997/98, the high-powered money created by fiscal operations amounted to around USh 60 billion, but this figure had risen more than 10-fold by 2001/02, to a level equivalent to 6 percent o f GDP. If the injection o f high-powered money was not to fuel an inflationary increase in broad money, the B O U had to sterilize most o f it through open market operations and by intervening in the domestic IFEM to sell foreign exchange inreturnfor domestic currency. - 14- 1.49 The problem faced by the BOUwas that both the domestic money market and the IFEMwere too shallow to absorb the level of interventionrequiredto mop upthe amount of high-powered money created by fiscal operations. The consequences were most clearly felt in the money market, where net issuances o f treasury bills rose from USh 30 billion a year inthe late 1990s to USh268 billion in2001/02. Increased net issuances o f this magnitude could be purchased only by the commercial banks, but at the expense of their lendingto the private sector. While the share of government securities inthe bank's asset portfolios increased by around 10 percentage points inthe space o f two years up to June 2002, the share of loans to the private sector fell by a similar amount. Private sector credit, which had recovered strongly during the 1 9 9 0 ~stagnated from 1999/00 to ~ 2001/02. In effect, the rising fiscal deficit had crowded out private sector borrowers because o f the large increase intreasury bills sold to the commercial banks to mop up the liquidity created by government spendinginexcess o f domestic revenue. 1.50 The BOU's net sales to the foreign exchange market were virtually zero inthe late 1 9 9 0 ~but they rose very quickly because o f the need to sell foreign exchange to help ~ sterilize the high-powered money created by the fiscal deficit. By 2001/02, around US$200 million o f foreign exchange was sold inthe IFEM, whichrepresentedclose to 20 percent o f private sector supplies o f foreign exchange to the IFEM. Selling this amount o f foreign exchange in the IFEM threatened to make the exchange rate less stable, and there were times during 2001 when the exchange rate appreciated sharply. Because o f its concern for exchange rate stability, the B O U was unable to sell all o f the foreign exchange it had planned to sell to the IFEM, which put an even greater burden on the domestic money market to control the expansion o f base money. 1.51 The reduction in the fiscal deficit achieved in 2002/03 brought rapid benefits to the domestic financial markets. Net issues of treasury bills were less than half the level inthe previous fiscal year, and this created more space for a rebound in lending to the private sector by the commercial banks. Private sector credit rose by 26 percent between June 2002 and April 2003. Nevertheless, eventhough the fiscal deficit had beenreduced in 2002/03, there was still strong pressure in money markets as indicated by very high treasury bill rates (between 15 percent and 20 percent) during a large part o f the fiscal year. 1.52 However, it i s debatable whether the government got the right balance interms o f treasury bill and foreign exchange sales in its sterilization efforts, such that a better balance might reduce the adverse effect. One could also argue that when more aid i s usedto finance imports, that will reduce the degree o f liquidity creation andthe problems o f sterilization will become less important. These issues are likely to remain an active part o f the aid management debate inUganda inthe medium term. 1.53 The third area o f concern articulated by the MFPED and the B O U relates to the risks to both public finances and the wider economy when half of the nation's budgetary resources are derived from sources outside government control and where the external sector relies heavily on sales o f foreign exchange by the central bank. A substantial reduction in donor funding, other than a purely temporary reduction that could be absorbed through sales o f the BOU's foreign exchange reserves, would risk destabilizing - 1 5 - the budget and the macroeconomy. The scope for rapid cutbacks in the government budget i s quite limited, because a large part of the budget is made up o f what are in reality nondiscretionary expenditures such as wages or statutory expenditures. Hence, a sharp reduction in budgetary resources would have the double effect o f severely squeezing the discretionary components o f the budget, which would inevitably damage public service delivery and force government to resort to domestic bank borrowing to bridge the gap in budgetary resources. At the same time, the B O U would have to cut back on its foreign exchange sales to protect its foreign exchange reserves: this would risk triggering a collapse o fthe exchange rate, which wouldbe exacerbated ifdemand for domestic currency contracted inthe face o f an erosion o f confidence inthe stability o f the economy. The economy's vulnerability to such a shock can be mitigated only if the dependence o fthe budget on donor funding i s gradually reduced and consolidated. 1.54 Another important risk i s vulnerability to substantial cost arising from contingent liabilities and potential pension arrears. The MFPED i s currently compiling a comprehensive list o f all contingent liabilities at the request o f the International Monetary Fund (IMF). The main contingent liabilities entail legal claims against the government, many o f which amount to several million dollars each and against which the government through the Ministry o f Justice and Constitutional Affairs has failed to mount an effective legal defense to protect public funds. Several stakeholders have on a number o f occasions at public forums voiced their concern about corruption involving legal cases against the government. This highlights that internal reforms to the Ministry o f Justice, and other organs relevant to addressing these issues o f competence and integrity inthose institutions, should be a component o f any future public expenditure reform program in Uganda. F. MEDIUM TERMCHALLENGES ANDTHEWAY FORWARD 1.55 In the medium term, Uganda will face challenges in sustaining the implementation of the poverty reduction strategy due to low revenue efforts, the need to reduce the fiscal deficit and the overextension o f government activities that has left expenditure in many sectors spread far too thinly, and the risks o f starving core areas o f the budget of essential levels o f funding. The government has already emphasized the need to control the fiscal deficit to maintain macroeconomic stability and contain pressure on real interest rates and the exchange rate arising from the large expendituresto support the poverty reduction program. As already noted, given the current trends inthe fiscal deficit, the appropriate balance o f sterilization instruments and the composition o f public expenditurewill remain important issues for discussion over the mediumterm. 1.56 There i s concern about commercial bank lendingrates, which have remained high (in the range of 20-30 percent) for a long time. High lending rates are partly being blamed on the mix o f instruments used to sterilize excess liquidity. One school o f thought i s that large issuance o f treasury bills i s the major cause o f highlendingrates. In this view, excess liquidity should be sterilized mainly through sale o f foreign exchange. The other school o f thought counters this argument with Dutch disease effects, which could lead to poor competitiveness o f Uganda-produced goods inexternal markets. - 16- 1.57 The government i s o f the view that the fiscal deficit should be reduced through cutting government spending, which would assist in reducing the issuance o f treasury bills and thereby drive interest rates down. The strategy o f reducing the fiscal deficit through cutting government spending i s said to be the most feasible, because increasing the tax revenue to GDP ratio is hardly possible. However, achievement of MDGs could be difficult ifthe government cuts spending. Inaddition, cuts ingovernment expenditure would not be feasible inthe short runwithout large disruptions inthe poverty eradication program. Inreality, government expenditures are set to increase in the mediumterm, in the context o f both national development goals and MDGs, therefore emphasis must be placed on the rate at which this increase takes place. At the same time, there are gains to be made on the revenue side through the implementation o f measures to improve tax administration. Ideally, in the short to medium term, the rates o f growth o f government revenueandexpenditure should be consistent with fiscal deficit reductiontargets. 1.58 There i s hardly any support for the proponents o f the fungibility effect o f aid flows, whereby external aid crowds out domestic revenue mobilization. More aid does not seem to be adversely affecting Uganda's tax effort. However, there are concerns about Uganda's revenue to GDP ratio, which has stagnated between 11 percent and 12 percent for about five years. Tax policy officials say that the tax base cannot be widened further, yet there are other countries at a similar level of development in Sub-Saharan Africa with a revenue to GDP ratio o f 18-20 percent. The issue o f domestic resource mobilization i s an ongoing debate that merits further research. There i s a needfor further dialogue on the revenue-GDP ratio issue with a view to exploring ways o f widening the tax base and registeringfurther improvements intax administration. 1.59 Initial efforts have already beenmade, bringingtogether the relevant stakeholders, the government, the central bank, the donor community, and research institutions in a forum to discuss the way forward on the fiscal deficit issue. In addition to more analytical work on the available options, further dialogue i s also needed on the way the fiscal deficit should be handled. There i s consensus that the fiscal deficit should indeed be reduced gradually, but there is no consensus on what strategy would be appropriate for reducing it. 1.60 Inview ofUganda's development needs, there is need for further dialogue on aid effectiveness and absorptive capacity. Much as aid could have destabilizing macroeconomic effects, absorptive capacity o f aid could be enhanced through efficiency measures, which are yet to be fully explored. 1.61 Although an overvalued exchange rate may not adversely affect traditional exports, it i s likely to adversely affect the performance o f nontraditional exports, which are giving a lot o f hope to Uganda for faster export growth. Further work on exchange rate effects on the real sector would assist inmaking o f best policy choices. 1.62 On the whole, there i s need for further dialogue on policy options with a view to addressing various concerns that have been raised-such as high lending rates, decreasing growth rate o f private sector credit, appreciation o f the real exchange rate, absorption o f aid, government crowding out the private sector from the credit market, - 17- government crowding out the private sector inconstruction (public construction i s said to be driving up the prices of construction), efficiency inthe use of aid resources, choice of a mix o f instruments for sterilizing excess liquidity, and so forth. 1.63 The ongoing PEAP revision process could address some o f these concerns. Furthermore, the process o f preparing a long-term expenditure framework (LTEF) could consider some of these concerns. These important issues will form part of the dialogue for the forthcoming discussion of the 2004/05 budget, which begins in October 2003 with the annual budgetworkshop. - 1 8 - 2. UGANDA'S BUDGETPROCESSAND MTEF, 2003/04 TO 2005/06 A. INTRODUCTION 2.1 Uganda's budget process has increasingly become participatory, involving key stakeholders such as government line ministries and departments, local governments (LGs), parliament, donors, civil society, and to some extent the rank and file of beneficiaries. Sector working groups (SWGs) play a key role inthe budget process. The budget process has two phases: phase 1 involves communication by the MFPED to SWGs o f their budget ceilings. It also involves preparation o f budget framework papers (BFPs) by SWGs. Phase 2 o f the budget process involves preparation o f the MTEF by the MFPED, approval o f the MTEF by the cabinet, presentation of the cabinet-approved MTEF to parliament, consultation with all other stakeholders in a PER meeting, and reading o f the budget. The process starts in October, and the reading o f the budget takes place inJune o f the following year. 2.2 The MTEF takes into consideration the overarching objectives o f economic management-namely, maintenance o f macroeconomic stability and economic growth. Thus, the macroeconomic framework plays a key role inthe budgetprocess. The size of the budget deficit and its financing options must be consistent with the macroeconomic framework. 2.3 The MTEF has substantially improved budgetpredictability. The transparency o f the budget process has improved the quality of budget planning, budget choices, budget discipline, and predictability o f the budget. However, recent developments in defense and public administration expenditure-namely, overruns and non programmed allocation late inthe budget process-are contrary to the budget discipline that has been increasingly firm since 1997/98. 2.4 The government's preferred mode o f receiving aid from donors is by way o f budget support. It has argued that channeling aid through budget support would increase budget predictability, enhance accountability, and improve the impact o f public expenditures. The government is concerned that donor inflows are sometimes not predictable and releases are often less than commitments, a situation that complicates management o f the budget process. It has decided that the MTEF ceilings should reflect both budget and project support, inrecognition that some donors are likely to take a long time before accepting to channel their assistance to Uganda through the budget support modality. 2.5 Several donors have responded favorably to the government's request o f providing assistance to Uganda in the form o f budget support. For example, the World - 19- Bank provides almost half o f its assistance to Uganda inthe form o f general budget, and the United Kingdom's Department for International Development (DFID) has moved almost exclusively to budget support. Inthe course o f the year, some donors have moved increasingly to budget support (for example, IrishAid, the EuropeanUnion [EU], and the Netherlands). The Danish International Development Agency (DANIDA) i s still considering the budget support modality. Some donors have preferredto provide general budget support, and others (especially Irish Aid and the EU) have preferred to give Poverty Action Fund (PAF) budget support. In addition, there are still a few basket fund-type arrangements in certain sectors (such as water and justice, law, and order) to which donors provide their assistance as earmarked budget support. The move to budget support enables the discussion o f the budget to move beyond allocations or budget choices to the overall outputs and outcomes o f the budget process. However, the move has raised the level o f complexity o f managing donor expectations o f reporting (for fiduciary concerns) and local government expectations o f decentralization. The mechanism has also raised the more complex issue o f low expenditure ceilings for social sectors such as health and education, where additional resources would be needed to address MDGs and for which some donors want to earmark funding. However, after a series o f workshops, various stakeholders have understood the logic behind sectoral expenditure ceilings. 2.6 The World Bank i s Uganda's leading partner in the budget process and has provided support to Uganda for budget consultations, especially the PER process. In addition to the Bank's traditional role in supporting the PER process, it has worked closely with the PER working groups to deepen participation o f various stakeholders in the budget process. The Bank has facilitated dialogue and research on issues o f budgetary concern, such as reviewing and assessing budget choices and ensuring value for money (through support to budget tracking studies). Thus, the Bank has supported both phase 1 and phase 2 o f the budget process. Various development partners have provided technical assistance through SWGs with a view to deepen the budget preparationprocess and improve budget choices over time. 2.7 After some new developments, especially o f the budget support modality of aid delivery, the agenda for discussion of public expenditure issues has expanded tremendously in recent years to cover the whole o f budget management issues and not merely a review o f budgetallocations. Thus, the discussion includes, among other issues o f fiscal decentralization, financial resources management at all levels o f government, revenue collection issues, donor commitments and disbursement issues, and the role o f civil society in the budget process. PER work is increasingly accommodating budget preparation, execution, accountability, and monitoring issues at the central and LG levels. The PER process has, therefore, provided a system for methodically tracking progress in budgetprocess and execution and identifyingremaining challenges. B. PHASE 1OF THE BUDGET:BUDGETCEILINGS AND BUDGET PREPARATION 2.8 The annual consultative budget workshop, which usually marks the beginning o f the budget process, was heldinOctober 2002 and involved all key stakeholders. As table - 20 - 2.1 shows, the budget process was in line with the usual budget program and, compared with the previous year, there was more time for the BFP preparation stages. Table2.I Budget Processfor the Year2003/04 Activity Date Consultative budget workshop Oct 2002 Budget workshops for LGs Oct 2002, Jan 2003 Preparationo f sector BFPs Nov-Dec 2002 Submissionof sector BFPs Dec 2002, Jan 2003 Interministerial consultations on BFPs Jan-Feb 2003 Submissiono f national BFP to the cabinet March2003 Submissionof macroeconomicplan and indicative April 1, 2003 budget framework to parliament Comments from parliament and PER workshop Mid-May 2003 Budget day June 12,2003 Source: MFPED 2.9 There were marginal improvements in this year's budgetary process in terms of frequency o f SWGs' meetings to discuss issues related to their sectors' budgets. However, sectors that have traditionally performed poorly inthis respect, such as P A and economic services, demonstrated a better effort. The PA and economic services groups sat occasionally and had rich discussions. This improvement notwithstanding, these two SWGs need to continue the process even after their BFPs have been submitted, mainly because-unlike the education, health, and justice, law, and order sectors, which have sector-wide approaches (SWAPs) and annual sector reviews-their activities are less permanent and less continuous throughout the year. In terms of content of their discussion, the P A group needs to develop more capacity for budgeting, particularly ensuring that planning is a lot more realistic than it i s at present. P A also needs to start making some hard choices, which they seem to have avoided so far. 2.10 The problems highlightedabove occur mainly because budgets o f SWGs are not realistic: there i s overbudgeting, with too many planned undertakings that are not consistent with given budget ceilings. This has tended to move the focus o f discussions away from results to be achieved with limitedresources to a demand for additional funds. An example of this i s the pupil-teacher ratio (PTR) inthe education sector: it may not be possible to attain the desired PTR within the given budget ceilings because the education sector allocation i s not set to increase substantially inthe near future. 2.11 Also, not all sectors abide by the budget process guidelines. For instance, the defense sector does not prepare detailed BFPs. This makes it extremely difficult to discuss the defense priorities in the context o f the overall budget process. It i s clear that classified aspects o f the defense budget cannot be part o f the BFPs, but the sector can prepare useful BFPs that reflect its priorities and allocations. Because defense increasingly absorbs a large share o f the budget, it i s important that defense i s encouraged to prepare BFPs like all other sectors. 2.12 Agriculture i s another sector whose BFP has become more complex. Many sectors contribute to rural development and P M A objectives, thus it i s important that the - 2 1 - agricultureBFP cross-references other sectors' input-output links. The health sector, for example,has madesubstantialprogressinthis area. 2.13 There was some improvement in output-oriented budgeting (OOB) and results- oriented management (ROM), which were highlighted in the BFP process. At the October budget workshop, special sessions were conducted to provide guidelines for SWGs on OOB and ROM. Sectors have set ROM as an operational objective at the institutional level andtried to matchthat with goals at the sectoral level. Although there has not yet been an audit to determinethe success o f this initiative, it i s encouragingthat efforts are being made to focus on results and outputs. The Ministry o f Public Service (MOPS) has gone a long way in terms of covering all line ministries and government departments on ROM and OOB. In SWG meetings, the MFPED encourages them to articulate their outputs and outcomes. However, some SWGs have yet to fully differentiate betweenoutputs andoutcomes. This couldexplainwhy ROM and OOB are not so well linked with the budget process in some cases. Where ROM has become somewhatestablished, as inthe healthand educationsectors, the challenge is to integrate it into staff performance appraisal. This year, the health sector did make some progress on this ROMarea as well. 2.14 The IntegratedFinancialManagementSystem(IFMS) links budgetingto financial managementandhas helpedthe operationalizationofthe OOB andROMprocesses. The IFMSis activity or output based. Disbursementof funds from the MFPEDto cost centers i s basedon progress in terms of achievement of outputs andresults. Such disbursements are not broken down in terms o f development and recurrent expenditure, but rather in terms ofplannedactivities or targetedoutputs andresults. It is at the levelof cost centers that differentiationbetweendevelopment and recurrent expenditure i s possible. Infiscal year 2003/04, all cost centers will use the new IFMS. Duringthe previous financial year, consultants developed the budgeting module and the general ledger module that cost centers will use inthe system's implementation. 2.15 Duringthe previous financial year, implementationofthe IFMS was pilotedin 10 cost centers, made up of six ministries (the Ministry of Finance, Planning andEconomic Development [MFPED]; Ministry of Health [MOH]; Ministry of Water, Lands and Environment [MWLE]; Ministry of Works, Housing and Communication; Ministry of Agriculture, Animal Industries and Fisheries [MAAIF];and Ministry of Education and Sports [MOES]); and four LGs (Bushenyi and Lira districts, Jinja municipality, and the Kampala City Council). Lessons learned from the pilot implementation of the IFMS were taken into considerationinthe development of the budgetingmodule andthe ledger module,Other cost centers implementingthe IMFS for the first time will draw uponthe experienceso fthe pilot cost centers. Whenthe IFMSbecomesoperationalinall LGs, the MFPED will be able to generate a user-friendly report that can be used by planners, accountants, andpolicymakers. 2.16 Intersectoral linkages are improvingthrough SWG meetings, and this i s reflected inthe BFPs, notablythose of the education, health, andwater and sanitationsectors. The health and education sectors refer in their BFPs to their memorandumof understanding regarding their mutual work. However, the implementation of the memorandum of - 22 - understanding i s running behind schedule. The health and water and sanitation sectors are jointly addressing issues in mortality indicators. The economic services sector also tried to carry the discussion into the area o f intersectoral linkages between certain ministries (for example the Ministry o f Energy and Minerals and the Ministry o f Tourism, Trade and Industry). However, despite the progress in intersectoral linkages, there will be a need to revisit the issue because some cost centers have yet to fully internalize the differences betweenoutputs and outcomes and some are unwillingto work as a team-especially those SWGs that are not natural constituencies, such as education or health. 2.17 Prioritization o f activities inthe BFPs i s still wanting inmany respects, especially inrelation to PEAP targets and MDGs. All sectors need further help inthis area. The PEAP priorities are seldom articulated in the BFPs, which means that PEAP is not a priority for BFPs. Inaddition, this PEAP linkages issue has not beenaddressed in depth in individual sectors, and it does not seem to be feeding through well to the budget process. Some sectors have done better than others-for example, prioritization i s better done in education than in PA, which does not budget as effectively. The PEAP revision process in progress has identified this problem and there i s hope that it will be fully addressed. 2.18 Cross-sectoral issues have been getting a relatively low priority in the BFPs. In this regard, integration of gender issues into BFPs has been highlighted as needing additional assistance. With the World Bank technical assistance, the MFPED and the Ministryof Gender carried out workshops to sensitize and introducetools for integrating gender issues in the BFPs at the LG and central government levels. The benefits o f this exercise are likely to be reflected in the forthcoming budget process and PEAP revision activities. 2.19 Duringthe previous year, fiscal decentralization was implementedin 15 LGs (13 districts and 3 municipalities). The Fiscal Decentralization Strategy (FDS) i s intended among other things to address the problem o f too many conditional grants that have led LGs to operate very many bank accounts-in most cases, more than 100. LGs are required to prepare a report and account to the MFPED for every conditional grant received, thus making them preoccupied with preparation o f reports and accountability at the expense o f activity implementation. The new chart of accounts developed under the IFMS i s intendedto address the problem o f too many bank accounts and the associated problem o f reporting to the MFPED. The system gives some flexibility to LGs by allowing them to reallocate 10 percent o f conditional grants to other uses in line with their priorities. However, for every conditional grant, there is a specified minimum proportion o f resources that must be spent on the earmarked activity. Over time, as LGs improve their financial management and planning, conditional grants will reduce. Achievements during the previous fiscal year inthis respect included the development o f the planning and budgeting manual and the financial management, accountability, and reporting manual. Furthermore, two important committees were set up: the Releases Committee in the MFPED and the Local Government Committee in the Local Government Commission. As pointed out already, the year marked the piloting o f fiscal decentralization in 15 LGs. - 23 - 2.20 Regardingpay reform, the MOPS and the MFPED prepared a draft position paper inJuly 2003. The paper takes into consideration (a) the size of public administration (a report on this has been completed and presented to the highest authorities, (b) an assessment o f wage bill pay reform implications o f establishments in social sectors (the report on this was completed inJuly 2003), and (c) a study o f skills gaps inpublic service (a draft report was completed in July 2003). The paper will reinstate the policy o f transparency in the pay system by calling for an end to recently introduced nontaxable fuel and mobile phone allowances for political leaders and top civil servants. These reports and the proposed Capacity and Performance Enhancement Project (CAPEP) are intendedto strengthenPublic Sector Reform Program implementation. 2.21 During fiscal year 2003/04, a provision of USh 15 billion has been made to provide for salary increases, with a view to achieving a single-spine salary structure. This is inline with the policy objective of enhancing salaries for staff with technical and professional skills, who are difficult to recruit and retain in the Ugandan public service. Thus, for example, salaries of systems analysts, accountants, auditors, economists, and so on will rise by 41 percent, while those of semiskilled and unskilled workers will rise by less than 10 percent. Among the social sector workers, teachers will receive the highest salary rise o f 9% percent, but their salaries remain low compared with those o f nursesand other health workers, who received an excessively high salary rise two years ago. On the whole, the salary adjustments in fiscal year 2003/04 as planned by the MOPS are consistent with the primary pay policy objectives. The medium-termprognosis o f the wage bill points to annual funding shortfalls o f USh3 0 4 0 billion, a matter that should be addressedthrough reform o fthe public service (see table 2.2). Table2.2MOPSBudget (MTEF)Proposal for WageBil. 2003/04to2005/06(USh Billion) Wage bill category 2002103 2003104 2004105 2005106 Defense 128.00 132.00 136.89 141.00 PAF-teachers 185.07 201.58 212.38 225.48 PAF-others 47.70 49.37 49.43 49.52 Other 266.26 262.47 282.25 296.23 Total 627.03 645.42 680.95 712.23 Allocationfor pay reform 15.00 34.95 48.25 Pay increasefor lower-levelstaff 30.80 33.90 37.29 Totalrequirements 698.45 737.91 773.28 MTEF 627.03 667.65 704.01 735.99 Fundinggap 30.80 33.90 37.29 Source: Governmentof Uganda2003a. 2.22 The future number o f government employees, which has a large bearing on salary levels, i s not yet determined. Intheory, it could increase or decrease. As noted already, there are ongoing initiatives to control the size o f the public service. A report on this has been submitted to the highest authorities, and after consideration, there is a general agreement and commitment within the government to reduce the size o f the public - 24 - service. The MOPS and the MFPED are expected to take up this matter with the Head o f Public Service/Secretary to the Cabinet. 2.23 Giventhe assumptions about growth inemployment that have beenacknowledged by the MFPED (and budgetedfor in the MTEF), the allocation for pay reform in fiscal year 2003/04 i s inadequate. The MFPED should review its budget for pay reform unless it intends to suspend implementation of pay reform in2003/04. Inshort, the MOPS and the MFPED should close the gap in their perspectives and budgeting for the wage bill, beginningwith the nextbudget. Under a scenario of achieving pay reform targets within three years, the wage bill would need to rise from about USh 607 billion today to about USh860 billion infiscal year 2005/06 (see table 2.1). Inother words, an annual increase inthe wage bill of about 20 percent would be required in each of the next three years. The MTEF, however, provides for only a 5 percent adjustment inthe wage bill ineach of the three years, mainly because of equity considerations. For example, the current figure inthe 2005/06 MTEF is USh 746. It should be noted that increases of 20 percent per year in the wage bill are unlikely unless there i s a drastic reduction in the number o f public service employees, which i s also very unlikely. 2.24 The proposed road map for the Integrated Payroll and Personnel System (IPPS), completed by the IPPS implementation committee (MOPS, MFPED, and Uganda Computer Service (UCS), i s on course to ensure its implementation. The total cost o f the IPPS will be approximately US$7 million, and its sources are not yet identified. The MOPS should prepare a funding proposal for the development partners. Underthe Local Government Development Program (LGDP), an initiative to decentralize payrolls to districts independent o f the IPPS has been under way for the past two years. This has raised two concerns on the part o f the MOPS: First, the initiative lacks strategic focus, thus the system may not dovetail with the IPPS; second, the decentralized payrolls are weak ininternal controls and checks and could be a new source o f "ghost workers" on the payroll. In August 2003, the MOPS mobilized an interministerial team to work with an independentconsultant in assessingthe system inRakai district. The team i s evaluating performance o f the system and lessons learned in the management o f decentralized payrolls. c. PHASE 2 OF THE BUDGET:PARTICIPATIONAND BUDGETPRESENTATIONTO PARLIAMENT 2.25 The PER workshop was held on May 19-20, 2003, to discuss the current year's budget execution (for the three quarters for which information is available). The vice presidentopened the PER meetings, and several cabinet ministers attended. Memberso f parliament, led by the Parliamentary Committee on the Budget, and representatives o f civil society, academia, and donors also took part. The participants discussed a wide range o f issues pertaining to public expenditure, including budget efficiency and allocation issues. This year, the PER focused on allocative efficiency and operational efficiency o f public expenditure in four areas: water; local government; justice, law, and order; and health. - 25 - 2.26 The MFPED presentedthe macroeconomic framework and the MTEF for 2003/04 to 2005/06 at the PER workshop. It was noted that because o f an increase in externally funded projects, there would be no reduction inthe overall budget deficit, before grants: the budget deficit will remain at 11.4 percent of GDP, the same as inthe previous year. The government's medium-term budget plans entail reductions in the overall budget deficit to about 9.7 percent. As noted in chapter 1, reductions in the budget deficit are deemednecessary to maintainmacroeconomic stability. 2.27 The PER workshop also discussed MTEF ceilings for fiscal year 2003/04. The government presentedan indicative MTEF to parliament on April 1, 2003, in accordance with the Budget Act o f 2001, which was very different from the October budget workshop ceilings. The changes were intended to accommodate additional defense expenditure o f USh57 billion. Giventhat this was adversely affecting PAF activities and some o f the other priority reform areas supported by the Poverty Reduction Support Credit (PRSC), the government had to ensure that the October ceilings were maintained. This ledto an MTEFthat hadto be revisedat the time o fthe PER workshop. 2.28 The revised sector ceilings were in line with the allocations presented at the budget consultative workshop in October 2002. However, the allocations were still provisional, pending input from parliament. The discussions at the meeting expressed concern about dismal allocation to agriculture and P M A as a whole, the persistently low pay reform allocations, no decline in allocations to defense, and marginal decline in the allocation to PA, among others. It was noted in the consultative workshop that health, education, and water and sanitation are areas needing additional resources to achieve some o f the critical MDGs. 2.29 As inthe previous year, this year's PER focused on broadening and deepeningthe involvement o f civil society and parliament. The capacity o f parliament to effectively participate in the PER process improved considerably over the year. Indeed, after the PER workshop, the Parliamentary Budget Committee met with MFPED officials for further detailed discussion o fcertain expenditureissues. 2.30 Based on inputs from parliament and discussions at the PER workshop, the MFPED revised the BFP's MTEF and presented it to the cabinet. Following further revisions, the MFPED submitted the fiscal year 2003/04 budget and the MTEF to parliament on June 12,2003. Medium-TermResourceEnvelope 2.31 As a part o f GDP, domestic resources have generally stagnated at about 10-12 percent. Projections o f revenueperformance do not show significant improvement over the medium term. However, unlike in previous years, revenue collection for 2002/03 improved significantly, to about 12.2 percent o f GDP, but still fell below projected amounts, Medium-term projections o f revenue are 12.8 percent o f GDP for 2003/04, 12.9 percent o f GDP for 2004/05, and 13 percent o f GDP for 2005/06. - 26 - 2.32 As table 2.3 indicates, the budget support i s supposed to be USh 843 billion, compared with a budget figure o f USh 811 billion, because o f larger than expected depreciation o f the Ugandan shilling against the currencies in which budget support i s denominated. The project grant i s supposed to increase from USh360 billion in2002/03 to USh 505 billion in 2003/04 (see table 2.3). Project loans are expected to remainmore or less stable, inline with the objective to reduce the debt burden. Table2.3ResourceEnvelope2003, 4 to2005/06(USh E !lion) 2002/03 2002/03 2003/04 2004/05 2005/06 zpproved outturn wojection projection projection budaet lomesticRevenue 1,432.6 1,418.3 1,690.9 1,856.2 2,052.8 URA 1,392.1 1,393.0 1,655.2 1,823.5 2,017.5 Nontax revenue 40.5 25.3 35.7 32.7 35.3 Loanrepayments 24.0 12.9 28.6 19.2 19.8 ;rants 861.2 851.5 1,042.7 1,063.1 1,109.2 Budget support 481.9 485.3 537.9 550.2 577.4 Project grants 379.3 366.2 504.8 512.9 531.8 .,oans 583.6 619.0 480.3 427.8 444.6 Budget support 330.7 358.0 199.9 180.8 188.5 Project loans 252.9 261.0 280.4 247.0 256.1 rota1resourceinflows minus Externaldebt repayment 165.1 181.9 181.1 188.7 195.6 Exceptionalfinancing -43.2 -43.2 -2 1.2 -4.8 -5 Domestic financing 9.7 28.7 -100.1 -85.1 -41.6 Arrears 50.0 56.4 45.0 41.0 41.0 rota1availablefor MTEF/budget 2,745.8 2,721.4 3,153.9 3,247.2 3,455.3 expenditures demo GDP 11,691.9 11,691.4 13,016.9 14,231.6 15,550.5 URArevenues as % ofGDP 11.9% 11.9% 12.7% 13.0% 13.2% 'AF 692.4 659.4 774.9 821.6 883.5 rransfersto Local Gover 670.7 660.8 742.2 775.6 820.6 mrce: MFPED 2003 2.33 Interms of revenue, the overall URA performance is very much in line with the target as a result o f reforms intax administration. Also, more than 70 percent o f revenue i s trade related. Consequently, changes in the exchange rate are a significant determinant o f revenue performance. This structure i s unlikely to change inthe near future. With the exception o f income tax, which performed above projections, other trade-related sources o f revenue generally performed marginally below projections. Income tax, VAT, and fees and licenses performed above target, enabling URA to meet its annual target despite a poor performance inimport duty. - 27 - TabIe2.4 URARevenuePerformance, July2002Afarch 2003(USh BiIIion) Tax head Budget Pro rata Outturn Performance(percent) Income tax 338 225 238 105.9 Import duty 721 542 537 99.0 Excise tax 376 263 266 101.9 VAT 240 168 178 106.4 Fees and licenses 40 28 28 101.2 Total 1,392 998 1,009 101.1 Source: MFPED 2003 2.34 Nontax collections are performing poorly, at only 25 percent o f projected collections. Loan repayments by some parastatals such as the Uganda Electricity Board and NPART were also below programmed levels. As a result o f these shortfalls, total domestic resource inflows for fiscal year 2002/03 are projected to underperform by UShs15 billion relative to the budget. However, this i s a significant improvement compared with a shortfall o f USh75 billion inthe previous financial year. 2.35 External financing as a proportion o f GDP was 5.7 percent, almost the same as the 5.8 percent in the previous year. The proportion o f external financing to GDP i s projected to increase slightly, to 6.2 percent in 2003/04, and then decrease thereafter to 5.8 percent in 2004/05 and 5.5 percent in 2005/06. However, domestic financing o f the budgetis projectedto be Ush 100billion in2003/04, compared with a saving this year. Medium Term ExpenditureFramework(MTEF) 2.36 Inaccordance withthe BudgetAct of2001, anindicative MTEFwas presentedby the government to parliament on April 1, 2003. The MTEF has been subsequently revised, by distributingadditional unallocatedresources to all sectors. The MTEF covers the period 2003/04 to 2005/06. Beginning this year, 2003/04, sector ceilings include donor-financed projects (and budget support), as agreed under the partnership principles. However, the quality o f information on donor project financing still remains imperfect. Table 2.5 shows the projected spending levels and sectoral trends o f the MTEF in the coming years, as well as the projected resources anticipated through donor project financing. - 28 - Table2.5Sector Ceilings,includingDonor Projects (USh Billion) 2001/02 2003/04 budget 2002/03 2002/03 budget 2004105 2005/06 releases approvedbudget budget outturn projections budgetprojections budgetprojections Total Total "otal Total Total Total 'otal Total iector rota1excl. 'otal excl. incl. Total excl. incl. xcl. incl. excl. incl. xcl. incl. ionor lonor donor donor donor Ionor donor donor donor 'onor donor roject Nroject project project project woject project project project lroject project Security 237.9 261.7 261.7 296.9 297.6 331.1 331.9 336.1 336.1 346.7 346.7 Roadsand works 156.7 150.5 316.6 155.0 223.2 168.9 344.5 172.9 396.8 176.1 408.4 Agriculture 42.4 46.8 133.1 49.1 114.8 48.6 103.3 53.9 100.2 62.5 110.4 Education 455.8 505.2 550.8 491.8 560.8 529.9 604.2 553.6 629.3 577.6 656.1 Health 162.9 196.0 337.9 189.9 307.1 218.8 375.1 229.9 358.9 247.9 381.7 Water and sanitation 49.0 48.7 126.1 55.2 90.3 61.9 114.2 64.5 108.2 67.8 114.0 Justice, law, and order 127.3 142.2 147.5 145.7 155.5 150.5 160.7 146.7 154.1 153.2 160.9 Accountability 21.5 26.6 27.3 26.1 30.6 28.9 32.2 29.5 32.4 30.2 33.2 Economic functionsand social services 122.4 149.9 337.5 151.9 382.3 172.0 404.0 167.4 376.4 180.2 398.3 PA 365.8 361.0 380.6 367.2 393.7 365.3 391.0 380.3 400.9 424.0 445.3 Interest payments 153.4 144.6 144.6 180.8 180.8 227.8 227.8 239.9 239.9 252.6 252.6 'otal 1,895.0 2,033.0 2,763.7 2,109.5 2,736.7 2,302.9 3,088.1 2,373.5 3,133.3 2,518.5 3,306.4 Source: MFPED2003 2.37 As in the previous year, the budget focus for 2003/04 is driven by three themes: continued attention to PEAP through the PAF, reducing the level o f the fiscal deficit, and promoting strategic exports. Just like last year's budget, this year's budget focuses on supporting PEAP implementation. Table 2.5 shows that all PAF areas have received increases inthe resources allocated to them. O f the overall increase o f USh214 billion, at least USh 80 billion have gone to the PAF. In fact, this puts the PAF share o f total Ugandan government expenditures at 37 percent, which i s higher than 36 percent for the previous year. Inabsolute terms, the PAF i s expected to move from USh692.4 billion to USh774.9 billionnext year. Thistrendo fincrease is expectedto continue inthe medium term. 2.38 However, the increases are not o f equal proportions. The share o f some sectors in the total budget has decreased, the increase in spendingnotwithstanding. In2002/03, the education sector used about 23.5 percent o f the total budget, but this year it has been allocated 23 percent. Similarly, the agriculture sector's share in the total budget has decreased from 2.35 percent inthe previous year to 2.1 percent in2003/04. The share for the health sector, however, has increased from 9.0 percent in 2002/03 to 9.5 percent in 2003/04. Thus, one may argue that some o f the priority sectors for achieving MDGs are not receiving their due importance. Sectors such as agriculture, health, and education do - 2 9 - not have adequate resources to implement their undertakings to meet the PEAP targets and MDGs. It i s imperative that more attention be accorded to the priorities o f the MDGs. 2.39 The largest increase in spending goes to defense, thus keeping the share o f defense expenditure in GDP above 2.5 percent. There i s needto ensure that this increase i s in line with the defense review outcome and priorities. At a time when all sectors are living within tight budget limits, there is need to get a better sense o f the composition o f the defense budget and its justification to maintain the credibility of the budget process. The sectoral distribution ofthe MTEF is described intable 2.6. Table2.6SectoraISharesofExpenditure over theMedium Term Sector 2001/02 2002/03 2002/03 2003/04 2004/05 2005/06 budget approved budget budget budget budget releases budget outturns projections projections projections Security 12.55 12.87 14.08 14.38 14.16 13.77 Roads and 8.27 7.40 7.34 7.28 6.99 works 7.35 Agriculture 2.24 2.30 2.33 2.11 2.27 2.48 Education 24.05 24.85 23.31 23.01 23.33 22.94 Health 8.60 9.64 9.00 9.50 9.68 9.84 Water and 2.59 2.40 2.69 2.72 2.73 sanitation 2.62 Justice, law, 6.72 6.99 6.54 6.18 6.08 and order 6.91 Accountability 1.13 1.31 1.24 1.26 1.24 1.20 Economic 6.46 7.37 7.43 7.00 7.10 functions and social services 7.20 PA 19.30 17.75 17.40 15.86 16.02 16.83 Interest 8.09 7.11 9.89 10.11 10.03 payments 8.57 Total 100.00 100.00 100.00 100.00 100.00 100.00 Source: MFPED2003 2.40 Funding of the strategic exports program in 2003/04 will amount to USh 40 billion. The strategic exports, inaddition to coffee, include fish, flowers, tobacco, cotton, tea, hides and skins, and maize. 2.41 Total Ugandan government spendinginthe revisedMTEF for fiscal year 2003/04 stands at USh2.305 billion. This is an increase o f USh215 billion, or 10.3 percent, over the outturn of the previous year. This spending i s consistent with the overall macroeconomic objective of maintaining macroeconomic stability. 2.42 Given the need to reduce the fiscal deficit and ensure macroeconomic stability, the government has highlightedthe need for slower growth o f government expenditures over the next three years-in the range of 5-7 percent per year, which is lower than the projected GDP growth. To achieve the projected government expenditure and GDP growth rates, there i s need to focus, more than ever before, on budget efficiency issues - 30 - and internal allocations. It i s also clear that fiscal deficit targets are becoming difficult to achieve because o f revenue shortfalls, on one hand, and expenditure increases, on the other, Both o f these constraints needto be addressedinthe mediumterm. 2.43 A review o f the projected spending levels and key sectoral trends of the MTEF reveals the following: 0 Health: In line with the government's plan to increase spending on health, the sector has been allocated an increase o f USh 20 billion in the 2003/04 budget, representing 9.5 percent o f the MTEF compared with 9.0 percent in 2002/03. The increase for the health sector i s for the improvement ofprimary health care. Although there i s an increase inthe MTEF to this sector, the weak performance o f the health indicators and the need for improvements invarious interventions in the health sector call for further resources for the health program in the coming year if MDGs are to be achieved. However, absorption o f more financial resources from various external sources, such as the global funds or donors with a special preference to fund the health sector, will depend on the sector ceiling, which should be increased with a view to providing sufficient fundingto the health sector. 0 Education: Innominal terms, the education sector receivedthe same amount as the health sector: an addition o f USh 20 billion in the 2003/04 MTEF for fundingthe Universal Primary Education(UPE) program. However, contrary to the expectations o f its SWG, education's share in the MTEF has come down from 23.5 percent in 2002/03 to 23 percent in2003/04. The increase in funding to education i s for recruitment of more primary school teachers, with a view to improve the PTR. The government still faces the challenge o f providing adequate financial resources to the education sector: although more inputs have been provided, including teachers, the education input indicators such as the PTR, the pupil-classroom ratio (PCR), and the pupil-textbook ratio have yet to improve significantly over their pre-UPE levels. The decreasing share of education in the MTEF i s not consistent with the UPE quantity and quality objectives. It i s important, therefore, that adequate resources be provided to the education sector to fully accommodate the rising enrollment inthe UPE. 0 Agriculture: The allocation to agriculture, including projects, decreased from USh 114.7 billion in 2002/03 to USh 103.3 billion in 2003/04. However, the agriculture sector's share in the MTEF decreased slightly, from 2.35 percent in the previous year to 2.11 percent in 2003/04. Nonetheless, the allocation is in line with investments in modernization of agriculture and production of strategic exports. Allocation to strategic exports in 2003/04 has come down to USh40 billion from the USh50 billion outturn o fthe previous year. Indeed, it i s extremely important for Uganda to focus on increasing exports. However, there i s lack o f transparency in funding selected interventions. It i s not clear in the budgethow the resources allocatedto strategic exports are to be used. Lack o f transparency in funding these activities, especially at a time when there are important priorities that do not have adequate resources, i s undermining the - 3 1 - budget process. Furthermore, these resources should not be used to provide unwanted and unjustified subsidies to private enterprises, as past evidence indicates. Inthe last two years, more than USh 100 billionhas beenallocated to strategic exports at a high opportunity cost in terms o f achieving other priority PEAPtargets. Related to this matter, activities financed from the budget toward facilitating private sector growth require careful consideration to ensure value for money and sustainable results. Subsidies to the private sector could be counterproductive, especially when they are selective. Such a policy will undoubtedly create an impression that there i s no level playing field for investors in the economy, thus discouraging some would-be investors from locating inUganda. 0 Justice, law, and order: The allocation to the justice, law, and order sector in the 2003/04 MTEF will increase to USh 150 billion, up from USh 143 billion the previous year. The increase is intendedto continue the implementation of the SWAP inthis sector. Furthermore, additional resources have beenallocated to the police to facilitate ongoing programs to protect the civilian population in Karamoja and northern Uganda. Because this sector has far-reaching implications for the achievement o f PEAP objectives, it should receive further attention and allocation o f resources. 0 PA: Inline with government plans to decrease the cost of PA, the allocation to the sector has decreased in both nominal terms and as a share o f total government expenditure. In nominal terms, the allocation to PA has remained stable at USh 365 billion in 2003/04, which i s the same amount o f resources allocated to the sector in the previous financial year. In line with the reform program, PA's share in the MTEF has come down, from 17.5 percent in 2002/03 to 15.9 percent in 2003/04. A further decrease in PA spending is desirable. The government i s inthe process of reviewingPA institutions with a view to reducingthis cost inthe mediumterm. 0 Pay reform: A total o f USh25 billion has beenprovided inthe MTEF for pay reform and pay awards. USh 10 billion i s intended to cover pay increases, and USh15 billion is for continuing the implementationofthe pay reform program. The amount provided for pay increases is to be distributed across all categories o f public service. This is in line with pay reform's objectives and principles o f ensuringthat the government recruits highly skilled people, pays them well, and retains them. However, the amount o f resources allocated for pay reform inthe mediumterm are inadequate to achieve pay reform objectives. Faster progress can be made only within the context o f comprehensive public sector reform that would generate the resources to pay much higher wages and salaries to highly skilled employees. 0Defense: As in the previous year's outturn, the increase in resources going to defense was the largest innominal terms this year. The allocation to security in the 2003/04 MTEF is USh 332 billion compared with USh 296 billion in the year before, an increase o f 13 percent. Defense's share in the MTEF increased - 32 - slightly, from 14.18 percent this year to 14.38 percent next year. Nonetheless, defense expenditure has become an area o f concern. Although it i s necessary to address northern Uganda security, defense expenditure needs to be closely monitored. The allocation to defense i s more than 2.5 percent o f GDP, which i s far above the government's past commitment to keep defense expenditure below 2 percent o f GDP. At a time when all sectors are living within tight budgets, there i s need to look into the composition of defense expenditure and its justification if credibility o f the budget process i s to be maintained. The government should hold discussions to address this matter inthe context of the defense review that i s currently under way. There should be greater accountability for defense expenditures to the responsible authorities. In the context o f competing priorities, there i s need to take a fresh look at the process o f setting expenditure ceilings for various sectors within the PEAP process and the LTEF. 2.44 Some development partners are willing to provide additional resources to fund social sectors, especially health and education. There is need for further dialogue between the government and its development partners to find a modality that supports additional funding to the social sectors. This will be a priority as Uganda discusses its trajectory inachieving the MDGs. ' 2.45 Allocation to district development, mainly through the LGDP, has received an increment o f USh 20 billion in the 2003/04 MTEF. Furthermore, National Agriculture Advisory Services (NAADS) has received an increment o f USh 4 billion. Thus, there will be increased spending at the LG level in fiscal year 2003/04. This demonstrates government commitment to the decentralization program. Also, because the resources directly allocated to LGs also contribute to sectoral programs, it i s important that these allocations be taken into account when discussing sectoral shares inthe future. This will be very important as more and more activities inthe mediumterm become decentralized. As figure 2.1 shows (see page 3.9, ifthe past LGDP sectoral allocation i s included, then the various sectors will receive additional amounts that will be directly channeled to LGs. 2.46 In line with the decentralization policy, there will be an increasing share of government spending at the LG level. It i s expected to increase from USh 657 billion in 2002/03 to USh 743 billion in 2003/04. Inthe last few years, this share has substantially increased overall, and by regions it has increased most in the north and east when compared on a per capita basis (MFPED 2003a). This also reflects well the poverty sensitivity o f the LG expenditures. However, this does not necessarily mean that they provide value for money. From a governance point o f view, there are concerns that these large allocations to the north and east may not be reaching the intended beneficiaries, an area intendedto be the focus o fthe MTEF inthe mediumterm. 2.47 Table 2.7 shows the distribution o f budget increases for 2003/04 by sectors. The highest increase is allocated to domestic interest payments to account for the under allocation this year. Education got the highestincrease, followed by health and economic functions and social services. Interms o f activities, inaddition to pay reform, which was - 33 - allocated USh 25 billion, primary teacher wages, health care, and LGDP all have recorded a USh20 billion increase. TabIe2.7Distribution ofBudgetIncreasesfor 2003/04, by Sectors Categories Allocation (USh billion) Security 5.0 Roads and works 14.0 Education 39.0 Health 30.0 Water 8.4 Justice, law, and order 7.2 Accountability 2.8 Economic functions and social services 29.0 Interest 47.8 Source: Budget SpeechJune 2003 (2003~). 2.48 Despite these challenges, the development partners supported the government's overall MTEF and the budget for fiscal year 2003/04, subject to the observations related to sectors that might need additional funding to carry out their mandates. At the annual PER meeting, the donors confirmed that the 2003/04 MTEF provides sufficient and appropriate basis for donor budget support. 2.49 Phase 1 o f the budgetkicked off in October 2002, whenthe MFPEDpreparedand communicated sector budget ceilings to SWGs and line ministries. Although the SWGs are now more institutionalized, technical capacity gaps in the preparation of BFPs remain. Furthermore, although SWGs are required to ensure that their BFPs are output oriented, many SWGs still find difficulties in doing so. Where outputs are a shared product between two or more sectors, there i s still hardly any collaboration between sectors inthe preparation of their BFPs. SWGs and LGs seem not to be adequately using the prepared guidelines that would assist them inpreparing output-oriented BFPs. These issues remain medium-termchallenges for the budgetprocess. - 34 - Figure 2.1SectoralIncreases for 2003/04 with LGDPAllocations Note: EF & SS: economic functions and social services; JLO:justice, law, and order Source; Ministry o f Finance,Planningand EconomicDevelopment Assessment of the BudgetProcess 2.50 There i s some improvement in the local government budget framework papers (LGBFPs). However, as in the previous year, synchronization o f LGBFPs with central government BFPs (PEAP objectives and development needs at the LG level) i s still a big challenge. Yet, as shown above, local governments are receiving more resources from the center, which should be managed better and reflect development priorities at the local level. Well-prepared, output-oriented LGBFPs would assist in the monitoring o f budget execution at the LG level. 2.51 The enactment of the Budget Act in 2001 marked the beginning of parliament's participation in the budget process. Initially, the Parliamentary Budget Committee, which i s responsible for taking the lead in the budget process, lacked the capacity to effectively participate. However, there has beenmarked improvement intheir capacity to effectively participate in budget discussions. This was evident in the PER meetings, duringwhich MFPED officials sat together with the Parliamentary Budget Committee to further discuss the allocations. In fact, there is an effort to streamline PER and parliamentary processes over the medium term so that they will mutually benefit and influence the budgetary choices in a coordinated manner. The budget that was finally read on June 12, 2003, reflected a position that parliament was comfortable with. This year, again, civil society played a role through various SWGs and also by making presentations at the PER meetings. Still, only a handful o f civil society organizations are able to participate: many are not able to attend because o f capacity constraints. It i s therefore necessary to develop the capacity o f civil society organizations and that o f parliament to effectively participate inthe budgetprocess. 2.52 Through participation, the discussions are beginning to go beyond allocation issues and are starting to touch on efficiency issues. Civil society representatives in - 35 - particular have been making a strong case about abuse o f public office, including misuse of government vehicles by public officers, corruption, conflict o f interest, and so forth. These issues are likely to determine the future course of dialogue between different interest groups in the budget process. Indeed, the payoff o f efficiency gains in terms o f improvement insocial service delivery could be huge. D. CHALLENGESINTHE BUDGET PROCESS 2.53 Inlight of this assessment of the budgetprocess, it is necessary to look critically at the challenges to address them in the medium term. First, we looked at technical capacity constraints in the preparation o f BFPs at both the sectoral and LG levels. The government should not be complacent about the minimal improvements in the BFPs, which largely remain inadequate because o f lack o f technical capacity and the short time normally given to SWGs to prepare them. SWGs and LGs need more time to prepare their BFPs. Training should be provided to the relevant public officers to enhance their technical capacity and enable them to play the role expected o f them in the budget process. 2.54 Funding of the social sectors i s still a big challenge, given the dilemma o f meeting the MDGs, on the one hand, and limiting expenditure to given budget ceilings, on the other. Although some donors are willing to provide additional funding to finance the social sectors to attain MDGs, the current sector ceilings for education and health make this difficult. Determination o f sector ceilings calls for participation o f all key stakeholders very early inthe budgetprocess cycle. 2.55 Integration of project aid in the MTEF i s still weak. The government has taken steps to ensure that donor project aid i s properly integrated into the MTEF. However, challenges still remain, the main ones o f which are (a) donor disbursements falling short o f commitments; (b) synchronization o f project aid with budget ceilings o f various SWGs, including local governments; (c) modalities o f integrating o f cross-cutting reforms and technical assistanceprograms; and (d) getting the cooperation o f every donor inthismatter. 2.56 Integration o f the wage bill into the MTEF i s also still weak, largely because o f the pending work on public service reform. Although consultation betweenthe MFPED and MOPS has improved, it will remain difficult to get meaningful medium-term projections o fthe wage bill untilthe public service reform exercise i s complete. 2.57 Because BFPs are hardly output oriented, monitoring o f outputs and outcomes has become a big challenge. Beginning at the stage o f BFP preparation, additional measures are needed to strengthenmonitoring and evaluation (M&E) to ensure an output or results orientation. The National IntegratedMonitoring and Evaluation System (NIMES), which i s being developed under the Office o f the Prime Minister, should take these issues into consideration. 2.58 The LGBFPs, which should be an interface between PEAP and LGs' priorities, have not featured very prominently in the preparation o f the BFPs, yet action should be - 36 - taken at the local level. Decentralization i s seen more as a fiscal matter than a planning tool. This deficiency has continuedto account for weak accountability, low transparency, and limited success interms o f service delivery or outputs. Consequently, because budget preparation-especially at the LG level-is not as it should be, the prepared BFPs fail to reflect LGs' real priorities. This area needs a lot o f attention to meet efficiency and equity goals in social service delivery. The government i s already beginningto address some o f these challenges. Inthe course o f the year, guidelines for the implementation o f the fiscal decentralization strategy were developed. They are being piloted in some selected LGs with a view to extendingthem to all districts, should they prove successful. 2.59 Lack of budget discipline in a few sectors still i s still one o f the budget process challenges. When public officers that incur budget overruns are not disciplined, others take it as normal and tend to do the same in subsequent years, thus renderingthe budget irrelevant. This matter i s being addressed through the Public Finance and Accountability Act. The government needs to demonstrate commitment to this act by punishing offending officers as the law stipulates. 2.60 With few exceptions, participation inthe budget process has yet to be geared to producing quality improvements in the BFPs, the MTEF, and the budget through incorporation of inputs from various stakeholders. For example, the extent to which stakeholders' inputs to the PER translate into changes to the MTEF i s not known- perhaps until budget day, when it i s too late to change anything. This i s an area for further work and support. 2.61 Civil society participation in the budget process i s not improving, mainly as a result o f capacity constraints. Support for representatives o f civil society to participate more effectively in the budget process i s very important. The gender budgeting workshops that were held for central and LG officials could be extendedto civil society to enhance the benefits interms o f BFP preparation. E. THEWAYFORWARD 2.62 The World Bank and other budget support donors will assist Uganda inimproving the budget process in the medium term. World Bank areas o f focus include improving the efficiency and equitable allocation of resources, as well as improved financial management at central and LG levels through PRSC and LGDP programs. 2.63 As pointed out in the PER report of the previous year, the PEAP-Uganda's broad development framework-is undergoing revision. The World Bank i s supporting the government inthe PEAP revision process, to provide support to PEAP target setting, sequencing, and costing actions. SWGs will use this information to realize budget linkages and PEAP targets. The MFPED i s also working on an LTEF that will assist sectors intheir longer-term planning. 2.64 The World Bank needs to support the whole budget process-starting with initial determination o f budget ceilings by the MFPED; BFPs preparation by SWGs and LGs; consultation through various forums, including the PER; and MTEF and budget - 37 - preparation-mainly because the Bank can assist various levels' technical capacity needs through funding. 2.65 Furthermore, other participants in the budget process, particularly civil society and parliament, are still facing technical capacity difficulties that the World Bank could support through funding. Although the participation o f parliamentarians has improved, more improvement i s desirable for effective participation inthe budgetprocess. Training o f some parliamentarians should improve their participation inthe budget cycle. 2.66 Recent reviews o f PAF show that its protection has been interpreted to mean protecting social service delivery sectors. It has become increasingly clear that activities of some nonprotected areas are critical to the realization o f PAF objectives. Hence, it is necessary to rework government expenditure priorities to ensure that PAF objectives are realizable. Some of these concerns should be addressed in the revision of the PEAP, which the World Bank and other donors are supporting. - 38 - 3. BUDGET EXECUTIONAND MONITORING, FISCALYEAR 2002/03 A. BACKGROUND 3.1 This chapter is about budget performance in terms o f revenue collection and expenditure releases and outturns. It looks briefly at revenue performance and then concentrates mainly on expenditure performance at the central government and LG levels. It examines issues related to budget execution, such as timely flow o f budgetary resources to cost centers, budget overruns and their management, effects o f budget cuts on affected sectors, payment o f arrears, inadequate counterpart funding o f donor- supported development projects, commitment control system (CCS) and its impact on sectoral performance, and budget execution challenges for the future. The chapter also touches on some key aspects o f financial management, such as capacity gaps, systems and processes o f expenditure management, accounting and evaluation, and the link between budget discipline and political involvement. These issues have a large bearing on the realization o f outputs and results o f PEAP, BFPs, or both. Expenditure tracking studies can provide valuable information on the proportion o f releases that get to the intendedbeneficiaries. The objective is to obtain information that could be usedto ease operational constraints and improve service delivery. 3.2 Once an agreed-upon budget i s drawn, it i s important to monitor not just revenue performance to see whether revenue i s in line with projected performance, but more crucially to monitor expenditure performance to see whether various sectors are receiving the resources they are programmed to receive at the time they should receive those resources. Furthermore, it i s important to monitor performance on the basis o f outputs and results outlined in the BFPs, because the BFPs are used as a basis o f arriving at the budget. However, as noted inchapter 2, BFPs o f most SWGs and LGs do not adequately outline results and outputs. This shortcoming has made M&E o f budget performance a rather difficult undertaking. 3.3 This chapter examines critical budget implementation issues that arose duringthe period under review. Major concerns include budget overruns and how they were managed within the MTEF, how payment o f arrears was handled, how adverse effects that arose from budget cuts were minimized (for example, irregular flow o f funds), and how the lessons learned could improve budget execution in the future. Once the government agreed to make budget adjustments to finance some activities that were not foreseen at the time o f preparing the budget, the problem became that o f managing the likely adverse effects on affected areas. The contingency budget was too small to absorb the increased demand for financial resources for the defense sector, leading to an across- the-board budget cut o f 23 percent for all unprotected areas. The key lesson to draw from - 39 - this chapter is howthe budget can accommodate unforeseen demands and how they could be better managed to minimize adverse effects on affected areas. B. REVENUE PERFORMANCE 3.4 During fiscal year 2002/03, domestic revenue was USh 1.418 billion, equivalent to 12.1 percent o f GDP. Domestic revenue collection for the financial year was 99 percent o f the approved budget estimates. The small revenue shortfall i s attributed to underperformance o f non-URA revenues, which are composed o f nontax revenues collected by line ministries and dividends from parastatals. Non-URA revenues underperfonned by USh 15 billion. 3.5 The best performing tax head in 2003/04 was income tax, which accounted for 25.3 percent o f total tax revenue, compared with 22.4 percent in the previous year. Compared with the previous year, income tax increased by 17.7 percent. Pay As You Earn (PAYE), withholding tax, and corporation tax performed well. PAYE and withholding tax registered considerable growth, with the latter registering a surplus o f USh5.14 billion. The improvement is partly attributedto improved tax administration. Import duty, which contributes about 10 percent o f revenue, registereda shortfall o f USh 2.7 billion due to external shocks and some inefficiency in the administration o f this tax. Excise duty performance was below budget estimates by about USh 27.3 billion, partly because o f lower than anticipated oil imports. 3.6 Donor project support (grants and loans) amounted to USh 627.2 billion comparedwith the budget figure o f USh632.2 billion, representing a slight improvement inperformance at the end of the fiscal year. Donor budget support overperformed by 4 percent (or about USh 40 billion) and amounted to USh 853.1 billion as a result o f depreciation o f the shilling. In dollar terms, however, by March 2003 donor budget support flows were below commitments by about 10 percent (around US$30 million) because o f differences created by the additional allocation to the defense budget. This i s an area where further progress has to be made. Part o f the answer lies in the government trying to limit such excesses in the future. If this is not checked, there is a risk that development partners may not have adequate confidence to provide resources to the budget and hence may have contrary incentives to extend assistance to the country through project aid. This i s also likely to undermine what the government o f Uganda wants interms o f consolidating project support into the MTEF. C. EXPENDITUREPERFORMANCE 3.7 As in the previous fiscal year, the 2003/04 budget experienced implementation difficulties mainly as a result of requests for supplementary expenditures, most notably by the defense and P A sectors, and higher than programmed domestic interest payments. During fiscal year 2002/03, budget release performance was generally not satisfactory. Performance by broad expenditure categories shows that wage and nonwage were above target, while development expenditure fell short by almost 20 percent in most sectors because o f additional resources provided to the defense budget. Contrary to the situation o f the previous year, central government votes performed better compared with LGvotes. - 40 - However, several central government votes performed below target to accommodate domestic revenueshortfall andsupplementariesfor other votes. 3.8 The year experienced expenditure pressures mainly from two sources: PA and defense. Inthe first half ofthe budget year ,defense expenditurewas above programmed levels by 18 percent, while PA expenditure was above target by 3 percent. A total of USh 22.3 billion was acquired through supplementaries, half of which were resource supplementaries, which had to be obtained through spending cuts in other votes, especially innon-PAF sectors. The other half of the budget year consistedof technical supplementaries, which financed rehabilitation of roads to the north, and a strategic exports initiative. Sectors mostly affected by the spending cuts included roads and works, health, education, andjustice, law, and order. Roads and works sector outturns were far below programmed levels-by about 20 percent by the third quarter-but expectedto meettarget bythe end ofthe fiscal year. 3.9 Individualsector performancevaried considerably, with some sectors consistently divergingfrom their pro rata budgets inthe first three quarters. Inparticular,the defense sector was 15percentabove budget, followedby water and sanitation andPA.The higher than programmeddefense spending, especially inthe first half of the financial year, and the lack of fiscal discipline in the PA sector rendered budget execution particularly difficult in2002103. 3.10 Wages performed at 97 percent of the program, largely because of the underperformanceof the wage bill in central ministries. Nonwage releases were above the programby 6 percent as a result of higher expenditure requirementsfor the Ministry of Defense and over expenditures in some line ministries mostly State House and the Office o fthe PrimeMinister. 3.11 Nonwage PAF releases performed at 97 percent of the program because of underperformance in some LG grants. Development releases performed at 93 percent, includingVAT releases. 3.12 Releasesto PAF developmentprojectswere at 99.5 percent of the program.Non- PAFdevelopment expenditureunderperformedby 15 percent,mainlybecauseofthe need to absorb budget cuts that were made to finance overexpenditure on nonwage items and interest payments. Overall releases, excluding interest payments, were in line with the programat 99.8 percent. Interest paymentswere abovethe programby 12percent. 3.13 A substantialandincreasingshare of governmentspendingis throughLGs, inline with the decentralizationpolicy. Transfers to LGs in fiscal year 2002/03 amountedto USh 654.0 billion, compared with USh601.89 billion in the previous fiscal year. Own generated revenues of most LGs amount to less than 15 percent of their expenditures. Thus, resource transfers from central governmentto LGs is a vital financing component of LGactivities. 3.14 The increase in transfers to LGs notwithstanding, it was less than the targeted USh670billionbecauseofunderperformanceof severaldevelopmentgrants attributedto -41 - noncompliance with the grant disbursement requirements, particularly delayed submission o f work plans by some LGs. MFPED data show that 20 o f 69 LGs did not submit their work plans or accountability for the first quarter. Also, many LGs had certain sectors that did not comply with the work plans for the first quarter. However, this could not always be verified with LGs, who claimed that they had complied with the requirements. LGDP funds did not flow as expected, which resultedinnoncompletion of many programmed activities. D. BUDGETOVERRUNSAND BUDGETCUTS 3.15 As in the previous fiscal year, the 2002/03 budget was not implemented as planned because o f requests for supplementaries, notably for the defense and P A sectors. The government decided to frontload a big portion of defense expenditure inthe first half o f the fiscal year to meet pressingsecurity needs. The problem became that o f managing the impact ofthese overruns onother sectors. 3.16 Budget overruns are usually financed by supplementaries. The MTEF allows sectors to receive supplementaries only when the equivalent amount i s cut from other sectors. Consequently, for the government to finance budgetoverruns for some sectors- notably defense, PA, and domestic interest payments-budgets o f other nonprotected areas were cut by 23 percent. At the sector level, the challenge became that o f managing with substantially reduced budgetary allocations. 3.17 The budget outturn ofthe first halfbest illustrates the budget execution challenges due to overrun: by that time, performance o f PAF development was at 93.5 percent but non-PAF development was at only 74.0 percent. However, many non-PAF projects were performing between 35 percent and 70 percent and were subject to release cuts o f about 50 percent. Similarly, release performance o f the unprotected nonwage recurrent budget was at 63 percent. 3.18 Broadly, the causes o f budget overruns were underprogramming and an inadequate contingency budgetthat could have beenusedto finance unforeseen demands, such as those o f defense. The next section looks briefly at the challenge o f underprogramming. E. UNDERPROGRAMMINGAND UNDERBUDGETING 3.19 Underprogramming refers to a situation in which resources provided for an activity are not adequate to finance that activity during a giventime. Usually, the activity i s a program that the sector must undertake to avoid jeopardizing attainment o f the broader development and poverty reduction objectives. This i s the case with budgetary provisions to finance domestic interest payments. These are mainly interest payments arising from Bank o f Uganda sale of treasury bills to absorb excess liquidity in the economy. The excess liquidity comes about from donor disbursements. The Bank o f Uganda has opted for a combination o f treasury bills and foreign exchange sales to absorb excess liquidity. This cost o f maintaining macroeconomic stability i s borne by the national budget interms o f domestic interest payments. - 42 - 3.20 Domestic interest payments for the first three quarters o f fiscal year 2002/03 amounted to USh 124.6 billion, representing an overperformance o f 14.9 percent. The annual outturn i s expected to be USh 180 billion against an allocation o f USh 144 billion. The domestic component o f interest payments is expected to overrun allocation by 60 percent. Expenditure of other sectors hadto be cut to finance these interest payments, as well as increased expenditure on defense and PA. For example, National Agriculture Advisory Services (NAADS) received only USh 13.48 billion this year, instead o f USh 17 billion expected from the budget. Early next year, this problem needs to be addressed through better prediction o f interest payments and sufficient resources allocated for defense and other programs, to ensure that it does not adversely affect other sector programs. 3.21 There were three major effects of budget overruns and associated budget cuts: inadequate counterpart funding, the irregular flow o f funds to sectors whose budgets had beencut, and payments o f arrears. The next sections examine these three broad issues in more detail. F. INADEQUATE COUNTERPART FUNDING 3.22 Insufficient counterpart funding partly explains the underperformance in donor project loans, which amounted to USh 101.8 billion in the first half-82 percent performance. Counterpart funds were inadequate mainly because o f budget cuts that were implemented to increase expenditure o f other sectors, notably defense, PA, and domestic interest payments. 3.23 It i s important to note that development expenditurecontributes to the realization o f some PEAP objectives. Therefore, protection o f some sectors under PAF may not guarantee realization o f PEAP objectives when other sectors are being starved o f funds. Thus, although the PAF wage and nonwage performance was satisfactory at 98 percent, inadequate funding o f other sectors couldjeopardize the realization o f PEAP objectives. 3.24 The shortage o f counterpart funds led to underperformance in non-PAF and nondefense development expenditure for the whole year, which i s likely to be only 75 percent o f the programmed level. Thus, budget cuts that were intendedto provide more resources to finance defense and other sectors that received supplementaries adversely affected development expenditure for most sectors through directly reducing the resources allocated for development expenditure and also making counterpart funding inadequate. 3.25 Inaddition, counterpart funds were inadequate inthe roads andworks sector. The issue o f inadequate counterpart funds, which Uganda had addressed quite well inthe past, i s now turning out to be a major concern, not only in the road sector, but for many other sectors as well. A special mention should be made o f the problems in the agriculture sector, where many seasonal activities under National Agriculture Advisory Services (NAADS) and National Agricultural Research Organization (NARO) could not be carried out this year because o f lack o f timely counterpart funds. As a result o f resource reallocations, many project counterpart funds were affected adversely, preventing - 43 - projects from continuing implementation. Thus, the intended objectives and results o f many projects could not be achieved. G. PROBLEMSOF IRREGULARFLOW RESOURCES OF 3.26 Budget cuts resulted in affected sectors receiving smaller budget allocations or receiving resources late. The point here i s about the irregular flow o f funds to some sectors, eventual receipt o f budgetedamounts notwithstanding. For example, by the first half o f the fiscal year, performance o f all sectors (with the exception o f defense, interest payments, and PA) was below 92 percent: 86 percent for roads and works; 87 percent for justice, law, and order; 87 percent for water and sanitation; and 89 percent for accountability. Performance was 89 percent for agriculture, 90 percent for health, 91 percent for education, and 90.5 percent for economic functions and social services. 3.27 Most o f the activities in the agricultural sector follow a seasonal pattern, which means untimely flow of funds will have a substantial impact on the agricultural activities to be carried out July-August and January-February in anticipation of rains in September-November and March-May, respectively. As figure 3.1 shows, NAADS releases for this year have beenvery uneven and not in accordance with need. A similar patternwas also observed for NARO activities. Figure3.1BudgetedandActual CashFlowsfor July2002/June2003 I 2 3 4 5 6 7 8 9 1 0 1 1 1 2 1 3 Months Source: MFPED 3.28 Delayed releases to the agriculture, primary education, drugs, and roads and works programs adversely affected activities o f those sectors. In the development budget, the situation was further exacerbated by irregular flow o f counterpart funds. Irregular flow o f funds mainly explains the underperformance o f development expenditure, which by March2003 was at 93.1 percent performance. H. PAYMENT OF ARREARS AND THEccs 3.29 In 1999/00,the CCS's quarterly cash limit was introduced for each vote, project, or item. It was first implemented for nonwage and development budgets at the central government level, and substantial progress was made in arresting arrears buildup. However, progress in extending the CCS to wage and all LG expenditures has been - 44 - delayed, and it i s clear that the gains can be immense if it i s implemented at those levels. This definitely could be combined with implementation of the IFMS, and it will help reduce arrears at all levels o f government. 3.30 Like other itemsthat sufferedlower than programmed releases because o f budget overruns in other sectors (mainly defense and PA), payment o f arrears lagged behindthe program this year. USh 50.9 billion had been initially budgeted to clear arrears, and additional USh4.3 billion supplementaries were obtained for arrears payments to Uganda Telecoms, among others, raising the budget for arrears to USh 55.1 billion. The approved USh4.3 billionbudget supplementary for arrears was intendedto pay for newly verified utility arrears. This was necessary to ensure uninterrupted service to the government by utility providers. However, only 74 percent o f the pro rata arrears budget was paid out during the first half o f the fiscal year. Wage and nonwage arrears underperformed while development arrears exceeded the pro rata budget. The arrears i s expected to overperfom annually by 11percent, which shouldbe avoided. 3.31 Arrears performance varied by sector. All sectors-with the exception o f health, economic functions and social services, and justice, law, and order-received supplementaries to settle utility arrears. Underperformance in arrears payments was registered in agriculture, education, PA, and justice, law, and order, where pro rata budget performance was less than 100percent. By the first half o fthe fiscal year, the pro rata arrears budget performance for education was only 36.4 percent, and that for justice, law, and order only 24.2 percent. 3.32 There are two concerns that arise from this matter o f arrears payments. First, the underprogramming o f resources called for processing and approval o f supplementaries. Second, the irregular and late releases o f arrears payments i s reflected in low pro rata releases o f arrears for some sectors. Such execution o f the budget adversely affects the realization o f the results and outputs originally envisaged inthe BFPs. I. MANAGEMENT OF BUDGETCUTS 3.33 As in previous years, the PAF mechanism was usedto protect some expenditure areas from budget cuts. Other non-PAF areas that are not protected from budget cuts had to bear the full effect o f the 23 percent budget cut. H o w did different sectors manage with significantly reduced resources? What lessons can be drawn from PAFto inform the budget execution process? Did protection o f PAF from budget cuts ensure total realization o fthe PAF objectives? 3.34 PAF came into being to channel HIPC savings and assistance from other donors toward pro-poor expenditures. Thus, PAF was established in 1998/99 as a "virtual poverty fund" to identify expenditures with a high poverty reduction impact. PAF funding is largely national, although some donors earmark budget support for PAF expenditures, and the budgeting o f PAF expenditures occurs through normal budgeting processes, However, PAF expenditures are given larger than normal spending increases, and they are protected from within-year spending cuts arising from domestic revenue shortfalls. However, protection of PAF from budget cuts does not ensure realization o f - 45 - PEAP or PAF objectives, because other areas whose budgets are cut have a large bearing on the realization o f those objectives. 3.35 Spending on programs covered by PAF roughly tripled between 1997/98 and 2000/01. Consequently, the proportion o f PAF to total public spending rose from 16 percent in 1997/98 to 35.2 percent in 2001/02. The share o f PAF programs in total expenditures is projectedto increase further inthe next three years, rising to 37.1 percent o f total expenditures in 2003/04. As a result, over the last four years, the Ugandan budget has become muchmore strongly poverty focused. 3.36 In terms of execution, by the end of the third quarter of fiscal year 2002/03, performance o f PAF nonwage releases was 97.0 percent and PAF wage was 98.2 percent. Performance o f PAF development was at 99.5 percent at the end o f March 2003. Clearly, these releases are generally within programmed levels. The slight underperfonnance i s explained by the slowness in reporting by SWGs and LGs to responsible authorities, which i s a requirement for subsequent release of funds. However, overall annual PAF performance i s expected to be slightly above 100 percent o f the programmed USh 692 bi11ion. 3.37 In a situation characterized by budget overruns for some sectors, protection of PAF expenditures from budget cuts has played a critical role in ensuring that resources flow to PAF activities. However, despite PAF's protection from budget cuts, achievement o f its objectives still remains difficult because o f budget cuts in non-PAF activities. In the absence o f budget overruns, protection o f PAF activities would not be necessary. 3.38 Spending units that received supplementaries did so at the expense o f others whose budgets were cut. Partly because o f inadequate contingency funds, supplementaries could not be fully accommodated within the budget. Consequently, some sectors received considerably fewer resources than programmed. Furthermore, the flow o f resources to most sectors was irregular. H o w did sectors whose budgetswere cut manage? Did they cut their planned activities to absorb the budget cuts, or were they innovative in some way and able to work with reduced activities to fund all planned activities? 3.39 Affected sectors handledthe issue o f budget cuts differently. Some few affected sectors absorbed budget cuts mainly through prioritization o f their activities and concentrated their spending on selected priorities. This i s what happened in non-PAF education areas: after the budget cut o f 23 percent, spendingon some activities was either not cut or marginally cut, while that on some other activities o f lesser priority inthe same category was either cut by a high percentage (say 70 percent) or activities were simply abandoned because o f lack o f funding. However, in most sectors (notably roads and works, agriculture), a lot o f activities whose implementation had started, ground to a halt. Thus, ifbudgetcuts are not properly managed, they can adversely affect the attainment of development targets. - 46 - J. BUDGETMONITORING REPORTINGAND 3.40 As highlighted in last year's PER, budget monitoring and reporting have been quite weak in Uganda. However, as a result of various efforts, progress has been made this year. The recently completedHIPC tracking report action planupdate (March2003) provides a comprehensivelook at how various monitoringand reportingchallenges have been addressed in the last year (see annex 5). However, as chapter 5 of this report demonstrates, the budget reporting and monitoring at the LG level is very weak and uncoordinated. Current monitoring and reporting practices do not optimally use the limitedcapacity at the LGlevel. 3.41 At the central government level budget support (PRSC) monitoring is done through the preparation of a quarterly budget performance report, typically prepared within 30 days of the end of a quarter. In addition, the half-yearly report is formally published and distributed. A final annual report i s usually expected to be produced within 90 days of the end of the fiscal year. All these improvements have helped to improvemonitoringofbudgetexecution. 3.42 The improvements in OOB and ROM have also facilitated better monitoring of outputs and outcomes. However, there are still many sectors where such informationi s hardto come by. The MFPEDis committedto improvingOOB, andMOPS has provided additional support for ROM to SWGs. In addition, the MFPED's Poverty Monitoring and Evaluation Strategy (PMES) and the Ugandan government's commitment to undertake NIMES are addressing the systemwide monitoring problems and will also improve budgetmonitoringandreporting. However, it i s clear that Ugandahas yet to see the full benefitsofthese improvementsinnational monitoringsystems. 3.43 These budget performancereports have helpedto identify and reduce the budget discrepancy betweenallocation and outturns. Inthe last four years, budget discrepancy has declined to less than 10 percent from double figures in the early 1990s. Measuring variation using the discrepancy index (defined as the absolute percentage deviation betweenbudgetand outturns andthentaking a weightedaverageofthe deviations) makes it clear that substantialprogresshas beenmadeover the last four years (seetable 3.1). Table3.1Average BudgetDeviations 1998199 1999/00 2000/01 2001/02 2002/03 Discrepancy index 9.8 4.9 7.6 5.8 5.5 Source: Estimatesbasedon various budget documents and MFPED 2003 a. 3.44 As a result, the Ugandan budget is gradually becoming more predictable in execution, providing a basis for donor finance to shift from project to budget support modalities. It is, however, clear that until the challenges of budget monitoring and reporting are satisfactorily addressed across sectors, donor needs of adequate fiduciary assurance cannot be met. Therefore, if they must rely on the government's current monitoringandreportingsystem, many donors will be reluctant to move fully toward the budgetsupport modality. - 47 - K. CHALLENGES FOR THE FUTURE 3.45 Budget execution has two sides: the revenue side and the expenditure side. On the revenue side, domestic revenues as a proportion of GDP have generally stagnated at about 12 percent. Uganda faces the challenge o f increasing this proportion to, say, 20 percent in the medium term. Increase o f domestic revenues would be a good strategy for reducing the country's fiscal deficit, and this seems to be high on Uganda's economic management agenda. 3.46 The biggest expenditure challenge is getting expenditure priorities right to ensure that government spending i s consistent with the attainment o f PEAP targets and MDGs. The government may need to review several points, including the shares of PA and defense in the total budget. There i s need to review the number o f P A institutions, including the LGs. It i s not enough to put a lid on LGs' expansion. It i s necessary to review &e number o f LGs with a view to attainment o f cost-effectiveness in PA. Are there too many districts, and are some too small and unviable administrative units?I s there need o f a review o f the institutions o f political governance to do away with duplication? Could some institutions be merged? Can Uganda afford the current size o f parliament, the number o f cabinet ministers, and the number o f political advisers, district resident commissioners, and so on? 3.47 The ongoing public service reform could address some o f these challenges, but not all o f them, because it i s not comprehensive enough. A comprehensive civil service reform that looks at the entire structure o f the public service, with special focus on PA, i s necessary. Some o f these issues have already beencaptured through a study the president commissioned on the possibility o f more efficient P A spending. The recommendations of that study could form the starting point for a comprehensive civil service reform. Donor support would be necessary. 3.48 The problems arising from budget overruns and underprogramming translated into irregular flow o f resources to sectors, LGs, and counterpart funds for donor- supported development projects. Lack o f counterpart funds also led to withholding o f donor assistance for project activities. This partly explains the underperformance o f donor-supported projects. This, too, has a bearing on the achievements o f the poverty reduction and economic development objectives o fthe country. 3.49 In recent years, budget execution challenges at the LG level have gained increasing attention, especially with a larger share o f resources beingtargeted at LGs. In fact, this i s clearly both a budget implementation and budget monitoring issue. The irregular flow o f resources cited by many LGs seems to emanate from the lack o f provision o f timely accountability for funds already provided. A closer review also shows that in addition to lack o f capacity, there i s also a deficiency o f clarity about roles and responsibilities, which need addressing. Whatever the reasons may be, it i s clear that delays infunding have substantially disrupted service delivery activities at the LG level. - 4 8 - L. CONCLUSION 3.50 The above review clearly shows that on the revenue side, Uganda still faces the challenge o f collecting a higher proportion of GDP than the current 12.1 percent. On the expenditure side, it is important to improve the predictability o f sectors. That there will always be new financial demands arising inthe process o f budgetexecution, thus the size o f the contingency budget should be adequateto cover them. Because o f budget overruns and associated budget cuts for many sectors, budget implementationwas characterized by an irregular flow o f funds to some sectors, failure to pay arrears timely, insufficient government counterpart funds to finance donor-supported projects, and underprogramming that led to insufficient resources for domestic interest payments and arrears payments. The trend o f providing supplementaries to defense and P A i s a matter o f serious concern because it leads to budget cuts in other activities that are critical to the realization o f poverty reduction and development objectives. The implications for the overall credibility o f the budget process and execution should be given due importance. Many o f these problems are rooted inpolitical interference inbudgeting and planning and execution-issues that are regarded as technical considerations. This situation calls for closer collaboration and dialogue among the relevant stakeholders to reduce the frequency o f such disruptions. 3.51 There is also the challenge o f ensuring the timely flow of funds to the various sectors. The tendency to frontload expenditure o f some sectors while others are starved o f funds should be avoided. There may be a needto review the cash budget management system so that it does not become detrimentalto budgetexecution. 3.52 Budget overruns by some sectors are the root cause o f budgetary problems experienced by other sectors starved o f financial resources. Budget overruns distort the budget and make it irrelevant. For the budget to remain credible, overruns should be avoided. They come about mainlyas a result o f underprogramming. Sufficient resources should be provided ina contingency budgetto minimize adverse effects o f budget cuts on nonprotected areas. The problem o f budget overruns needs immediate attention if budget preparation and budget execution are to remain credible. 3.53 The implementation o f the Public Finance and Accountability Act (2003) will be an important instrument in arresting the past trends. It i s intended to limit the incidence o f supplementaries by requiring parliamentary approval before the release o f any supplementary funds. This should help contain overexpenditures o f some sectors and lead to a balanced execution o f the budget. However, with the objective o f deficit reduction and an overall underfunding o f most activities, it will be very difficult to manage the culture o f supplementaries. 3.54 Full implementation o f the CCS i s important to cover all central government and LG expenditures. The CCS currently covers only central government nonwage and development budgets. Arrears should be minimized on all accounts by further extension o f the CCS. Payment o f arrears should also be a high-priority matter. Arrears reflect services renderedand not paid for. Failure to clear arrears adversely affects performance o f the various sectors, especially inthe case o f utilities, as was experienced this year. - 49 - 3.55 With the government committed to reducingthe fiscal deficit, there are likely to be even greater pressures on budget execution because many more sectors and programs are likely to have underbudgetingproblems. Hence, if adequate checks and balances are not designed and implemented, it is possible that some of the problems discussed inthis report may intensify in the medium term. Thus, budget indiscipline may worsen in the future because many activities will not be fully funded, which will make demands for supplementariesunavoidable. - 50 - 4. EFFICIENCY OF PUBLICEXPENDITURES 4.1 The issue o f efficiency has gained attention because o f the dilemma faced by governments in distributing scarce resources among competing needs. This i s compounded by the infeasibility o f sustained increases in the overall resource envelope and therefore the inability to adequately meet individual sector funding expectations. Increasing the efficiency and equity o f public expenditures i s essential to improving public resource allocation and raising the productivity and cost-effectiveness o f public investments. 4.2 Uganda, like many countries, is directing its resources toward achieving national development goals (PEAP goals and MDGs), but it i s not clear whether these goals can be met within the current levels o f resources. It is also unlikelythat a large scaling up o f these resources will occur in the short or medium term because o f the limited tax base and Uganda's already high aid dependence. Therefore, achieving outcomes that are consistent with the overall poverty reduction objectives will mean an increased need for the efficient use of public expendituresacross sectors. One way to maximize the impact o f overall budgetary expenditures i s sector-by-sector improvement o f the efficiency o f expenditures, especially given that there is a limited amount o f resources available to sectors through the budget. 4.3 To improve efficiency, the government has at least two options, which have been advocated widely in economic literature. First, changing the allocation mix o f inputs or reallocating resources to their most cost-effective interventions could dramatically improve outcomes. This alternative i s feasible in the short run. Second, the government has the option o f reducing the unit costs o f outputs. The latter option may be slightly more ambitious, involving the long run, because reducing the cost o f outputs may consist o f implementing wide-ranging institutional reforms to improve such variables as the overall level o f bureaucratic quality and corruption, with the hope that this in turn will improve the efficiency o f public expenditures(Jayasuriya and Wodon 2002a). 4.4 A number o f explanations have been given for why public expenditures may not result in the desired quantity or quality o f public services. First, even targeted public expenditures sometimes disproportionately find their way to nonpoor people and private goods. Second, the transfer o f funds from the central (frontline) provider may often be subject to leakage, misdirecting resources away from the intended service provider (for example, a school or a health unit) and ultimately from the intended beneficiaries. Third, even if funds reach the frontline producer, there may be low efficacy inthe provision o f goods and services. Last, even if the public services were provided and delivered, households may be lacking in demand (Reinikka, Svensson, and Lindelow 2002). A mismatch between supply o f services and lack o f demand, and the resulting creation o f excess capacity, will inturn result inlow efficiency o f expenditures. This mismatch may - 51 - be the result o f a high degree o f centralization in both the policymaking and fiscal systems, where the supply o f services i s defined at the central level and the central government is still largely responsible for implementing the national development agenda. 4.5 This chapter tracks the progress o f four key sectors in the Ugandan economy- water and sanitation, health, education, and justice, law, and order-in institutionalizing allocative and operational efficiency issues. The education and health sectors have a longer history o f expenditure tracking and linking sector inputs to outputs, and in the water and sanitation and justice, law, and order sectors, the discussions are relatively new, Yet all four sectors have made remarkable progress toward allocation and effectiveness o f spendingat both the national and local levels. A. WATER AND SANITATION SECTOR Background 4.6 Inboth the Poverty Status Report (PSR) 2003 and Uganda Participating Poverty Assessment Project (UPPAP) 2002 reports, water and sanitation are noted with concern as essential for improving the quality o f life o f the poor, especially given recent reports about lack o f improvement in several important mortality indicators-infant, child, and maternal mortality. Statistics from the 2000/01 Uganda Demographic and Health Survey (UDHS) showed that infant and child mortality rates have worsened, while maternal mortality rates have stagnated over a number o f years. Consequently, the government did not meet the 2002 infant mortality target and i s unlikely to achieve the 2005 target if trends persist. Past mortality trends are highlighted in table 4.1. The progress o f the determinant variables for mortality indicators, including water and sanitation variables, i s also shown inthe table. 4.7 Access to safe water and sanitary conditions i s a proximate determinant o f infant mortality and as such the water and sanitation sector i s one o f three sectors (including health and education) that are key to improving the performance o f mortality indicators. Increasing poor communities' access to safe water and their sanitary environment has positive linkages to the reduction o f highinfant and maternal mortality rates. 4.8 Despite the deteriorating infant mortality situation between 1995 and 2000, table 4.1 shows access to safe water and sanitation improving considerably during the same period. Given that access to safe water and sanitation are determinants o f the infant mortality indicators, it appears that the gains in this sector over 1995-2000, however considerable, were not sufficient. Therefore, a greater effort i s needed in the water and sanitation sector (as well as in health and education) to increase rural access even more dramatically and have a positive impact on infant mortality. - 52 - Table4.1 Trendsin SelectedMortality Indicators Indicator 1995 2000 Trend Mortalitv Infan; mortality (per 1,000 live births) 81.3 88.4 Deteriorated Neonatal mortality (per 1,000 live births) 27.0 33.2 Deteriorated Postneonatal(per 1,000 live births) 54.3 55.2 Deteriorated Maternalmortality (per 100,000 live births) 527 505 Improved Proximate determinants of infant mortality Access to safe sanitary disposal(pit latrineor flush 79.9 82.3 Improved toilet) 39.4 53.8 Improved Access to safe water inrural areas 42.9 31.4 Improved Share ofteenage women who are pregnant or mothers 6.9 6.9 Constant Total fertility rate (all) 29.3 29.2 Constant Medianbirth interval 14.8 22.8 Improved Contraceptiveprevalence(any method) 47.4 36.7 Deteriorated Child immunizationcoverage (all) 53.7 41.7 Deteriorated Tetanustoxoid for pregnantwomen (fully covered) 35.4 36.6 Constant Deliveriesat a healthfacility 37.8 39.0 Constant Deliveriesassistedby skilled healthworker 30.6 21.9 Improved Women with no education 56.0 59.8 Improved Women with primary education 13.5 18.4 Improved Women with secondaryeducation or above 50.2 35.2 Improved Headcountpoverty Source: UDHS2000101. 4.9 Usingdata on toilet facilities derived from two surveys in 1995 and 2000/01 as a proxy for sanitary conditions, an MFPED study on infant mortality (August 2002) shows that infant mortality i s generally higher for households without toilet facilities compared with those that have either a flush toilet or a pit latrine. Similarly, for households with access to safe water, the infant mortality estimates are halfo f those o f households without access to safe waterq4(See figure 4.1, adapted from the same study.) Figure 4.1 TrendsinInfant Mortality, by SanitationFacility and WaterSource Infant mortality by safe and unsafe water Infant mortality by toilet facility source (2000) IO0 I50 80 IO0 60 1995 40 50 2000 20 I 0 0 1 2 Safe water source Unsafe water source Pit latrinelflusb Nofacility Source: MFPED2002a 4.10 There has been increased government investment in the water and sanitation sector over the last three years. The sector's share of the budget continued to rise in 4. Safe water sources are defined as pipedwater, protectedwells, boreholes, gravity flow schemes, and bottledwater; unsafe water sources includeopen wells, surface water, rain water, and tank truck. - 53 - fiscal year 2001/02 to 2.59 percent o f total expenditures, up from 1.44 percent o f total expenditures in fiscal year 1999100. Despite this increase, budget expenditure outturns declined for the first time intwo years, from the sector's 100 percent rate to 91 percent, and this was attributed to nonwage releases at the center and underuseo f resources by the Water for Production Project. 4.11 The emergingbudgetary picture has attracted stakeholders' attentionto a disparity between increased resource allocation to the water and sanitation sector and improved outputs and outcomes. In a background note in May 2002, the MFPED argued that a threefold increase in the input-output ratio in the sector between 1997/98 and 2000/01 reflected a substantial amount o f inefficiency within the ~ e c t o r .It~ was questioned whether increased inputs to the sector had been successfully transformed into higher outputs. The input-output disparity has been attributed to the sector's tendency to allocate a larger proportion o f additional resources to capacity building and maintenance activities at the district level, which has not necessarily translated into increased water points in the short run. (Refer to PER 2002 for a detailed discussion o f this issue.), Nevertheless, this highlights a need to analyze the impact o f additional resources in the sector, as well as its level o f efficiency across subsectors, so that future resources will indeedbe able to maximize outputs. 4.12 The current state o f the water and sanitation sector's coverage is illustrated in Table 4.2. The statistics, taken from the results o f the 2001 National Service Delivery Survey, are quite positive, indicating that the sector i s within target o f its coverage outcomes. Table 4.2 Water CoveragePatterns (2002) Source: Householdand ServiceDelivery Survey2001. 4.13 Furthermore, interms of rural coverage, there are indications that half o f the rural population have a supply o f clean water, representinga 3 percent increase since 2000. 5. The MFPEDcomparisono f the total water and sanitation budget (whichincludes ruralwater, urban water and sanitation, water for production,and water resourcesmanagement)to ruralsector outputsonly was criticizedby the Directorateo f Water Department(DWD) study, "Water Sector Performance" (2002), althoughtheir own comparisonoftotal inputsand outputs to ruralwater confirmed that the input-output ratiohad infact tripled. See also PER(2002) - 54 - Table4.3Rural Clean WaterSupplyby Source, 1999/2001 (Number ofSourcesofSupply) Source 1999 2000 2001 Spring 17,282 17,843 19,029 Borehole 15,374 16,520 17,915 Shallow well 2,774 3,417 4,734 Gravity flow system 82 125 134 Taps 2,655 3,866 4,058 Rural population, % coverage 46.6 47.5 50.2 Source: Uganda Bureau of Statistics 2002. 4.14 Inshort, the water and sanitation sector could be described as "having breadthbut lacking depth." Despite the apparent acceleration in coverage, reports indicate that the water quality does not measure up and resource allocation i s not efficiently targeted. There is a need for better analytical work to inform the subsequent choices o f the sector and ensure that resources are targeted where they will generate the most value. 4.15 Insanitation, the situation is not particularly encouraging. Uganda has one ofthe lowest access and coverage rates for sanitation inthe world, and the MOHEnvironmental Division estimates that 80 percent o f the disease burdeni s associated with poor sanitation and hygiene. National household latrine coverage rates are relatively low, especially in rural and remote areas, and piped sewerage services are accessible only to an estimated 5 percent o f the population. A greater priority needs to be attached to the sanitation problem, especially given the importance o f hygiene for overall health and mortality indicators. Without adequate attention to the sanitation issue, the health impact o f investment inthe water sector will remain minimal-in short, resources will be wasted. The 2003 Poverty Reduction Strategy Paper notes that one o f the main challenges in addressing the sanitation problem i s that there i s no responsible lead agency and, according to government policy, sanitation remains an individual household responsibility. Allocative Efficiency 4.16 Therehas beenlittle progress on the issue o f allocative efficiency inthe water and sanitation sector, with intrasectoral distribution o f resources still disproportionately favoring the urban areas. From the resource allocation perspective, the central issue in the sector is low coverage o f rural water and sanitation and services for the urban poor. As inthe situation in2001/02, 50 percent o f the total water sector budget is still spent at the center, despite the fact that 90 percent o f the beneficiaries live in rural areas. Furthermore, a review by donors in the SWG of the fiscal year 2002/03 BFP reports inconsistency in the allocation o f the budget to items that do not fully justify or show strong alignment with the PEAP objectives o fthe sector. 4.17 The water and sanitation sector i s still skewed toward activities that the MWLE Directorate o f Water Department (DWD) implements directly, and the budget allocation process appears to be driven by a perceived institutional mandate rather than by the incidence o f sector needs. Although this situation could be explained from the perspective that the sector i s still in a state o f transition toward greater decentralization, - 55 - increased private sector participation, and the addition o f new mandates such as water for production, the current share of more than 50 percent o f activities implemented by the MWLE-DWD is worrisome, given the size of the population that these center agencies directly serve (see Table 4.4). Table4.4 WaterandSanitationSectorResourceAllocation (2002) Subsector Total % of total % of the Coverage (USh millions) population Rural 38,073 32% 85% 55% Small towns 28,365 24% 3yo 70% Urban 36,840 31% 12% 85% Water for production 5,000 4% Water Resource Management 6,034 5% RM Human Resource Development 4,498 4% and policy reform Total 118,810 100% Source; Uganda Bureau o f Statistics 2002. 4.18 Small towns currently take up 24 percent o f the water and sanitation sector's total budget and serve approximately 3 percent o f the total population. Water coverage in small towns reached 70 percent in 2002, representing an increase o f 18 percent since 2000. Nonetheless, there i s a need within the sector to explain why about 60 percent o f the budget is spent at the center for small towns, water for production, water resources management, and general sector support. This presents a distorted picture in resource allocation and indicates a lack o f alignment with PEAP objectives. 4.19 The water supply and sanitation project for small towns began in 1995 with the aim o f (a) improving health through improved personal hygiene, excreta disposal, and waste management; (b) reducing time and effort saved in fetching water and the daily routine o fpersonal hygienethrough a more convenient location o f sanitation facilities; (c) improving water supply to an estimated 160,000 people; (d) improving sustainability o f services beyond the duration o f the project; (e) reorientation and institutional strengthening o f the DWD; (f) contributing to poverty alleviation; and (g) reducing the water collection burden o f women, enabling them to focus on other household and income generation activities. Small towns are now a large subsector in water and sanitation, taking up 24 percent o ftotal resources in2002. 4.20 Questions have been raised about the allocative efficiency o f small towns' operations, especially given the large amount o f resources used by the subsector relative to its outputs (see table 4.4). In 2003, operations in all small towns except Wobulenzi were being run by private operators through management contracts signed between the private operators and each town's Urban Water Supply Authority. The majority o f small towns have experienced growth in connections since May 2002, albeit at a rate far less than was projected, and have either surpassed or recently reached the break-even point and sustainable operations. To break even intheir operations, the privately run facilities insmall towns have had to be subsidized by the sector. The question arises whether the sector can continue allocating a large share o f its resources to the smallest percentage o f the population. - 56 - 4.21 At the district level, there is aneedfor better poverty targeting. Resultsvary from district to district, butthe war-torn northern districts are worse off. As a result o fthe war, resources allocated are largely not reaching the districts-or in cases where they do arrive, they are not creating much value. This i s especially true inthe case o f the water for production subsector. This subsector comprises such activities as small-scale irrigation and water for livestock. The conflict inthe northern region, which has resulted inlarge-scale displacement ofthe population, has also affected the implementationofthe water for production program inthe area. Operational Efficiency 4.22 Operational efficiency in the water and sanitation sector has been difficult to evaluate inthe past, mostly because o f the lack o f thorough analytical work inthe form o f efficiency studies. Needs remains to better systematize the process, improve the methodologies used and the overall quality o f measuring efficiency, and integrate the findings ofthe analyses into sector policy. 4.23 A culture of expendituretracking and efficiency analysis is slowly developing in the water and sanitation sector. Although the sector has frequently carried out financial audits in the past, these were not always concerned with value-for-money issues until recently. The MFPED, in a note entitled "Is the Water Sector Delivering?" (2002a), questioned whether the sector's increased resource allocations over the previous four years had been matched with value for money. The note observed that although resources flowing to the sector had increased threefold between 1997/98 and 2000/0 1, average annual outputs had declined over the same period compared with the previous six fiscal years. In response to the note, the DWD undertook a study o f its own, using a slightly different methodology, which verified the tripling o f input-output ratios in the rural subsector. Consequently, a technical audit-value-for-money study was launched to identify the sector's efficiency problems. The results o f this study have already been circulated, and although there are questions about its overall quality, its mere completion i s a first step in bringing the efficiency discussion to the forefront within the sector. Other studies are planned that will increase the focus on efficiency, notably an analysis o f performance measurement in the water and sanitation sector, was discussed at the water and sanitation sector review inSeptember 2003. 4.24 The technical audit of the rural water and sanitation conditional grant, also referred to as a value-for-money study, was carried out during fiscal year 2001/02. It was evident inthe final report that accurate methodologies for measuringvalue for money had not been used. Eight separate audits were carried out in each o f the Technical Support Unit (TSU) regions, and the final report of the technical audit was a synthesis of all of them. Hence, there were significant discrepancies in the way individual studies were conducted and inthe findings o f the final report. The process o f auditing also seemed to have been plagued by a lack o f credible data at the district level, and, consequently, the results of the studies were quite vague and mostly poor from a quantitative (and even qualitative) standpoint. - 5 7 - 4.25 Nonetheless, there are indications from the final report o f existing loopholes in the districts analyzed in terms o f adherence to procurement and financial management standards. District procurement processes are characterized by inadequate tender documentation, inappropriate technical evaluation, and political interference in Tender Board decision making. Inaddition, there i s inadequate internal and external auditing o f financial accounts. Water and sanitation committees are invariably either dysfunctional or nonexistent, which results in communities failing to meet their responsibilities toward operation and maintenance. 4.26 Furthermore, the weak capacity o f the private sector was a recurring theme across most districts. In many instances, the private sector firms engaged to construct new facilities are technically incompetent and do not comply with guidelines, specifications, and contract agreements, resulting in substandard outputs. For example, the generally poor quality o f school sanitation facilities was directly linked to the hiring o f nonqualified contractors and to insufficient contract supervision, which have also led to poor quality material and workmanship. 4.27 Last, there i s a top-down approach to sector planning, budgeting, and management, with a lack of coordination between districts and subcounties and inadequate involvement o f NGOs, instead o f a preferredbottom-up approach that would involve communities in quality assurance. The lack o f a bottom-up approach to sector planning, budgeting, and management was also flagged at the PER workshop in May 2003 as a constraint to poverty reduction, because the poor have little or no power to influence the provision o f water and sanitation services. In addition, a top-down approach will prevent the achievement o f a demand-responsive approach and the associated efficiency gains. Therefore, a shift inapproach i s requiredwithin the sector to put the needs of the poor at the center and ensure that sector policies are designed to respondto these needs. 4.28 The DWD has a responsibility to address these issues, first, by increasing support to districts through the TSUs, to which a significant amount o f funding has been committed. It i s not yet clear how well the TSUs have beenable to fulfill their functions. Second, the DWD can increase the effectiveness o f the M&E structures in place, which should enable a continuous assessment o f district performance. Ad hoc M&E activities are currently taking place and needto become better systematized. B. HEALTH SECTOR Background 4.29 The biggest challenges confronting the health sector are the worrying trends in infant, child, and maternal mortality rates and the total fertility rate, which i s the third highest in the world. The poor performance of the mortality indicators i s not only the health sector's responsibility, given the influence o f other factors-such as household income, female education, access to clean water, security, gender disparity, cultural practices, and nutrition-on health outcomes. A wealth index from the 2001 UDHS dataset shows that infant mortality i s almost 80 percent higher among the poorest 20 - 58 - percent compared with the richest 20 percent o f the sampled population.6 Infant mortality i s also shown to have risen significantly in the northern region o f the country, consistent with an increase inpoverty. 4.30 In addition, the HIV/AIDS prevalence rate (measured from prenatal sentinel surveillance sites), which declined consistently from 6.9 percent in fiscal year 1999/00 to 6.1 percent at the end o f 2001, appears to have stagnated for the first time ina decade. It should be noted that Uganda has made substantive progress over the past 17 years in bringingthe pandemic under control, achieving reductions inHIV/AIDSprevalence from 18.5 percent in 1992 to 6.1 percent in 2002. Despite this progress, the current tapering off o f the decline i s cause for concern and there i s a needto reevaluate the situation. TabIe4.5Summary ofHeaIthIndicators2000/01to2002/03 Indicator Baseline value 2000/01 2001/02 2002/03 forecast OPD use .40 .43 .60 .63 (total government and PNFP) DPT3 coverage 41% 48% 63% 65% Deliveries in health facilities 25.2% 22.6% 19% 25% (government and PNFP) Approved posts filled by 33% 40% 42% 43% trained health workers Urbadrural-specific HIV 6.8% 6.1% 6.5% 6.3% seroprevalence (national average) Note: Diphtheria-Pertussis-Tetanus vaccine (DPT3); Out Patients Department (OPD); Private, Not- For-Profit (PNFP). Source: Health Sector Strategic Plan Midterm Review Report, March 14,2003 4.3 1 Inview ofthe declining mortality indicators, the government set up atask force to analyze the situation and determine a way forward. The underlying factors identified by the task force as contributing to high mortality have already been outlined above in the water and sanitation section. The task force proposed a wide range o f interventions for reducing mortality rates, which will require the joint efforts o f an extensive cross-section o f players, ranging from the health, water and sanitation, education, and LG sectors, to the energy, roads andworks, agriculture, and gender ministries. 4.32 Given the multiple factors affecting mortality indicators, there is a need to increase the coordination o f sector strategies in health, water and sanitation, and education to deal with the problem o f declining mortality indicators. The health sector recognizes the key area for intervention as the improvement o f reproductive health by increasing access to reproductive health facilities, particularly at the health subdistrict level. 4.33 The intersectoral linkages between the health and water and sanitation sectors, and the health and education sectors, as well as the multidimensional nature o f mortality 6. Government o f Uganda 2002 - 59 - indicators, mean that the funding and overall performance o f these sectors are important for health indicators as well. Water and sanitation are critical determinants o f infant, child, and maternal mortality indexes, and households with access to safe water and sanitation facilities experience lower infant and child mortality rates. Per capita funding inthe water and sanitation sector has increased quite dramatically, from approximately US$0.38 in 1997198 to US$2.79 in 2001/02.7 Iftranslated adequately to increased access to safe water and sanitation facilities (especially for the rural poor), increased funding over the last three years inthe water and sanitation sector should contribute to improved infant and child mortality indicators. However, as shown in the mortality indicators in table 4.1, the increased funding has not been synonymous with an equal reduction in mortality rates. 4.34 The health sector needs to increase the number of skilled health workers to staff health units: there i s indication that planned targets for achieving the minimum staffing norms are not being met, and the conversion plan for multivalent nursing cadres i s well behind schedule. There are also problems related to minimum entry requirements for admission to courses and inadequate management and funding o f nursing institutions (both government and private, not-for-profit [PNFP]) since their transfer to the MOES, which has severely limitedthe number o f new comprehensive enrolled nurses. Although the share of skilled health workers in approved filled positions has increased from 33 percent in fiscal year 1998/99 to 40 percent in fiscal year 2000/01, there i s still a need for further investment in human resource capacity and, in particular, to attract and retain workers inremote places. Allocative Efficiency 4.35 Infiscal year 2001/02, the health sector completedthe second year of the Health Sector Strategic Plan (HSSP) for 2000/05, which was developed withinthe framework o f the PEAP. Fundingfor the sector has increased considerably (by an average o f 9 percent annual rate o f growth over the past three years) but still remains short o f the sector's expectations. The resource allocation to the sector increased from USh 108 million in 2000/01 to USh 184 million in 2001/02 and USh 196 million in 2002/03. It is noteworthy that the increase in resources i s mainly attributable to increases in the government's budget facilitated by donor support. The budget process itself is more consultative and transparent. 4.36 However, inadequate funding, shortage o f trained health personnel, an inadequate network of functional health infrastructure, and serious drug shortages are still challenges to the implementation o f the HSSP. The sector's resource constraint has highlighted the needto improve prioritization o f activities within programs and at the different levels o f the healthsystem, as well as to addressefficiency issues. 4.37 There is clearly a needto reconcile the fundingexpectations outlined inthe HSSP with the healthsector resource envelope. Per capita public health expenditure currently stands at US$& a full US$22 less than the amount the World Health Organization Report 7. MFPED2002 - 60 - o f the Commission for Macroeconomics and Health (2002) recommends should be attained by 2015. This i s not very different from the HSSP recommendation o f US$28 to be attained by 2019. It should be noted that per capita expenditures inthe health sector have grown significantly over the past decade. In 1992/93, public spending was only US$4.38, while private spendingwas US$5.36 (Reinikka and Collier 2001). The amount doubled between 1992/93 and 2002/03. 4.38 The overall health sector expenditure outturn was 95.8 percent, and the shortfall was mostly driven by the overall wage bill and development releases. These figures relate to the first half. They should be updated with third-quarter numbers, which will be available soon. The shortfall in the sector's wage bill was attributed to the presence o f unfilled vacancies at the MOH, Butabika Hospital, the Health Service Commission, and district referral hospitals. The primary health care bill overperformed as a result o f more people accessing the payroll than had been planned. The process o f cleaning up the payroll has beencompleted, and the results have beenforwarded to the MOPS. 4.39 The increases inresources allocated inthe budgetto health over the mediumterm appear to be driven primarily by the Primary Health Care (PHC) grant as well as PAF expenditures (see Table 4.6). In 1998, the health sector's PAF expenditures as a share o f its total expenditures amounted to 5 percent. This figure increased to 42 percent in fiscal year 2002/03, and it will reach 69 percent o f total health sector expendituresby 2005/06. 4.40 There i s a concerted effort in 2002103 BFP to reflect sector priorities as well as clearly incorporated criteria for resource allocation and allowances for hard-to-reach areas. There i s also an emphasis on increasing the amount o f resources allocated to the district level, and increasing subsidies to PNFP units have been addressed through an increase infunding and improved budgetperformance. - 61 - Table 4.6PAF (Health Sector)Expenditures,Fiscal Year2001/02 to 2005/06(USh billion) 2001/02 2002/03 2003/04 2004/05 2005/06 Primary HealthCare MinisterialDevelopment Budget 7.02 20.61 20.61 21.52 22.79 NationalPHC ServiceDeliveryPrograms 20.74 21.1 21.1 24.32 25.58 PHC ConditionalGrant Wage 36.74 43.86 45.31 45.38 45.42 Nonwage Recurrent 14.86 19.67 22.16 24.21 27.06 Development 10.61 7.58 9.04 11.91 16.54 NGOs 11.58 16.61 19.72 23.4 26.57 District hospitals 8.87 8.71 10 11.4 11.52 UgandaAIDS Commission 2.23 2.42 2.42 2.59 2.72 Supportto AIDS orphans and children's rehabilitation 1.43 1.86 1.86 2.69 2.84 PHC subtotal(excl. arrears) 114.08 142.42 152.22 167.42 181.04 Total healthsector expenditures 302.47 337.92 352.2 350.67 261.09 PHC share o ftotal healthsector expenditures 0.38 0.42 0.43 0.48 0.69 Source: MFPED2003h. 4.41 There are efforts to improve the poverty targeting o f resources in the health sector, Per capita allocations o f the PHC grant try to reflect indications from household survey data that poverty in Uganda i s mainly a rural phenomenon and regionally concentrated in the north and east. More resources are gradually being channeled to health subdistricts and lower-level units, resulting in greater coverage in the rural communities where the poor are concentrated. Table 4.7 shows the allocation o f the PHC conditional grant service delivery funds to the six highest and four lowest ranked districts according to the district poverty index, the presence o f health projects, and the demand for health services. Table4.7 Variationin Per CapitaAllocation of the PHCConditionalGrant toDistricts, 2002/03 Note: The ranking is accordingto district poverty levels, health needs, and presence o f healthprojects. For example, Bundibugyo, one of the most disadvantageddistricts in terms health coverage, with a high level o f poverty and highdemand for healthservices, receives morethan 20 percentadditional per capitafunding than Rakaior Masakadistricts. Source: HSSPMidTerm ReviewReport, 2003 - 62 - 4.42 Although the health sector budget i s particularly sensitive to the goal o f poverty reduction from an allocation perspective, it i s not entirely clear inthe BFP how explicitly the poor and vulnerable groups can benefit. This question would be better answered through analysis o f how the poor access and consume health services. An ongoing district PER aims at providing information on the performance o f the health budget at the district level against health sector objectives and targets. 4.43 There are indications from other studies of an increase in the demand for health services demonstrated through increased shortages o f essential drugs, trained staff, and training materials. Despite this increase indemand, the funding o f district health training schools remains a low priority in the education budget, which will have medium- and long-term implications for the health sector. 4.44 Given that the health sector has received the most significant increase in its budget among all sectors, there is a particular need to demonstrate improved sector performance and effectiveness and continued management o f these resources interms o f both efficiency and equity. It i s becoming clearer that inthe short term, the expectations o f the sector's strategy need to be lowered so that they can better be aligned to the sector's resources. An M O H review o f the HSSP in2003 recommendedthat with effects from fiscal year 2003/04, priorities for allocation should be more focused on the major contributors to the national disease burden, with particular attention to the most cost- effective measures for reducing mortality and morbidity. Furthermore, a more careful stock taking should be carried out to scale up the strategies that have worked over the last two years. 4.45 The role o f the private sector in the provision o f health services should be expanded to relieve the overburdened, underfunded national health care system as a measure o f efficiency. Although the national health policy recognizes the important role of the private sector, even subsidizing two of the three private subsectors (facility-based, PNFP and non-facility-based PNFP), there i s still a need to better harness and capitalize on the comparative advantage o f private providers in complementarity with government health services. 4.46 Recent findings by Reinikka and Svensson (2003) highlight the importance o f both for-profit and not-for-profit health providers in Uganda. They find that religious, not-for-profit facilities are more likely to provide pro-poor services and charge lower prices for services than for-profit units, although they provide a similar (observable) quality o f care. Religious, not-for-profit and for-profit facilities both provide better- quality care than their government counterparts, although government facilities have better equipment and provide a wider range o f services. The findings provide an argument for a public-private partnership approach to health care strategy and the needto strengthenthe private sector's participation. Operational Efficiency 4.47 The health sector i s one o f the government sectors that has adopted R O M as one way o f increasing efficiency. R O M i s an approach to management that seeks to use - 63 - available resources in the most efficient way by setting clear and attainable goals. Inthe health sector, output and outcome indicators have been defined, and the sector monitors i t s performance against these agreed targets and outputs in national and district work plans on quarterly and annual bases. 4.48 There i s evidence from a number o f data sources that the removal o f user charges inall governmenthealthfacilities has significantly improvedthe use ofhealthservices by the poor.' An analysis of use patterns carried out after the abolition of user fees shows that once access barriers have been removed, the quality of the service becomes crucial. For example, once drug availability (an indicator o f use o f free Out Patients Department [OPD] services) becomes stable, there i s a sustained increase inOPD use.' The available data on perceptions o f quality (UPPAP I12002; .UNHCO Baseline Survey o f Patient's Rights, 2002 data) suggest that the concerns about quality o f service deteriorating with the abolition o f user fees have not significantly reduced satisfaction levels, although this has tended to vary by socioeconomic status. Figure 4.2New OutpatientAttendancein Ugandan Governmentand Private, Not-for-Profit HeaIth Units 16 11 12 10 E 8 i =0 6 1 2 0 1997198 1998199 1999/00 2000/01 2001/02 Source: Reinikka and Svensson 2003 4.49 Much attention has been given to efficiency at the macro level on issues such as improved resource allocation formulae and budget funds predominating in bilateral or multilateral donor vertical project funds. However, there i s a need to address efficiency at the micro level at the district and service delivery levels, as well as issues such as procurement and construction. 4.50 Value-for-money and tracking studies have typically been carried out on an ad hoc basis, and there i s a need to better systematize this process and integrate it into the sector's plan. Despite this, the sector has aggressively tackled issues raised by a number 8. World HealthOrganization-MOH study, 2000,2001, and 2002 data; Annual Health Sector Policy Review (AHSPR) 2001/02 data; UPPAP I12002 data. 9. World HealthOrganization-MOH study, 2000,2001, and 2002 data; Annual Health Sector Policy Review (AHSPR) 2001/02 data; UPPAP I12002 data. - 64 - of studies, most notably, the tracking study o f Primary Health Care (PHC) conditional grant funds, which was finalized in 2001. That study attributed the bulk o f delays in releases (67 percent) to the M O H and MFPED. After release from the MFPED, the delay inthe flow o ffunds was either due to internalbureaucracy (MOH), lacking accountability from the lower levels, or both. The recommendations o f the 2001 PHC grant tracking study were forwarded by the M O H to the MFPED and have resultedinimproved flow o f funds. 4.51 The drug tracking study was undertaken in 2002, and its recommendations were forwarded for action to the MOH. As a result, efficiency and rationalization o f drug procurement have improved. During the current fiscal year, the M O H will undertake a tracking study o f the PHC conditional grant for shared services, which i s managed at the center. In addition, a health sector PER analyzing efficiency at the district level was carried out infiscal year 2002/03. 4.52 Using a broad definition o f efficiency as the relationship between inputs and productivity, making the following observations:I outputs, Reinikka and Svensson (2002) com fare labor costs and health worker 0 Interms ofproductivity (output per worker): >> There are considerable differences across individual facilities. There are some differences in output per worker across ownership categories, > with low numbersfor nonprofit facilities. There appear to be notable regional differences, with low output per worker in the central region for both government andprivate facilities. 0 Interms o funit costs: > The higher average remuneration in government facilities appears to more than offset the higher productivity, such that unit costs are higher in government than infor-profit and nonprofit facilities. k This difference is largely driven by the very high unit costs in government facilities in the central region, which i s consistent with higher than average remuneration and lower than average labor productivity. 4.53 There is an increasing recognition o f the importance o f private providers in the health sector, and there are even suggestions that the government, for-profit, and nonprofit sectors should provide roughly equal portions o f the health care inthe country (Hutchinson 2001). The individual comparative advantages o f each sector are not clear, nor are the equity o f private provision and the overall efficiency o f the health sector. Ongoing research inthis area by the World Bank and the M O H analyze how the mix and targeting o f health services are affected by the introduction o f a performance-based contract betweenthe purchaser (the government) and private sector (nonprofit) providers of health services. The goal o f this research i s to improve access to health services by the poor, thus it addresses not only efficiency, but also equity issues. ~ 10. Reinikkaand Svensson(2002) report on findings from a baselinesurvey o f dispensaries(with and without maternity units) and districtadministrationsthat was implemented inUgandain2000. - 65 - C. EDUCATION SECTOR Background 4.54 The education sector has the longest history o f all the sectors in terms o f addressing budget efficiency through its SWAP, Reinikka's and Collier, 2001 work on service delivery in Uganda pioneered the systematic analysis o f sectoral expenditure flows in the education sector. The first service delivery study analyzed the hypothesis that public spending-and, inparticular, education sector inputs-had not beentranslated into increased outputs despite substantial increases inthese resources over 1990-95. The study's findings disputed official records, which appeared to have considerably underreported the increase in primary school enrollment between 1991 and 1995 after increased funding o f the education sector. Despite the increases in primary enrollment, the study also showed a significant amount of leakage of resources before they reached the schools and serious accountability problems with respect to nonwage education spending." The study found that less than 30 percent o f funding intendedfor nonwage public spendingactually reached the schools in 1991/95. District authorities retained and usedmost o f the capitation grant meant for schools, and, similarly, schools were allowed to keep only one-third o f the mandatory tuition fees they collected, with the rest going to the district education officers. 4.55 A second survey in 1999/00 showed significant improvements: schools were by then receiving more than 90 percent ofthe intendedcapitation grant. Since the release of the survey results, there have also been some changes in the system to improve the transparency o f the process o f transferring funds and institute standard practices o f procurement. Monthly transfers o f public funds are now reported in the media, school- based procurement has replaced the central supply o f construction and other materials, and an effort has been made to institute basic public accounting systems in the public sector, including districts (Ablo and Reinikka 1998). These improvements were achieved through a public information campaign by the government at the national, district, and school levels. 4.56 There has beena continual focus withinthe education sector on seeking efficiency gains that will lead to better indicators across all the subsectors. Tracking studies are now an annual exercise in the sector and a means for shaping the policy discussion and increasing the efficiency with which resources are used. The education sector review o f October 2002 highlighted the recent "Draft Tracking Study Report on Teacher Recruitment, Deployment and Payroll Management," which was carried out in early 2002, as well as the studies on management and use o f UPE capitation grants (1999 and 2001). 4.57 The overall aim o fthese studies inparticular has beento provide a mechanism for ensuring the efficient and effective use o f funds, especially at the lower levels o f LG. The MOES made a presentation at the Education Sector Review (ESR) on the 11. Only 2 percento fpublic nonwage education spendingwas found to be reaching schools in 1991, and four years later, this share had increasedto only about 20 percent (Reinikka Collier 2001). - 66 - recommendations o f these studies, as well as the status o f their implementation. A significant number o f the (27) recommendations from the UPE capitation grants studies-for example, development o f a strategic plan for management o f UPE funds, standardization of all UPE documents, and efforts to enable UPE funds reach the schools at the beginning o f term-have been carried out, although a few remain outstanding. There is also a detailed matrix inthe tracking study on teacher recruitment, deployment, and payroll management that outlines specific actions to be implemented. 4.58 The education sector has placedmuch emphasis on improving equitable access to basic education as well as improved quality through the Education Strategic Investment Plan, Systematic monitoring o f indicators such as the PTR, PCR, and pupil-book ratio (PBR) has enabled the sector to track progress over time (see Table 4.8). Furthermore, school facility grants (SFGs) and UPE capitation grants have contributed to the realization o f equity and quality targets. The sector has also benefited significantly from the PAF, given that all primary education-related expenses are protected from revenue shortfall-related budget cuts. Table4.8Education Sector Targetsfor QuaIityInputs Fiscal Year Ratio 2002/03 2003/04 2004/05 PTR 54:1 50:1 44:1 PCR 95:l 90:1 84:1 PBR (P3P4) 3:l 2:1 1:l 4.59 In line with the recommendations of the October 2002 ESR, the sector also recognizes the needto increase equity by providing more opportunities for admission to secondary school, especially for the underprivilegedand for girls. O f similar importance i s the need for further training o f school officials, including teachers, head teachers, and boards o f governors. To improve the quality o f education, there i s a recognized need to review the curriculum and provide sufficient copies o f instructional materials to support it. Finally, the ESRacknowledgedthe necessity o f scaling sector responseto HIV/AIDS and improving coordination among the implementers o f the HIV/AIDS interventions. Duringfiscal year 2000/01, budgetary allocations to educationconstituted31.2percent of the total government recurrent discretionary budget, and the primary education allocation was 69.2 percent o f the overall education budget. In addition, the actual allocated share of releases to primary education was 69.4 percent as a result o f supplementary expenditure in the teachers' wage bill. Despite these global increases in the education resource envelope, there i s still a considerable amount o f pressure on resources and many subsectors face a funding gap. There i s a need to reevaluate intrasectoral allocations and, in particular, the share o f the budget currently allocated to primary education. 4.60 There are two main long-term financing issues arising withinthe education sector, Inthe first case, the question arises whether it is reasonable and sustainable for the sector to continue receiving more than one quarter o f the budget and whether the balance betweenallocations to healthand education are equitable. Withinthe education sector, it - 67 - i s not clear what the optimal balance o f resource allocation between different subsectors should be, and it i s not entirely clear how, for example, to maintain and even improve access to primary education and increase access to secondary education withinthe current resource envelope at the same time. Allocative Efficiency 4.61 The education sector budget continues to be skewed heavily toward primary education. However, in the 2002/03 approved budget for the education sector, 65.5 percent was allocated to primary education, down from 69.0 percent in 2000/01. The primary education budget has increased by 40 percent since 1999, and this increase has been matched by an increase in primary school enrollment (150 percent). Secondary school enrollment also increased duringthe same period (57 percent), yet this increased enrollment was accompanied by only a 12 percent increase in the secondary school budget. Given the infeasibility within the short to medium term o f any dramatic increases in the overall education resource envelope, the sector must emphasize gains in efficiency as the means for increasing the impact o f its activities. Nonetheless, in the long term, there should be a greater share o f the budget allocated away from the primary education subsector toward secondary education. However, inthe short to medium term in the postprimary subsectors, a greater focus needs to be directed toward addressing inefficiencies in the postprimary education and training (PPET) subsector. This should include the general secondary; business, technical, and vocational education and training (BTVET); and tertiary sectors. Table4.9EducationSector ProgrammedExpenditurein theMedium Term(USh Billion) Source: MOES. 4.62 The share o f the PPET subsector is expected to increase marginally in the next fiscal year as the share o f the primary education subsector declines. This i s in line with the ESR's recommendations and the shifting objectives within the sector, in recognition of the needto share resources more equitably within the sector. 4.63 The recent ESR recommendedthat the government shouldreviewthe share o f the budget for primary education in accordance with completion and progression o f the current UPE pupils through the PPET and tertiary education levels. Flexibility that - 68 - permits reallocation from underspending programs to other critically needy programs should be negotiatedbetweenthe MOES and MFPED. 4.64 Uganda's successful UPE policy has qualified it for the Education for All (EFA) Fast Track Funding Initiative. The EFA provides further external assistance, and the implications o f this commitment will have budgetary as well as macroeconomic consequences. There are concerns about the absorptive capacity of the sector and the economy itselfthat may not allow for the additional resources. Operational Efficiency 4.65 The education sector has gradually developed a culture of assessing budget efficiency by minimizing its costs and maximizing its outputs and outcomes. Unlike other sectors, the education sector now conducts annual tracking studies. The sector has also used annual reviews, internal audits, and end-of-project evaluations to improve the flow o f resources as well as its overall operational efficiency. To ensure that reforms are followed through, the findings o f tracking studies are usually discussed during the ESR and incorporated into the sector's action plan for the following year. Implementation o f recommendations arising from the tracking studies i s carried out through the Education Sector Consultative Committee andthe ESRs. 4.66 The most recent study carried out in the education sector tracks teacher recruitment and access to payroll. The results identify specific loopholes, including gaps inthe overall system of recruitment and payroll, which have ledto delays inrecruitment and a lagging commitment to teacher recruitment at the district level. The study found that teacher recruitment i s further encumbered by an inadequate knowledge o f the guidelines among chief administrative officers (CAOs), who delayed hiringnewteachers, and among new teachers, who delayed submission of their acceptance letters and taking uptheir appointments-especially iftheir deployment was to remoteareas. 4.67 The tracking study also identified problems with payroll and financial management that were causing delays in payroll access, including the appointment of teachers in contravention o f standing regulations. Recommendations to improve teacher management based on the results o f the study were presented by the ESR, and a number o f actions to be taken for improving the number o f teachers on the payroll (especially for the 10 districts with the worst PTR) will be incorporated into the sector's action plan. This process reflects the institutionalization of budgetary efficiency issues in the education sector whereby tracking studies and the operational constraints they identify are linkedto sector action plans through the ESR. Giventhe progress made inidentifying allocative and operational constraints through tracking studies, there i s an increased need for the education sector to effectively integrate the recommendations into sector plans, to implementand subsequentlytrack their progress and improve overall efficiency. 4.68 Five years into the UPE program, sector policy is focusing increasingly on postprimary education, given the current pressures on existing facilities as a result o f the UPE bulge. For example, in 2002, 401,555 students sat for the Primary Leaving Examination and applied to secondary schools. Currently, the available postprimary - 69 - schools and institutions can absorb only 50 percent o f these primary school leavers. The situation will be compounded in 2004, when about 900,000 UPE candidates will be seekingadmission to postprimary institutions (MOES 2002). Year 1997 I998 1999 2000 2001 2002 2003 Numbero f PLE 211,749 250,726 270,449 304,636 363,801 401,555 461,788 candidates YOincrease 0% 20% 8Yo 13% 19% 10% 15% Source: Uganda National Examinations Board, The 2003 figure is an estimate. I 4.69 InApril 2002, a task force was established by MOES to draft a PPET policy and costed framework that makes provisions for absorbing the UPE bulge into postprimary education. To arrive at the PPET policy framework, the MOES commissioned eight policy studies in2001/02 on thematic issues to form the basis for the policy. 4.70 The efficiency measures agreed upon for implementation o f the PPET strategy include increase inthe teaching load (implementation o f this has begun), selective use o f double shifting in areas where it i s appropriate, particularly in urban day schools, where the student concentration is particularly high. Establishmentof a double shiftingpilot for primary schools has been slow, although scaling up of the effort i s planned during the 2003 school year. 4.71 The report on costed options for the PPET policy recommends a range of efficiency measures and outcomes that would respond to the urgent need to expand access to PPET as the number o f primary graduates increases. Among the recommendations are a requirement that secondary teachers teach at least two subjects, slowly increasing the PTR from 17 to 30, expanding 0-level general secondary education, and slowing downthe construction o f community polytechnics. 4.72 A value-for-money analysis ofthe school facilities grant program has beenagreed upon for fiscal year 2003/04 and the Terms o f Reference are being developed for discussion with various stakeholders. In addition, a tracking or a value-for-money audit on the SFGprogram will be completed for the October 2003 ESR. D. JUSTICE,LAW, ANDORDERSECTOR (JLOS) 4.73 The JLOS carries the mandate for implementing issues outlined inpillar 2 of the PEAP. The stated sector mission-"to enable all persons in Uganda to live in a safe and just society"-is supported by a sector strategic plan that focuses on reform o f the criminal justice system through ensuring access to justice, efficiency and effectiveness, and the quality o f justice. The outcomes o f the sector reform strategy will improve individual safety o f all persons, ensure property rights, improve access to justice, and support other economic and social reforms inthe country. 4.74 The JLOS faces some unique challenges in carrying forward its mandate. Its complex institutional structure encompasses a number of institutions with differing - 70- mandates and differing institutional priorities, which must all be lined up under a common sector vision and a common sector budget and budgetary process. Despite this uniqueorganizational characteristic, the sector is making progress intackling the issue of efficient management o f expenditures. The sector comprises nine institutions: the Uganda Police, prisons, judiciary, Directorate o f Public Prosecutions, Judicial Services Commission, Ministry o f Internal Affairs, Ministry o f Justice and Constitutional Affairs, Ministryo fLocal Government, and Uganda Law Reform Commission. 4.75 The JLOS developed a strategic investment plan in fiscal year 2001/02 for the subsequent five-year period (beginning in 2001/02), emphasizing the areas o f criminal justice, administration o f justice, civic and legal education, and criminal law reform, Emphasis for improving commercial justice i s on establishing a commercial court, companies and registries reform, commercial law reform, and streamlining the practices o f commercial lawyers. There has been progress in arriving at a common sector investment plan, but the institutions within the sector have yet to align their spending plans in accordance with the sector priorities. In addition, the sector still needs to streamline its programs inline with the MFPED's MTEF figures, 4.76 Interms ofresources allocated from the central government, the sector's share of total expenditures was 6.72 percent in 2001/02 and 6.86 percent in 2002/03. This share was 7.0 percent in 1998199 and has remained quite stable over the last four years, neither increasing nor decreasing dramatically. 4.77 A study on the financial management systems o f the JLOS institutions was recently completed, with the aim o f evaluating the sector's capacity for efficient use o f its funds and identifying weaknesses in the system and areas o f intervention to reduce fiduciary risk and increase overall efficiency. The financial management o f a sector i s crucial for overall efficiency in the use o f resources because it points to the sector's capacity to manage its resources. In fact, a sector can only function effectively in the delivery o f its programs where there is a robust financial management system. Therefore the baseline study on financial management systems o f the JLOS institutions shows progress within the sector in addressing efficiency issues. It i s a first step inthe direction o f setting up an effective system o f monitoring fiduciary risk within the JLOS institutions. 4.78 In general, the study identified a high level of fiduciary risk in the sector. The sector was found to be lacking in efficient means, such as analytical sector reviews, to evaluate progress across the key institutions. Inaddition, the manual accounting system found in most o f the institutions was identified as a root cause o f delays in reporting and generation o f inaccurate information. There was a severe shortage o f qualified personnel, which translated into inadequate execution o f financial management procedures and functions, leading to fiduciary risks. Furthermore, the study also found a culture o f nonaccountability and noncompliance withinthe JLOS institutions. 4.79 Among the key recommendations were the needs to develop analytical reviews in all the institutions; develop an output-focused approach to planning, management, and reporting; and improve accountability and reconciliation measures, especially at the - 71 - district level. The main conclusiono f the report-a highlevel of fiduciary risk, implying that there are sufficient loopholes within the financial system to provide opportunities for mismanagement o f funds-necessitates prompt interventions within the sector to mitigate this risk. The study suggests that a substantial reduction of the fiduciary risks identified can be achieved ina relatively short period o f one to one and a half years. 4.80 The donor community has raised concerns about whether the practical recommendations offered by the study team will be effective in the short term if addressing the more general key findings i s a much longer-term effort. Inthe end, these recommendations will be effective, it seems, only if parallel efforts on cross-cutting financial management issues under the Economic and Financial Management Project 2 are successful. The concern at the time o f the PER workshop in May 2003 was that very little progress has been made in implementingthe recommendations since the study was presented during the December 2002 review. The good start made by the sector in identifying its fiduciary weaknesses needs to be compounded by timely and adequate implementation of the recommendations to fully realize the benefits for efficiency. E. CONCLUSION 4.8 1 The above sections aimed.to underscore the growing importance o f the issue o f public expenditure efficiency. Inthe water and sanitation, health, education, andjustice, law, and order sectors, there i s a growing awareness of the limitations of government funding and a concerted effort to improve the effectiveness o f sector funding so that the sector objectives are achieved. As was noted inthe fiscal year 2001/02 PER, the work on efficiency issues needs to be deepened in the sectors where it has begun and broadened beyond the water and sanitation, health, education, andjustice, law, and order sectors. 4.82 There are varied lessons to be learned from the experience of the four sectors in increasingthe effectiveness with which resources are used. The education sector, with its relatively longer history o f resource tracking, should be emulated for its ability to translate diagnoses made in analytical studies to policy recommendations, and the implementation of those recommendations i s featured in the biannual ESR. The level o f institutionalization of the process o f increasing efficiency forms a good basis for deepening this process across subsectors and all levels o f government to address deeply rooted problems. The health sector, which also has a longer history o f expenditure tracking and value-for-money analysis, i s focusing on the impact o f its resources at the district level. Inthe water and sanitation sector, there i s a recognition o f the needfor in- depth value-for-money studies on the basis o f sound methodologies, which can effectively be used for improving the sector's efficiency. The efficiency issue is relatively newer inthe justice, law, and order sector, and this sector is to be commended for initiating an exhaustive and in-depth analysis o f its financial management systems. Streamlining the areas o f financial management and procurement i s particularly crucial to a sector's overall efficiency, and the work carried out by the JLOS inthis area should be replicatedby other sectors. - 72 - 5. LOCAL GOVERNMENT BUDGET AND FINANCIAL MANAGEMENT ISSUES A. INTRODUCTION 5.1 This chapter describes Uganda's decentralization process, with the aim of pointing out areas in financial management that need improvement. Thorough analysis o f financial management problems in LGs, which would identify possible assistance required in budgetingand financial management, was not possible because o f paucity o f data and the short time LGs have existed. This concern will be addressed in subsequent PERs. As o f now, LGs are relatively new, and there i s a lack o f information that would facilitate good analysis to inform the process o f improving financial management in LGs across the country. This chapter is based on the premise that decentralization is a fundamental reality within the country and prospects for a policy reversal are remote, and there i s a needto understand this with a view to identifying areas o f support. 5.2 The core problem the country faces in decentralization has three interrelated dimensions. First, Uganda has only partially decentralized. Although some o f the fundamental formal steps intrinsic to any decentralization process have been taken (for example, legal establishment o f formal authorities with clear functional responsibilities, legislation o f local financial management procedures, and so on), some critical elements o f a decentralized system are missing(for example, resource allocation autonomy, strong LG institutions, and so forth). Second, most LGs suffer capacity constraints in several critical areas. Third, decentralization i s threatening to weaken rather than strengthen efficiency-promoting incentives by encouraging upward accountability o f local councilors to the central political structure rather than downward accountability to communities. 5.3 The focus of this chapter, therefore, is on how LGs can improve budget efficiency and performance. It examines the rationale and legal framework for decentralization, the practice o f fiscal decentralization, and actions that could strengthen planning participation and financial management inLGs. Theoretical Frameworkfor Decentralization of Public Service Delivery and Emerging Challenges 5.4 Decentralization o f public service delivery was implemented in Uganda on the rationale that it would improve efficiency and effectiveness inservice delivery as a result o f enhanced accountability to beneficiaries o f social services at the LG level. Decentralization started in the form o f political decentralization, which started with the National Resistance Movement's (NRM) creation o f "resistance councils" during the "bush war" inthe early 1980s. - 73 - 5.5 At the time, centralizationo f service delivery was facing many challenges-such as high-level corruption, resource slippages, poor prioritization o f development interventions, and delays-all o f which were combining to bring about poor service delivery at the grassroots. Fiscal decentralization was implementedpartly as a response to these challenges andpartly to be inline with the political developmentso f the day. 5.6 However, there i s currently a lot o f concern within government and the donor community about the state o f financial management inLGs and its impact on delivery o f social services at the LG level. This arose from desk reviews o f several financial management studies that were carried out over the last 3 years.I2 The studies exposed several common problems. Some line ministries and sectors conducted expenditure tracking studies (education, health, water, and so on), and some bilateral donors did studies specific to financial and budget management. In preparation for the budget support modality o f channeling donor support, the World Bank and the government undertook a Country Financial Accountability Assessment (CFAA). Because the CFAA did not go in depth into the area of LG financial accountability, this chapter reviews CFAA issues as they relate to LGs inUganda. LGandLegalFramework inUganda 5.7 The Local Government Act o f 1997 provides the legal framework for decentralization in Uganda. The act empowers different layers o f LG to exercise political, administrative, and financial management for the development o f the people in the respective areas. LGs include higher-level (designated as LCV) districts and municipalities. These are the focal points o f public administration. They have elected political leadership: an executive, headed by an elected chairperson, and an elected council. The LCIII (subcounty) i s another important level o f LGs. It also has an elected chairperson and council and plays a substantial role in PA. Although the L C I (village) level has an elected chairperson and council, its P A role i s not as significant as that played by the L C V and LCIII levels. L C I activities involve little use o f public financial resources, and at that level, there are no technical staff to undertake day-to-day PA. The L C V and LCIII levels have technical staff that perform P A functions. The Process of FiscalDecentralization: A PhasedApproach 5.8 Because o f uncertainty and also to draw useful lessons, fiscal decentralization was implementedin phases, beginning with 12 districts in 1993. The districts were given a status o f "votes" like line ministries and government departments. As a result, decentralization was unevenly implementedat the district level and also within different sectors. 5.9 Inline with the government's program of fiscal decentralization, more LGs were added to the list. All LGs today have some degree o f budget and financial decentralization. However, fiscal decentralization has been possible only for the recurrent budget and not for the development budget. The development budgethas yet to l2The desk studyreviewedof studies inthe field o f financialmanagementthat were carriedout by the Ministry o fLocal GovernmentMinistry of Financeand Sector Ministries. - 74 - be fully decentralized, mainly because o f the technicalities involved. More than 80 percent o f Uganda's development budget i s funded using borrowed resources from donors, and implementation i s in the form o f projects and development programs. The delivery and reporting requirements o f different donors have hitherto made decentralization o f the development budget difficult. The extent to which the donor support modality can address this challenge i s limited because o f the sheer number o f LGs as well as their inadequate capacity. 5.10 The sections that follow describe financial management in LGs from three dimensions: budget execution, budget reporting, and verification o f expenditures and results. The description and main findings reported are based on an in-depth study o f three districts-Mbarara, Masindi, and Soroti-and one municipality, Kabale. In addition, this chapter draws on the work carried out in the context o f the preparation o f budget and financial management guidelines for the FDS implementation, which will be launchedon a pilot basis in 15 districts beginningin2003/04. B. PLANNING 5.11 The planning and budget preparation process inUganda's LGs is comprehensive and elaborate. All LGs are required to have a three-year development plan. Their planning units work out development plans,13 with input from the sectors and the lower local governments (LLGs). The quality o f most LG plans has improved, although both the quality ofthe plans andthe accompanying processesvary greatly. Some plans include analysis (although sometimes rather superficial) o f poverty concerns, gender, environment, demography, economy, infrastructure services, and so on. The plans are increasingly beingprepared using a bottom-up approach, with input from L C I up to LCV. Citizens' involvement in planning i s very limited-restricted to the yearly budget conferences, with hardly any attendance by civil society. 5.12 Although planning has improved over the years, the quality o f the plans leaves room for improvement, especially interms o f analysis o f per capita sector coverage, costs estimates and linking to the budget and available resources, the realism in the planned activities and outputs, and the way the plans are rolled. Some o f the LGs merely copy large sections o f the plan from one year to another. Other LGs make an effort to update the analysis. 5.13 Weaknesses identified at the planning level include (a) lack o f primary data and data banks; (b) inconsistencies between the District Development Plan (DDP), BFP, budgets, and work plans in terms o f inputs, activities, outputs and objectives; (c) unrealistic plans because o f lack o f cost-constrained planning (the priorities against sources are done in the final stages o f the planning instead o f simultaneously); and (d) rolling o f the DDP by postponing activities without due consideration o f the realism o f the plans. Inaddition, the planning at the LLGs needs considerable mentoring support, and there i s also further needfor internalization and support to use o f recent FDS budget guidelines. l3InKabaleMunicipalitythe planner fromthe district supportedthe development o fthe MDP. - 75 - C. BUDGETING 5.14 LGs kick off the budget process by issuing a budget call and a fixed time and activityplanto heads of departments (HODS),but sometimes this is not possibleinsome districts because of time limitations. The BFP is generally less well understood by district-levelstakeholders, and its preparationi s seen largely as a technical exercise. The planning department usually compiles the BFP, and only one person in each sector is involved inpreparingthe sector BFP components. The BFP framework is therefore not effectively usedas a tool for sector prioritizationand decision makingby technocrats or politicians. 5.15 The budget is preparedby the ChiefFinancialOfficer (CFO)/Finance Department (FD) with some inputsfrom the sectors andthe LLGs.A budgetconference, programmed for December or January according to the FDS, i s normally held in the middle of or late inFebruary. There is hardlyany participationof the citizenry inthe budgetconferences. In districts where fiscal decentralization is more advanced, the LG authorities have started making radio announcements of the budget figures and other relevant LG activities. The indicative planning figures for the key grants (for example, the SFG and LGDP) are posted on notice boards of the administrationbuildings, and all LGs inform citizens throughmore or less regular radio programs. 5.16 Annual work plans for most sectors are preparedinresponse to the requirements o f central government and PAF guidelines. Some of the non-PAF sectors have also adoptedthe idea, althoughthis is not widespread. 5.17 An attemptto link the planningandbudgetingprocesses is evident inall LGs, and most politicians and technical staff say that the DDP, BFP, and budget were all important. However, this was not reflectedin the planning and budgeting documents, which is partly explained by timing differences in the preparationof these documents. For example, the BFPhas to be submittedby the end of December,a longtime beforethe final stages of the budget and planning process. Because of concern that there i s a proliferationof planningand budgettools, some stakeholdersthink that the DDP andthe BFP should be merged. Adoption of the budget in most LGs is delayed for several weeks, typically because of lack of political agreement on priorities. Developmentplan problems include unrealistic budgeting, especially poor estimation of revenue, and poor prioritization of expenditure. Lack of politicalwill to make unpopular decisions further compoundsthe problems. 5.18 Revenue and expenditures are planned over the medium term, but the development plans, BFP, and budget do not present planned targets in an explicit and comparable manner. It is also often difficult or impossible to identify and link the activities identifiedinthe DDPto those inthe BFP andthe budget. Exceptina few LGs, the cost implications(operationalandrecurrent costs) ofthe developmentprojectsare not explicitly incorporatedandmeasuredinto the budget andBFP. 5.19 Although a lot of effort is put into budget preparation, there i s little effort at the stage of implementationof the budgetto ensure achievement of set targets, especially on - 76 - the revenue side. Consequently, cash flow problems characterize budget implementation and make achievement o f the intendedobjectives difficult. D. COMPOSITIONOF EXPENDITURE AND REVENUES 5.20 Analysis o f the expenditures and revenues show that LGs are becoming more and more dependent on transfers from the central government. The districts raise only between 0 percent and 10 percent of their budgets. Inmany districts, the proportion of locally raised revenue to their budgets i s on a downward trend, mainly because small local revenue bases have hardly expanded in recent years. For example, Kabale's LG contribution to the district budget fell from 40 percent in 1994/95 to 15 percent in 2002/03. In all sample districts, education accounted for the biggest share o f budget resources, amounting to approximately 50 percent o f the budget. General administration (administration and finance departments), which accounted for approximately 15-20 percent, came second. E. INTERNAL CONTROLS 5.21 Apart from the central government regulations (e.g. LG Financial and Accounting Regulations [LGFAR], treasury bill regulations, and so on), there are no written procedures or standards for internal control in most districts. Thus, mechanisms for preventingconflict o f interest are not adequate. Separation o f duties and regular rotation o f positions could be used as tools o f internal control to complement internal audit, which initselfis inadequate. 5.22 Increasing revenue collection i s one o f the greatest challenges facing LGs. Experiences vary greatly, and some LGs have put measures in place to increase revenue collection. Initiatives identified include (a) more advanced printing o f tax tickets; (b) better control o f the tax collectors; (c) more support from the districts to the subcounties and the parish chiefs (and the tax collectors); (d) delinking politicians from tax collection; (e) contracting out o f some o f the tax collections; and (f) sensitization o f the citizens. Cofunding requirementshave increased attention on revenue collection by LGs. Ideas on improving revenue collection by LGs are emergingfrom studies and practices in selected LGs. These should form the basis for exchange, dissemination, and mutual support to enhance future revenue generation14. 5.23 Nevertheless, revenue collection has increasingly been influenced by negative political interference from all levels o f government-especially the L C chairpersons, who "protect" their citizens from paying taxes-and political interference in the actual tax collection (some o f the LGs have only managed to collect approximately 50 percent o f their budgetedrevenues). This interference has made ithard for the tax collectors andthe administration to improve or even just stabilize the tax yield. Other problems that characterize revenue collection in LGs include flaws in the enumeration and assessment process, suboptimal timing o f the enumeration, assessment and printing o f tax tickets, l4The Studyunderlinedthe importance of the new initiative on development of bestpractices initiated under the LGFC and supported by UNCDF, Danidaother donors. - 77 - poor maintenance and updating o f the tax registers, and lack o f control o f the revenue collectors. (There i s evidence that revenuei s usedat source and not properly banked). 5.24 Some o f these problems are caused by the poor organization o f tax collection, where many levels o f LGs are involved without proper checks and balances. Weak intemal audit with inadequate capacity compound the problem and make control o f transactions at the LLGs difficult. In some districts, there are a lack o f sufficient tax registers and problems in the horizontal coordination betweenthe various tax authorities (for example, divisions). Finally, all LGs experience severe problems in the vertical sharing o f revenues among LCV, -IV, and -111, caused by the setup and division of tasks and responsibilities in the Local Government Act. Both the politicians and the administration at the district level have pointed out the necessity o f changing the Local Government Act in terms o f the revenue-sharing arrangements to improve the tax administration and avoid the present fragmentation o f limitedresources.l5 F. PLANNINGFORPROCUREMENTAND TENDERING 5.25 Procurement and tendering are areas where significant improvements have been made. The bodies required by law to be established (the tender boards and technical evaluations committees) are up and running in all LGs. Most o f them have elaborate procurement plans and regular meetings to handle procurement. Most districts and municipalities have a list o f prequalified contractors. The list i s updated every year, and evaluation criteria for bid assessment are announced to the contractors as part o f the tender material. 5.26 Although procurement has generally improved, the following areas still need a lot o f attention in some districts: (a) lack o f quarterly returns o f purchases to control successive purchase; (b) weak transparency; (c) failure to publishminuteso f tender board meetings; (d) failure to update the list o f contractors; and (e) poor follow-up on the winning bids. There is also a need for better linkages between the tender board procurement and work plans and cash flow projections to better plan for procurement and help ensure that the LGs do not enter into too many contracts and overcommit themselves. Some LGs have experienced severe problems with various types o f conflict of interest (hidden tenders, political appointment o f tender board representatives, and so on) and lack o f explanation and transparency when the tender board overrules the technical evaluation committees. In future, these issues are expected to be reviewed under the Country Procurement and Accountability Assessment, which currently deals mostly with procurementissues at the central government level. G. CASHFLOW MANAGEMENT 5.27 Cash flow management i s especially poor in LGs. Very few o f them have a real cash flow budget, which i s a budget prepared at the beginning o f the financial year to predict when revenues will be received or when payments will be made, broken down into manageable periods (for example, quarters) and then updated regularly according to l5This is also supportedby another review:RevenueSharing, Phase11, Local GovernmentFinance Commission, July 2002. - 78 - new projections. Therefore, there have been insufficient efforts to plan in advance to ensure that the district always has enough working capital. 5.28 Nevertheless, most LGs have formed committees that sit weekly or monthly to consider the pressingneeds o f the sectors against available funds. Some LGs involve the executive committee inthese decisions, and others leave the business to the CFO and the finance department. Ironically, neither the budget nor annual work plans feature prominently in the management o f the cash flow, their preparation notwithstanding. Conditional grants and earmarked donor and cofinancing requirementare often the main reason i s adherence to budgetand work plans. 5.29 Insome LLGs, there is a view that persons that lobby most get the biggest share o f local revenue disbursed to their areas, with total disregard o f budgetary allocations. This is particularly the case for elements of budgets funded from local revenue. Some districts require departments to identify activities for the month, and these are not necessarily consistent with the annual work plans. These arise because local funds are generally less than the budgeted resources. Current decisions on priorities are thus o f utmost importance. 5.30 Despite the attempts to improve on the cash flow, all LGs had severe problems with meetingthe current payments because of significant outstanding arrears. Cash flow decisions are often political and concentrated at the executive level, which undermines the importance ofpolitical decisions made by the whole council at the time ofthe budget process. Commitmentand ExpenditureControl 5.3 1 Purchases initiated by the HODSgenerally comply with the LGFAR requirements in terms of number of signatures, supporting documents (for example, Local Purchase Orders [LPOs]), and control by the internal auditor. Checks are the usual means o f payment in all transactions. The time it takes to approve payments and commitments varies from two days to three weeks-some LGs approve payment much faster than others. Similarly, after approval o f payment, the period for processing payment varies a lot among LGs, and insome cases, payment o f bills takes a long time even after approval o f payment. 5.32 Common challenges in commitment and expenditure control include failure to update the vote books, weak control o f commitments against the available cash inflow, and making correct entries in the vote book (updating the vote books and commitment columns inthe same books as new commitments are enteredinto). 5.33 There are significant outstanding and delayed payments o f bills in all LGs. It is not easy to establish the magnitude o f the problem because unpaid bills are not recorded. Bills are often not paid even when they are based on commitments and cash is available. - 79 - Often there are "higher priorities" than paying bills that arise from political interference. The problem o f unpaidbills inLGs needs immediate attention.l6 5.34 Some LGs have neglected the role o f HODs as vote controllers for their respective sectors, in managing sector finances, and in controlling the vote book. Where HODs are not involved and vote books are held by the finance department, there i s generally more o f a problem in updating sector books o f account. This makes it impossible for the HODs to follow up on their work plans against available funds. It undermines the ability of HODs to manage their own finances and hence their responsibility for their department's performance. PayrollManagement 5.35 The conditional grant payroll i s issuedmonthly by Uganda Computer Services o f the MFPED. On its receipt in an LG, the personnel department examines it for verification and reconciliation. Salaries are paid through the banks by credit transfers directly to the individual employee's bank account. Generally, there are separate bank accounts for teachers, delegated staff, and district employees (paid by the districts' own revenuesources). 5.36 Practices in administration o f salary advances vary a lot across LGs: some fail to update salary advance registers or have no salary advance registers, and others keep them fully up to date. Advances are usually given inthe amount o f one month's salary with a three-month recovery period. Delayed access to the central government (MOPS) payroll by newly recruited staff is a big challenge. Accessing the payroll could take as long as a year. Other payroll challenges confronting LGs include minor delays in payment o f salaries o f "own" staff and considerable outstanding pensions and gratuity in some o f the LGs. Accountingand Record-Keeping 5.37 Compared with the situation three years ago and regional experiences, the situation in accounting and record-keeping in LGs has improved tremendously. However, there i s a lot o f variation betweenLGs in this regard. In some, the situation i s inserious needofattention, butinothersthere are well-managed systems andprocedures, Most o f the LGs have maintained the required books o f accounts, most o f them up to date. There are several instances o f problems with keeping the books up to date and ensuring timely internal reconciliation, especially with the general ledger. Bank reconciliation i s generally made for all required registers, although some o f them are usually months behind. 5.38 All districts use a cash-based system of accounting. Municipalities use a form of modified accrual accounting (the system does not immediately capture commitments, creditors, and assets). l6Some LGs' estimatedoutsandingarrears is inthe tune o f approx. 10-20% o fthere total budget, i.e. approx. one year of own revenue sources. - 80 - 5.39 Typical challenges LGs face in keeping the books o f accounts include problems with reconciliation, mixing of various ledgers, missing signatures, and poor dating for posting. These flows in bookkeeping details17 seem not to be captured in the audit reports, various assessments, or both. There i s evidence that the maintenance o f books o f account by accountants within sector departments i s more timely than by the finance department centrally. 5.40 LGs runtoo many bank accounts. The numberranges from about 30 to more than 100 in different LGs. Funds are banked daily (or weekly in subcounties), and there is generally bank reconciliation every month, although the control i s not optimal in some o f the LGs. The large number of bank accounts imply a significant cost for the LGs inboth financial and other transaction costs. The FDS proposal, which intends to address these capacity concerns at the LG level, i s therefore highly welcomed. 5.41 Record-keeping and security o f accounting documents are important areas that need serious attention in LGs. In most LGs, there are generally no procedures for safeguarding records from fire, water, other environmental risks, and unauthorized access. No one i s responsible for management o f financial records assigned to a particular officer. The office space i s normally not enough for storage o f the volume o f records. Invariably, LGs have inadequate space and facilities for storage o f paper and electronic financial records (fireproof safes and adequate backup procedures for highly important accounting materials). There i s hardly any investment by LGs in records management (staffing, training, equipment, storage facilities) and, above all, no procedure for destroying expiredrecords. Furthermore, there are no registers or overview o f documents. The yearly board of survey, which i s inadequate in many respects, i s the only control initiative so far. Although LGs are aware o fthe challenges confronting them inthis regard, hardly anything is plannedto improve the situation. 5.42 Except in a few LGs, information technology i s used only for word processing. All accounting i s done manually.18 In addition, the computer skills of personnel in the finance departments vary greatly across LGs. Generally, three to five staff have basic computer knowledge. Introduction o f a computerized system o f accounting has shown that there i s a great need for follow-up, on-the-job assistance, and adjustment o f systems to the needs o f the client (system, personnel, legal framework, reporting requirements, and so on). This has hitherto often been neglected, and the important lessons learned could be usedfor IFMS implementation. Deviations from Budget 5.43 There are considerable deviations betweenbudget outturn and budgetedamounts, arising mainly from overoptimistic budgeting for LGs' own revenues. The weak relationship between cash flow management and the budget largely explains the The ReviewTeamwent through all bookso f accountsinthe FinanceDepartment,Headquarterinall four sample LGsand discoveredmany flaws, which were not addressedinthe relatedaccountability control reports(IA, PAC, OAGreports, etc.). A computer packagewas introduced lastyear inMbararaDistrict, butthe FDwas not able to use it due to lack oftraining, fine-tuning of the programand adjustmentto the needofthe District. - 81 - deviations. The LGFAR procedures on authorization o f virements, reallocations, and supplementary budget changes are generally understood and adhered to, although all LGs have problems in reconciliation with the vote books and update o f the posting when changes are made inthe appropriations. 5.44 By locating vote books in the finance departments o f LGs and allowing the finance departments to control the vote books, HODShave lost control o f their votes. They get only an overview o f the changes inthe budget against the votes and the work plans. The LGs lack defined procedures for updating the books when virements and reallocations are decided upon. Finally, the LG virements, reallocations, and supplementary provisions are not reflected and explained in the monthly financial statement to the councils (LGparliaments). Debt Management 5.45 LGs have hardly any bank loans but have considerable outstanding payments and credits. The control and overview o f the outstanding payments are weak in most LGs. Creditor registers and ledgers are sometimes missing, with only a list o f creditors that i s usually not updated. When in place, creditor ledgers are often not updated. LGs generally do not have procedures in place for accounting and controlling the number and size o f creditors. 5.46 Authority to make decisions on prioritization o f the outstanding payments varies from the CFO, CFO/CAO to LGs, which involve the executive committee. All LGs seem to lack a systematic policy and procedures in place for the payment o f creditors. The situation underlines the need for a stronger system and guidelines on commitment control. 5.47 Although specific figures are not available, most LGs have arrears inthe range o f USh 100-300 million, equivalent to 20-40 percent of their yearly own revenues. There appears to be hardly any strategy or apparent provision in the LG budgets for the clearance o f arrears. 5.48 The sections that follow look at reporting on and verification o f expendituresand results. Issues addressed include a comparison betweenupward and downward reporting by LGs, political interference in reporting and verification, internal audit, external audit, preparation o f final accounts, external factors influencing reporting, and other financial management-related issues. H. PERFORMANCE MEASUREMENT AND VERTICAL REPORTING WITHIN THE LG ADMINISTRATION 5.49 Overall, the health and water and sanitation sectors appear to have the most systematic techniques o f performance measurement, through their Management Information Systems (MISS). In other sectors, such as education, performance measurement and reporting vary considerably. Some LGs have developed innovative ways o f performance measurement and reporting. For example, in Masindi district, the education department has the most outstanding system o f performance measurement, - 82 - which involves a scoring system carried out by inspectors for head teacher performance. This i s related to other performance indicators, such as examination results and number o f teachers and facilities. All this i s fed into a computerized database and used as a means o f targeting inspections toward poorly performing schools. However, there appear to be no operational management information systems in place, even in the Masindi education department. Most sectors relied on more haphazard methods o f information capture. 5.50 Some o f the LG HODs provide monthly general reports to the CAO, but the information i s not structured and there i s no system o fverification, control, and follow-up on these reports. In most instances, the LLGs and service providers do not provide performance-related information to the Higher Local Government (HLGs). The quarterly reports to the central government , which are transferred through the CAO, provide the main management information on activities and outputs. 5.51 There i s a needto streamline the reporting from the LG to the central government and introduce more structured internal reporting to improve the use o f information, analysis, and follow-up. Reportingto Politicians 5.52 It appears that politicians are taking an increasingly proactive role in monitoring performance at the LG level, and they are demanding performance reports. The CFO prepares and submits monthly financial statements to the executive in all LGs. However, it is not possible to relate the monthly financial statements to the changes in budgetary allocations (that is, adjustments o f the vote books). In some LGs, the monthly financial statements prepared by the CFO form the basis for updating the vote books and for the monthlydiscussion o fthe budgetby the LGexecutive. 5.53 Most o f the information from the sectors i s transmitted through the sector committees (of council), which sit four to eight times a year. Sector committee meetings are usually held quarterly to consider the quarterly reports to the central government . Because o f the technical nature o f the reports, committees normally request a narrative report from the technical staff o f LGs. 5.54 The council receives the quarterly internal audit reports. In most LGs, the councils also receive the regular quarterly reports from the PAC. 5.55 Lower-level reporting and accountability remains weak, and there i s hardly any formal reporting from the administration to lower-level councils. ExternalReportingto the CentralGovernmentandDonors 5.56 The reports from the LGs to the central government are prepared by the HODs and signed quarterly by the CFO and CAO. The number o f quarterly reports has increased over the past three years and i s in the range o f 10-20 reports. In addition, the LGs submit monthly financial statements to the MFPED. The LGs complain about the hightransaction costs o f preparing and presenting elaborate reports. Many reports have - 83 - to be preparedand deliveredto the center, andmost of them have to be deliveredby the HODSto ensure that they are not lost inthe machineryinthe LMs. 5.57 Often, analysis andfollow-uponthe reportsis of limitedvalue comparedwith the time andresources investedinreporting. Althoughsome LGs are moreusedto preparing the comprehensive and time-consuming reports and documents, the many reporting systems, each with different modality and guidelines, pose a serious challenge to most LGs. Reportingto the Beneficiariesand CivilSociety Organizations 5.58 Reportingto beneficiaries appearedto be the weakest point in the LG reporting system. Apart from the notices in some grant schemes of the size of transfers received, there were hardly any other established mechanisms and practices for reporting to the beneficiariesonfinancial matters, output, activities, andperformancemeasures. 5.59 To a varyingdegree, the citizens couldtake part inthe yearly budget conferences, but apart from this opportunity for a selected few, day-to-day contact and exchange of informationwith citizens i s limited. Some of the LGs have started experimenting with new means of communication, such as radio programs-a promising initiative that is appreciated by the NGOs. Other positive initiatives includethe establishmentof project management committees under the LGDP, where citizens are requested to supervise investments. 5.60 NGOs are generally not involvedinmonitoringand supervisingLGperformance, but rather in the implementationof programs and service delivery activities. Limited downwardaccountabilityis an importantarea of concernwhen elaboratingnewreporting systems. MonitoringandAnalyzingPerformance 5.61 There is generally no system in place for the comparison of performance across sectors. Monitoring of achievement o f outputs against the work plans i s hardly done. There has been a tendency to invest a lot of resources in the preparation of detailed reports (paper accountability)instead of usingthe datafor analyses of people's welfare or needs, A databank storage system, which is being developed in the Ministry of Local Government,will therefore fill an importantgap inall LGs. 5.62 Involvement of beneficiaries in the management and monitoring of service delivery varies across sectors. School management committees are the most effective groups because the parent-teacherrelationshipis the strongest,partlybecause of frequent meetings. In other sectors, especially health and roads and works, it has been more difficult to maintainuser committees, becausethe relationshipbetweenservice providers and beneficiaries is more sporadic (for example, you visit a health center only when you are ill,not every day). Internal Audit 5.63 Internal audit in LGs has improved over the past three years, although the situation i s far from optimal. Most LGs have internal audit units in place with 4 to 10 auditors inthe unit, some o f them with professional accounting skills. The internal audit units are more up to date with the audit reports and control than they were three years ago, but they are typically half a year behind the schedule." Critical challenges internal audit units face include (a) Lack o f facilitation (for example, no means o f transport), (b) limited funds for field visits to LLGs and institutions, (c) lack o f computers and secretarial support, and (d) budget cuts. Inaddition, internal audit units are not considered to be independentor strong inthe internal institutional arrangement-that is, they are not a separate directorate or sector in the LG setup. As such, their power becomes compromised. The internal audit units are mostly uninformed about new initiatives: they lack training in the multiple new guidelines on conditional grants. Finally, district councils hardly consider or discuss internal audit reports and very rarely take actions on the basis o f the reports. Similarly, other stakeholders-such as the Ministry o f Local Government, the MFPED, and the OAG-hardly use the internal audit reports. 5.64 With the existing staff positions, the mandate o f the internal audit units needs to be reviewed with a view to establishing whether they should cover the LLGs and the service providers (schools, health centers, and so on) and be a major factor in internal control. Internal audit units' work i s predominantly pre-audit. Post-audit-based on sample checks, spot visits to critical areas o f possible fiduciary risk, or both-is hardly done. The LLGs are scarcely covered by internal audit. Internal auditors visit subcounties two times a year at most, and their visits have barely any consequence on the way finances are managed inLLGs. 5.65 Normally, funding of internal audit i s often inadequate and irregular. This i s mainly because internal audit units rely on local revenue for operational funding (there i s some limited contribution from PAF). This means that they are at the mercy of the decisions made in the monthly cash flow meetings, where they must lobby for funds alongside sector departments. If the internal audit units are to play a role in financial management and control, and in mitigation o f risks, they need to be better facilitated- their status inLGs enhanced and the internal relationship within the LG structure spelled out more clearly.20 Local GovernmentPublicAccounts Committee (LGPAC) 5.66 LGs have functioning LGPACs. These committees review the internal audit reports and the reports from the OAG, and they may render support to special cases-for example, in cooperation with the Inspectorate General o f Government (IGG). Like the internal audit units, LGPACs are poorly facilitated. Although LGs are expected to contribute to supplement the PAF monitoring and evaluation (M&E) funding that See e.g. Public ManagementIssues,May 2000, Danida,Noradand SIDA. 2oThe Study by MOPSof the fbture organizationalstructures in LGs has suggestedthat the CIA shouldbe downgradedfrom U1to U2. This will be very inappropriateandunderminethe alreadyweak authority o f the IA. LGPACs receive, LGs make hardly any payment. Furthermore, PAC reports are scarcely considered by LG councils and, in some instances, some LG councils do not receive a copy o f the PAC reports. There are very few instances o f any reactions and punitive sanctions taken based on the reports from the LGPAC. 5.67 The LGPACs do not have the capacity to carry out detailed analysis o f the actual financial management practices, field visits, and so forth. Consequently, they rely entirely on internal audit reports, which are o f low quality, writtenon pre-audit basis, and lacking inmany other respects. Inmost LGs there i s a problem o f backlog and delays in producing LGPAC reports. Time lags inproducing these reports limit their relevance. It i s absurd that councils deal with matters o f financial management that are as far back as three or four years, instead o f considering the most current issues. ExternalAudit 5.68 External audit reports o f the HLGs are produced for every financial year, but they are prepared with time lags of two to three years. Delays in audit are caused by the procedures between the submission o f draft final accounts and the approval o f final accounts, followed by the audit report. The LG submits a draft account for preliminary review by the regional OAG, then the LG makes comments on a note from the OAG about outstanding issues and concerns. Lack o f transparency characterizes the process. Production o f final accounts and audit reports may be delayed, especially if there are long discussions and negotiations between concerned parties. The councils and the control bodies see only the final outcome o f the OAG's deliberations, not the original problems encountered when the draft report was available. A lot o f time may be spent on retrievingvouchers, getting bank statements, reconciliation, and so forth. External audit i s an area that needs review, paying particular attention to linkages with the internal audit work and filling ingaps-especially inmanagement auditing. ParliamentLGPAC andOther ControlMeasures 5.69 The national parliament has constitutedan LGPAC to look into issues o f financial management in LGs. Stakeholders across many LGs appreciate this development. However, there i s concern that in scrutinizing the LG accounts, parliamentarians treat CAOs o f LGs where problems are found in an overzealous manner. The long-term impact o f the parliamentary LGPAC i s questionable because where problems are found, there are hardly any sanctions except for mere threats to management o f LGs. Furthermore, LG council members are elected, just like members o f the national parliament, and they needto address the question o f social accountability. 5.70 Generally, the control measures and authorities-internal audit, LGPAC, OAG, Ministry of Local Government, MFPED, IGG, line ministries (LMs), and so on-are not working optimally. Their activities need to be synchronized and streamlined interms o f inspection, monitoring, review and analysis, reporting and accountability, follow-up, and, ifnecessary, sanctions. FinalAccounts 5.71 All LGsprepare audited accounts for every financial year, with a time lag of one to two years. The Office o f the Auditor General makes comments on the audited books to produce final accounts, or audited books are submitted to the OAG to prepare final accounts. The delays (typically two to four months) have a negative impact on the timing o f the work o f the OAG and impede the whole process o f accountability and information to the council and other stakeholders about the financial management status o f an LG. However, the time lags have been lessening over time, and by regional standards, they are rather good.21 5.72 The situation at the level of LLGs is more problematic. Fewer than half o f the subcounties have submitted any accounts to the OAG over the past three years. In four sample LGs in mid-2002, only approximately 110 o f 3,000 subcounty accounts were audited for the past three years (1,000 accounts multipliedby three years). This situation has improved with support from the LGDP for production o f accounts and audit, but there i s still a long way to go before the backlog o f all these accounts i s cleared. 5.73 The next sections look at the factors affecting financial management, such as transparency and accountability, institutional factors, political interference, and other external factors that influence financial management inLGs. Transparencyand Accountability in FinancialManagement 5.74 Inadequate transparency and accountability are both a symptom o f weaknesses in financial management performance and a cause o f those weaknesses. Generally, there i s still very weak downward accountability in most LGs. First, none o f the LGs publish or display their plans, budgets, work plans, accounts, and audit reports for citizens, NGOs, and civil society organizations, and there are no signs of a well-developed information and participation strategy in the LGs. The involvement o f citizens in the planning and budgeting process has improved only slightly and is still limited to the yearly budget conferences, which are rather formal and provide only limited opportunity for discussions. 5.75 Citizens' involvement in control and monitoring activities hardly exists. However, there are important new initiatives that could be built upon or developed further. LGs have started involving citizens in the project management committees around the LGDP investments, and there are many scattered activities indifferent sectors to involve the communities in service delivery (with varying success). There are also other signs o f improvements, such as public notice o f the LGDP grants and other development grants on the notice boards o f the administration building. These, however, are usually central government requirementsrather than LG initiatives. 5.76 Finally, some LGs have started information dissemination and exchange o f views on new developments in the LGs through local radio programs. Compared with the OAG is working on an update of the status, but an update from July 2002 show that out of 129 HLGs only 4 had no audited accounts for 2000101. - 87 - efforts made in compliance with the central government requirements and the reporting systems, upward reporting and accountability still seem to be the prevalent force and decisive factor. Institutional,Management,and PoliticalFactors 5.77 The many layers o f LGs and the large number o f LLGs pose a great challenge to financial management capacity, and they tend to fragment skills and intemal leaming. Revenue sharing among different layers o f LGs i s especially an area o f intrinsic conflict: different layers o fLGs fight over limitedLG own revenue sources. 5.78 In the HLGs, the organizational culture sometimes has a negative impact on performance, especially lack o f performance monitoring, incentives, rotation o f staff, delegation o f responsibility (bureaucracy), too much focus on hierarchy and top-down management, and lack o f vertical and horizontal information exchange. Together with low salaries for the key people in the finance departments-who have little access to benefits (as for the HODS)interms o f allowances, travel, and so on-poor organizational culture may negatively affect workers' morale, 5.79 There is a mutual mistrust betweenthe political and administrative levels inmost LGs and a tendency for the politicians to focus too much on narrow personal interests instead o f service provision and compliance with the rules. 5.80 Capacity building has focused too much on general accounting and budgeting skills, typically for private business accounting, and to a lesser degree on targeted, practical exchange o f experiences in financial management and targeted LG courses in key areas o f financial management. Numerous workshops with little relevance to staff performance have taken up too much staff time. 5.81 Although the financial management capacity i s to a varying degree available in terms o f staffing and formal skills at the HLG level, there i s a need to transform these skills into LG financial management practices and internal learning processes. Typically a district has between 20 and 30 staff in the FD (some o f them with professional accounting skills) at the headquarters and one accountant in each subcounty. There i s a problem o f lack o f accounting capacity at the level o f the sectors, because many functions require intemal decentralization in the accounting functions. Municipalities seem to be much less endowed with capacity, and the LCIII level (both urban and rural) i s still very weak interms o f financial management capacity. 5.82 The management culture inLGs has an adverse impact on the performance o f the financial management functions. New developments-such as rotation, intemal promotion, review o f job performance, facilitation in terms o f new tools (computer software and hardware), clear demarcationo f responsibilities, delegation, field visits, and monitoring-would have a positive effect on the performance o f the staff. - 88 - ExternalFactors 5.83 On the whole, the financial and accounting regulations are appreciated and followed, although there are provisions inthe LGFAR that are hardly complied with (for example, control o f commitments). 5.84 There are problems in the relationship with the central government, including changes in the conditional grant Indicative Planning Figures (IPFs) during the budget process, delays in grant transfers, and the irregular amount o f funds transferred in each release. These have an adverse effect on LGs' performance, especially in planning, budgeting, and budget execution. Often, funds are released two or three months behind schedule. However, some o f the delays are a result o f nonreporting or late reporting by LGs. Delays make it nearly impossible to follow the work plans, complete the work, and ensure sufficient cash inflow. 5.85 Multiple reporting requirements and different modalities across the sectors, as well as multiple bank accounts and district accounts, have a negative impact on the financial management performance in LGs. Implementation o f the FDS recommendations offers hope for improvement. The multiple accounts (50 to 100 inthe typical district) make it hard to get an overview, control, and follow up on each account and thereby reduce the overall accountability. Better coordination o f the various ministries in terms of reporting systems, workshop arrangements, production o f guidelines, and training would reduce the transactions costs and ensure a coherent and coordinated development and also reduce the number o f worthless workshops. 5.86 The multiple conditional grants have had both positive and negative impacts on the financial management performance. On one hand, they have ensured that the key sectors receive the expected share o f the resources, inthe absence o f effective cash flow management, fostered a culture o f accountability (although this has beenmostly upward), and improved the financial management control o f a large share o f the funds spent at the LG level. On the other hand, the rigid earmarking of funds and the many grants and budget lines have made budgeting and budget execution inflexible and undermined performance inareas that are supposed to be covered by own revenue sources. Often the funds would have beenspent more efficiently and better adjusted to local needs iftransfer betweensectors andbudgetlines hadbeenallowed duringbudgetpreparation. 5.87 The over determination o f the budget priorities by the central government may also be one o f the reasons why the citizens, NGOs, and civil society organizations have shown little interest in participating in the planning and budgeting process, because the room for maneuvering i s very limited. As much as 80 percent o f LGs' expenditures are financed and determinedby earmarked conditional grants. Improvement in own revenue sources and increased flexibility in the use o f the conditional grants are therefore important tools for improved financial management practices, as envisaged inthe FDS. 5.88 This chapter i s based on a study o f these issues in three districts and one municipality. The study was carried out this year as part o f the PER. The results illustrate the heterogeneity o f the LG budget and financial management performance and - 89 - challenges: to make progress in these areas in the context o f supporting decentralization in Uganda, there is a need to benchmark and assist LGs accordingly. The study also shows that there i s lot of scope for learning from others within Uganda. In conclusion, one could say that Uganda's LGs face enormous budget and financial management challenges, which could be improved with appropriate assistance. LGs needassistanceto improve budget processes-budget allocation, execution, efficiency, and financial management. - 90 - 6. CONCLUSIONSAND RECOMMENDATIONS A. INTRODUCTION 6.1 This year's PER has attempted to address the challenges in the budget process, budget execution, and budget efficiency at the central government level but to a greater extent at the LG level. As outlined inthe concept paper, the PER has identified critical issues surrounding the budget process, progress as well as remaining challenges, and it has also analyzed the underlyingfactors. Given that the credibility o f the budget process will depend on how progress made in upstream processes translates into downstream activities, this PER has focused on these issues at each stage o f the budget process. This has also beendiscussed inthe context o f changing aid modalities and the implications for future effective use o fpublic resources. The underlying rationale is that untilthese issues are understood and addressed, it will be difficult for budgets to make a useful impact on the growth andpoverty reduction challenges that Ugandafaces. B. MACROECONOMIC PERFORMANCE 6.2 This chapter provided a synopsis o f Uganda's recent economic performance, reviewing the overall macroeconomic stability o f the economy, as well as prospects for future growth. Selected macroeconomic indicators were usedto review developments in the fiscal and monetary sides of the economy. The chapter identified three main challenges facing Uganda's overall economic performance: the slowdown in real GDP growth, management o f the rate o f inflation, and the fiscal deficit. 6.3 Uganda's economy i s currently facing a growth slowdown. The real GDP growth rate fell below the 5 percent mark in 2002/03 for the first time inmore than 17 years. O n a positive note, real GDP growth was positive and quite significant this fiscal year, and, unlike inprevious years, the slowdown comes at a time of a general global growth slump. Yet there is concern that inthe worst case scenario, lower GDP growth rates infiscal year 2002/03 represent a downward trend that would prevent Uganda from reaching its macroeconomic and poverty reduction goals. For Uganda, the question now i s how much its decline in economic growth i s due to a larger global economic slowdown or whether, in fact, the steam is running out of Uganda's growth engine. Discussions in the upcoming PEAP revision will need to address this question as well as what policy realignments will be necessary to ensure a rebound inreal GDP growth. 6.4 Although inflation rates have largely been kept under control, there has been a significant amount o f volatility this fiscal year, in both headline inflation rates, resulting from unstable fuel prices, and underlyinginflation due to volatility infood crop prices. 6.5 The issue o f the fiscal deficit, which peaked at 13 percent of GDP in 2001/02, is now a focus o f economic policy because o f the threat posed by the size o f the deficit to - 91 - both monetary and fiscal policies. There are concerns about the impact o f the deficit on the macroeconomy, especially because the deficit is being funded primarily through donor aid, which has resultedin excess liquidity and higher interest rates in addition to a crunch on private sector credit and an appreciation o f the real exchange rate. 6.6 On the fiscal side, there are questions about the sustainability o f public finances, because growth in government revenues has consistently lagged behind growth inpublic expenditures. In2002/03, 53 percent of the government's budget was funded by donor aid, and there are legitimate concerns about the vulnerability o f the government budget to sudden cutbacks indonor aid. c. BUDGETPROCESS, MEDIUM-TERM BUDGET FRAMEWORK, MTEF 6.7 This chapter on the budget process provided an update for the BFP preparation process for 2003/04. It commented on the progress made inphases 1 and 2 o f the budget process and clearly shows that phase 1 has become more organized and more time was given to this process than last year. The phase 2 external evaluation process also has become more systematic. The parliament has become very actively involved in the budget process. In addition, there i s a synergy developing between local processes and donor interaction on the budget, which needs to be harnessed inthe coming year. 6.8 It is expected that the World Bank and other budget support donors will assist Uganda inimproving the budget process inthe mediumterm. World Bank areas o f focus will include the efficiency and equitable allocation o f resources, as well as improved financial management at the central and LG levels through PRSC programs and the LGDP. 6.9 As pointed out in last year's PER report, the PEAP i s Uganda's broad development framework. The PEAP i s now in the process o f revision. The World Bank i s keeping up with its commitment to the government to support the process o f PEAP revision in PEAP target setting, sequencing, and costing actions. SWGs will use this information to realize budget linkages and PEAPtargets. 6.10 The World Bank also needs to support the whole budget process, starting with initial determination o f budget ceilings by the MFPED, BFP preparation by SWGs and local government, consultation through various forums (including the PER), and finally the MTEF and budget preparation. Although many o f the reforms to strengthen the budget process are continuing, the progress has definitely been slow. This is mainly a result o f technical capacity needs at various levels that the World Bank can help to fill. 6.11 Furthermore, other participants in the budget process, particularly civil society and parliament, are still facing technical capacity needs that the World Bank could improve through funding. Although the participation o f parliamentarians has improved, more improvement i s needed for truly effective participation in the budget process. Training o f some parliamentarians should enhance their participation inthe budget cycle. 6.12 Recent reviews o f the PAF show that its protection has been interpreted to mean protecting social service delivery sectors. It has increasingly become clear that activities - 92 - o f some nonprotected areas are critical to the realization o f PAF objectives. Hence, it i s necessary to rework government expenditure priorities with a view to ensuring that PAF objectives are realizable. Some o f these concerns should be addressed in the revision o f the PEAP, which the World Bank and other donors are supporting. D. BUDGET EXECUTION 6.13 This chapter set out to understand the issues inbudget execution as they occurred this year. There were some unforeseen budgetpressures early inthe fiscal year, and this naturally created substantial instability inbudget management and program execution for many sectors. The review clearly demonstrated the indiscipline that i s beginning to characterize Uganda's budget execution. Critical dimensions o f indiscipline that have emerged include budget overruns by some sectors, irregular flow o f funds to some sectors, failure to pay arrears on time, insufficient government counterpart funds to finance donor-supported projects, and underprogramming that led to insufficient resources for domestic interest payments and arrears payments. 6.14 The trend o f using supplementaries for defense and PA, which has plagued the Uganda budget system for the last few years, i s a matter o f serious concern because it leads to budget cuts in other activities that are critical to the realization o f poverty reduction and development objectives. The implications o f these developments for the overall credibility o f the budget process and execution should be given due importance. Many o f these problems are rooted in political interference in budgeting, planning, and execution. 6.15 Ensuring the timely flow of funds to the various sectors is a challenge. The tendency to frontload expenditures o f some sectors while others are starved o f funds should be avoided. There may be a needto review the cash budget management system so it does not become detrimental to budgetexecution. 6.16 Budget overruns by some sectors are the root cause o f budgetary problems experienced by other sectors starved o f financial resources. Budget overruns distort the budget and make it irrelevant. If the budget i s to remain credible, overruns should be avoided. Budget overruns come about as a result o f either underprogramming or sheer indiscipline. This problem needs immediate attention if budget preparation and execution are to remain trustworthy. 6.17 The implementation o f the Public Finance and Accountability Act (2003) will be important in arresting the past trends. It is intended to limit the incidence of supplementaries by requiring parliamentary approval before the release o f any supplementary funds. This should help contain overexpenditures o f some sectors and lead to a balanced execution o f the budget. However, with the objective o f deficit reduction and an overall underfunding o f most activities, it will be very difficult to manage the culture o f supplementaries. 6.18 Full implementation of the CCS is important to cover all central and LG expenditures. Currently, the CCS covers only the central government's nonwage and - 93 - development budgets. Arrears could be minimized on all accounts ifthe CCS i s extended to all areas o f government budget. Also, payment o f arrears should be a high priority. Arrears reflect services rendered and not paid for, and failure to clear them adversely affects performance o f the various sectors-especially in the case of utilities, as was experienced this year. 6.19 The government is committed to reducing the fiscal deficit, and there i s likely to be even greater pressure on budget execution because many more sectors and programs are likely to have underbudgeting problems. If adequate checks and balances are not designed and implemented, there i s a possibility that some o f the problems may intensify inthe mediumterm. These budget executionpressures are likely to remainan important area o f concern for budget support donors and could undermine the Ugandan government's desire to deepenbudget support modality. E. BUDGETEFFICIENCY 6.20 This chapter underscored the importance o f budget efficiency as resources are becoming increasingly scarce. As a result o f the drive to meet national development goals, the government budget i s faced with considerable pressure from escalating, competing, and overwhelmingly legitimate demands for the same resources across sectors. Individual sectors inthe economy can meet the challenge o f scarce resources by increasing the effectiveness with which they use current resources. For example, by reducing leakages, tightening loopholes, and reducing unit costs, sectors can create more outputs and outcomes with a constant amount of inputs. 6.21 The chapter tracks the progress o f four sectors-water and sanitation, health, education, and justice, law, and order-in institutionalizing budget efficiency. The sectors are clearly at different stages of this process. Inthe education and health sectors, there i s a longer history o f tracking their own progress. Biannual health and education sector reviews have become an opportunity to confront efficiency issues raised by sector audits, reports, and tracking studies and integrate these issues into strategic action plans. Inthese two sectors, the critical issue is now the effective implementation and follow-up o f recommendations from the various studies and reports. Inthe water and sanitation, as well as justice, law, and order, sectors, the sectoral processes for assessing budget efficiency are much more recent and a culture i s not yet established. However, there have been some recent developments in the form o f tracking studies in both sectors, as well as a high-levelfocus onthe issue. 6.22 There is still a needto increase the allocative efficiency o f resources inthe water and sanitation sector so that resources are better targeted at PEAP objectives and also more equitably allocated among the subsectors. Concentration o f resources remains at the center, even though the majority o f the population is located in rural areas. Small towns are also still consuming an inequitably large share o f resources. In addition, despite increases in water coverage, the problem o f infant mortality-which has been linked to the cross-performance o f a number o f sectors, including water and sanitation- highlightsthe needto increase the impact o f expenditures on outcomes. - 94 - 6.23 The water and sanitation sector has experienced encouraging progress in addressing operational efficiency. The first comprehensive value-for-money study was completed inthe sector, and its results formed the basis for discussions inboth the sector review meeting in September 2002 and the PER meeting in May 2003. This should be made an ongoing process, to improve the methodologies used inthe study and make the results more useful for policy changes, as well as to integrate the results o f the efficiency studiesinsector policy. 6.24 The health sector i s also confronting the issue of negative trends in infant and maternal mortality rates in addition to the stagnating HIV/AIDS prevalence rate. Because o f the multidimensional nature o f these indicators, there i s a need to strengthen the cross-sectoral approach to addressingtheir poor performance. Specifically, the health sector needs to increase its coordination with the education and water and sanitation sectors. 6.25 There is an increased focus on improving the poverty targeting of health expenditures because it is still unclear (in the sector BFP) how the poor and vulnerable benefit. The ongoing district health PER will provide some answers by analyzing the flow o f funds from the center to the district and subdistrict levels, as well as the efficiency o f healthexpenditures at the district level. 6.26 In the education sector, there is concern about the competing needs o f the subsectors, especially the growing demand for secondary education as a direct result o f the UPE program. There is obviously a needto share resources more equitably between the primary and postprimary subsectors, but a dramatic short-run reallocation across subsectors i s unlikely. Thus, greater immediate emphasis on increasing efficiency i s needed, not only in the primary subsector, but also in the postprimary subsectors. A culture o f expendituretracking and addressing efficiency issues exists within the sector as a whole, and the challenge i s to ensure that the results and recommendations made by the various tracking studies are implementedand followed up. Currently, biannual education sector reviews provide an opportunity to address the issues raised in the tracking exercises; however, a systematic stock taking o f progress made on prior recommendations should also be done to increase the sector's accountability for these reforms. 6.27 Thejustice, law, and order sector has brought the issue o f budget efficiency to the forefront by focusing on financial management within its core institutions. A baseline study o f financial management in seven o f its nine institutions, which evaluated the sector's capacity to efficiently use its resources, was recently completed. The study identified a high degree of fiduciary risk within the individual institutions, raising concerns about the opportunities this presents for mismanagement o f resources. There i s also a crucial need for the sector to embark urgently on implementing some o f the practical recommendations made by the report to reduce the level o f the fiduciary risk andavoid unnecessary leakages o fresources inthe short run. -`95 - F. LG BUDGET AND FINANCIAL MANAGEMENT 6.28 This year, the PER wanted to give more than average attention to LG issues because o f the budgetary challenges at the local level that are especially related to decentralization. There is a need for greater planning, participation, and budget and financial management, which are all closely linked at the LG level. Ultimately, budget management determines the development outcomes o f the various stakeholders involved in public service delivery at the LG level. This chapter discussed the intemal and extemal factors affecting LGs' financial and budget management in detail, using an in- depthstudy that was carried out inthree districts and one municipality this year as part of the PER. 6.29 The chapter highlightedthree core problems facing the country's decentralization efforts, First, the process up to now is only partial, and critical elements of decentralization, such as resource allocation autonomy and strong local institutions, have yet to be implemented. Second, there are still severe capacity constraints at the LG level. Third, there are problems with efficiency as a result of a distorted incentives structure at the LGlevel (upward rather than downward). 6.30 The discussions demonstrated that poor financial management, although a general state, i s not uniform across all LGs. The chapter highlighted seven factors that affect budgetary and financial management: political leadership performance; administrative leadership performance; central government performance; systems, tools, and procedures; legal framework; organizational structure; and staff number and performance. Problems infinancial management may occur ifan imbalance betweentwo areas arises as a result o f a conflict betweenpoliticians and the management group. The financial management problems will inthis case be reflected inthe political and administrative leaders' inability to initiate and handle necessary changes in the organization because o f internal or extemal influences. 6.31 Internal factors likely to affect financial performance adversely are dysfunctional political leadership; dysfunctional administrative leadership; unclear or illogical organizational setup; ineffective management systems, tools, and procedures; and inadequate staff number and performance. Political disagreement and personal conflicts among politicians may lead to lack o f accountability and transparency in the budgeting process and consequently poor budget performance, because no one feels any responsibility for the implementation or follow-up. Political misconduct and mismanagement also encompass corruption, which generates political and administrative inefficiency and ineffectiveness and undermines the legitimacy o f political leaders and political institutions. Corruption might also lead to financial crises because o f its misallocation and waste o f resources. In general, one can argue that partial implementation o f the public leadership code may cause dysfunction in the organization, which as a consequence may leadto poor financial management at the LG level. 6.32 Poor management and action paralysis, or even mismanagement, within the management group could arise from misconduct and mismanagement by the politicians. In such a situation, improvement in budget and financial management may depend - 96 - heavily on changes in the political leadership. However, the study shows that poor budget and financial management could also arise from incompetence within the management group. 6.33 LGs face many challenges in improving budget processes: budget allocation, execution, efficiency, and financial management. The results o f the study illustrate the heterogeneity o f the LG budget and financial management performance and challenges, as well as the need to benchmark and assist LGs accordingly to make progress in these areas inthe context o f supporting decentralization inUganda. It also shows that there i s a lot o f scope for learning from others within Uganda. LG budget and financial management i s the focus o f the medium-termbudgetary reforms inUganda. - 97 - ANNEXES - 98 - e M r3 I I a r. PI 0 m 2 x "l x vi z0 I * 0 c 0 "l x d4 r I I I 3 2 I 0 erm- v ioo o - e s - 0 0 : O P N O O aEE.l:l I 0 M M I :::e% 8 8 8 8 8:: $8ci - 0 - - 0 0 0 0 0 0 ooIm m 01 P 01 -2=s m 10 m - P m r - 01 - N P 010 0 0 1 0 0 0 = 0 0 0 7 0 mr. g p 0 o m 0 0 0 f - p0 : 8 Z X 0 0 0 r. m 10 r- 2 0 x z s7"a 0 - m v l o o m 0 0 - 0 P 8 R , % 4 A 1 8 ;S 0100 - O - - O O N * 0 c P O 5 - r- o m m 8 m m I " 000 0 - 8 8W;ZaWg W 0 10 2 8 0100 - O - - O O N P 0 x 10 P O d 0 3 I a 10ha x10 a10 0' r- a c - N m - - N x 01 8: o w x 0 60 X 0 8 n W N -:: 0110 m vml 0 0 - w m ANNEX3: STATEMENTBY THEWORLDBANK,ONBEHALFOFDEVELOPMENT PARTNERS, ONBUDGET PERFORMANCE,2002103 INTRODUCTION We, on behalf of all development partners present, would like to thank the Permanent Secretary/ Secretary to the Treasury for his detailed presentation on performance o f the current year's budget and an exposition o f how they met the challenges they faced. At a time when global economic trends have beenunfavorable and the increases infuel prices have resulted in economy-wide price increases, one has to commend Uganda's budget and macroeconomic performance over this year. The growth rate, fiscal deficit, and inflation targets are slightly off the target for this year. The government has, however, kept the budget largely on track for the overall PAF expenditures despite pressures for supplementaries from all quarters through the course o f the year. However, some o f the cutbacks in the budget have not been without adverse effects on undertakings o f most PAFand non-PAF sectors, as we will see further below. The budget performance over the current year has highlighted a number o f areas of concern. BUDGETINCREASESFORDEFENSE PUBLICADMINISTRATION AND This year, the budgetwas again affected by nonprogrammed budget increases inthe early part o f the fiscal year from defense (USh 33 billion) and public administration (USh 15 billion). The budget increase for defense was especially damaging, because it breached earlier commitments to limit defense spending, pendingthe outcome o f the defense review, and seriously undermined donor confidence in the transparency and integrity of the budget process. As a consequence, donor budget support releases were reduced by almost US$30 million this year. The cuts made in the budget that were due to the increased defense spending,made worse by overspending on public administration, have resulted in difficulties for other sectors in carrying out their intendedprograms. This was especially the case inroads; health; education; justice, law, and order; and agriculture. Although we commend the MFPED for ensuringthat all PAF budget lines were fully protected, cuts to non-PAF budget have had an adverse effect on many PAF sector undertakings. We urge that the government does not allow this trendto continue inthe future. To ensure transparency and credibility in the budget process for SWGs and development partners, we would like to request the government to provide additional information as to what activities were funded through the unprogrammed increases in the defense and public administration sectors this year. TAXANDNONTAX REVENUES We very much welcome the fact that the URA revenue target will be substantially met this year, though note that this has beenhelpedby favorable exchange rate movements. However, the nontax revenue target will have a USh 15 billion (40 percent) shortfall. -105 - This confirms a pressingneedto address revenue issues through comprehensive reforms and ensure that unprogrammed revenue shortfalls do not adversely affect the Uganda government's objective o f fiscal deficit targets. AVAILABILITY OF COUNTERPART FUNDS This year has seen a number o f project activities in roads; justice, law, and order; and agriculture being delayed and postponed because o f nonavailability o f counterpart funds. This has affected many of the undertakings of sector programs and further undermined the budgetprocess. The shortfall incounterpart funds has also resultedina large number o f commitments that could not be honored, resulting inthe buildup o f arrears and interest payments on many activities, especially inthe roads and works sector. PREDICTABILITY AND TIMELINESSOF BUDGETFLOWS INTRASECTORAL BUDGET AND FLOWS The predictability and timeliness o f transfers to sectors and local governments remains a problem, which seriously undermines project and program planning, as well as the sectors and local governments' confidence in the budgetary process. Even though the problem was exacerbated this year as a result o f the budget overruns in defense and public administration, it i s not a new problem. Inthe first half, expenditureson non-PAF recurrent budget performed at 55 percent o f budget, while non-PAF development performed at 50 percent o f budget. Ifwe do not address this problem, we risk losing the budgetdiscipline and efficiency that we have built up over the years. On a related note, in some sectors, considerable differences persist in the distribution o f resources within the sector itself. The problem might be the most notable injustice, law, and order sector, where the police and Local Defence Units (LDUs) performed much betterthan any o f the other institutions inthe sector. Also, inthe roads and works sector, road maintenance has been especially underfunded. In addition, in the water and sanitation sector, more than 50 percent o f the budget i s devoted to urban areas, with only 15 percent o f the population residing there and more than 80 percent having access. These intrasectoral allocation problems have been accentuated by the irregular budget flows and unprogrammedbudgetcuts. DOMESTIC INTEREST PAYMENT AND STERILIZING LARGE RESOURCE INFLOWS We note the rapidly rising costs o f domestic interest payment, well above its planned allocation (USh 44 billion overrun this year), and recognize that this i s a direct result o f the highfiscal deficit and the need to mop up excess liquidity. The government should reexamine the balance betweenthe various options used for sterilizing its large resource inflows, which would result in issuing fewer treasury bills than otherwise. Given the medium-term fiscal pressures due to revenue shortfalls and expenditure increases, this will be a problem that has to be managed with limited adverse effect on interest rates and the overall economy. - 106 - Low SECTOR EXPENDITUREOUTTURNS The low outtum figures in the agriculture; roads and works; education; health; justice, law, and order; and accountability sectors indicate how the priority reform activities (PAF and non-PAF) have been affected by budget cuts incurredduring the course o f the year. Further details on these are presented in an annex to this statement and also discussed in greater detail in the recently completed education and health sector review aides-memoires. These adverse effects are mainly due to reallocations to defense, strategic exports, interest rate payments, and related cash constraints in the first half o f thisyear, as highlightedearlier. CONCLUSION We development partners welcome the fact that the revisedbudgettargets for 2002/03 on defense, public administration, and PAF spending will be met, as agreed between the government o f Uganda and donors in the PRSC aide-memoire. We also value the improvements in the legislative framework and the commitment to full and frank consultations, and look to the government to ensure that the problems highlightedabove are not repeated infuture years iffurther cuts inaid are to be avoided. - 107- ANNEX4: STATEMENT BYTHEWORLD BANK,ONBEHALFOFTHEDEVELOPMENT PARTNERS, THEMTEF ON 2003/04 To 2005/06 We would like to thank the Director, Economic Affairs, for his clear presentation o f the MTEF and budget for next year. We feel that it clearly articulates the challenges and constraints under which the government has to plan the fiscal program for the coming three years. It also highlightsthe difficulties the government faces in achieving the PEAP targets and MDGs over the medium term. We welcome the protection o f priority expenditures despite the budget increases necessitated by outstanding defense contracts late in the budget process. We note, however, that growth and the revenue targets are probably ambitious. Although we believe that macroeconomic targets and fiscal projections are achievable, the potential for slippages due to in-year budget pressures and external factors i s not to be underestimated. Please allow us to highlight some o f the issues we feel are o f major importance with respect to the MTEF as just presented. DEFENSE SPENDING While welcoming assurances by His Excellency the President that future defense spending, from 2004/05, will be arbitrated by the defense review, we are disappointed at the scale and justification given-to cover outstanding contractual obligations-for the increased defense spending in 2003104. It demonstrates that defense spending has not been kept under control and that past agreements on defense spending have not been kept. This is a serious threat to the budgetary process as a whole. Steps mustbe takento ensure greater accountability for defense expenditures to the responsible authorities. In this context, we very much welcome the fact that the O A G will get access to classified accounts as from the coming financial year, and that defense procurement will be subjected to annual procurement audits. We also feel that defense needs to prepare BFPs, as other ministries do, and that the commitment control system, already functioning reasonably well in other ministries, will be strictly implemented in the Ministry o f Defence. We would like to stress the importance we attach to compliance with the understandings reached in the meeting o f May 6 and as outlined (and to be confirmed) in the correspondences betweenthe MFPED and the Ministry o f Defense for our support for the MTEF. The assurances detailed inthe correspondence, particularly with regardto details o f outstanding commitments and the maintenance o f strict budget discipline, will for some development partners influence both the size and timing o f budget support releases. We welcome the publication of inquiry reports and look forward to their implementation. - 108 - IMPROVEMENTS IN THE BUDGETARY CYCLE We very much welcome the enactment o f the Procurement Act and the recent passing o f the Public Finance and Accountability Bill, which we hope will be enacted soon. Demonstrable improvements inpublic financial management are essential ifdonors are to be able to continue to provide budgetsupport at recent levels. Furthermore, we commend the government's commitment to improve output orientation of the budget and putting additional effort through SWGs to ensure that this does happen. We also welcome the regular publication o f budget performance reports. We hope, too, that further action will be taken to internalize ROM in the budget process, an area that has seen very limited progress inthe last few years. REVENUES As mentioned in our reaction to the budget execution for fiscal year 2002/03, revenue collection (both tax and nontax, at central and local government levels) i s an issue that merits special attention. The funding o f the budget for fiscal year 2003/04 depends substantially on a significant increase in tax revenue. This foreseen increase is considerably bigger than inearlier years, and it depends partly on measures that still have to be agreed to within government. Failure to meet the target could have serious consequences for the fiscal deficit and make in-year adjustments to sector budgets necessary, thereby once again undermining confidence in the budgetary process and making it difficult for donors to support the budget. Therefore, we urge the government to pursue vigorously the reforms for modernization o f the URA, including the swift finalization and implementation o f recommendations o fthe Justice Ssebutinde report. THEFISCAL DEFICIT To accommodate defense obligations while preserving an acceptable level o f growth in the budgets for poverty reduction sectors, we recognize that the original medium-term fiscal deficit reduction targets will not be met. Although donors find it acceptable that the 2003/04 deficit will be maintained at this year's projected level, we would underline our agreement with the Ugandan government's policy to reduce the fiscal deficit in the mediumto long term. Inthis context, greater attention should also be paidto the measure o f the deficit after grants. Fiscal constraints reinforce the need to focus on budget efficiency and appropriate intrasectoral allocations, to ensure that priority activities in each sector are adequately resourced. PAF AND THE SECTORAL DISTRIBUTION THE BUDGET OF We welcome the continued increase inthe share o f the government discretionary budget going to PAF, and that last October's proposed ceilings for priority sectors have been restored. Inaddition to our concerns about defense allocations outlined above, we would urge the government to contain and reduce expenditures on public administration and increase over time the shares going to priority sectors, notably health, education, - 109- accountability, and the judiciary (some o f which are currently not in line with commitments made by the government o f Uganda). INTEGRATION OF PROJECTSINTHE BUDGET We welcome government plans to integrate all donor-funded projects into sector ceilings, to improve budgetary planning and help ensure that all expenditures within each sector, whether funded through the government budget or through external project modalities, are fully consistent with sectoral priorities and the overall fiscal constraints. But we and the SWGs needgreater clarity on how this integration will be managed, particularly with regardto projects that cut across several sectors and the treatment o f technical assistance. We also need clarity on the role of SWGs. We therefore request the government to discuss this issue with development partners as soon as possible and put the mechanism into operation after reaching a consensus among all stakeholders. These agreements need to be consistent with the partnership principles that are currently being finalized between the government ofUgandaanddevelopment partners. STRATEGIC EXPORTS PRIVATE SECTOR DEVELOPMENT AND We support the government's objective of boosting exports and encouraging private sector investment, but we believe that this should be done ina more transparent manner. Inthe last two years, more than USh 100 billionhas beenallocatedto strategic exports, resources that have a high opportunity cost in terms o f achieving other priority PEAP targets and which we fear will provide poor value for money. The government should set out clear principles to guide its interventions in the export sector and avoid handing out selective subsidies (such as cheap loans, tax breaks, subsidized rail freight, and so on) to individual investors or bailing out private companies, however superficially attractive these initiatives may appear. Inthis regard, we would like government to provide a full and transparent accounting for all the funds provided through the budget or through the Uganda Development Bank to private companies, such as those inthe textile and clothing sector. PAY REFORM One o f the important areas that need the government's MTEF focus i s pay reform. We appreciate that next year's budget has made a provision o f USh 25 billion for pay increases to cover both inflation-related rises across the board and additional rises for higher cadre professional and technical personnel. The government needs to ensure that the MOPS and MFPED clearly articulate how they intendto implement the pay reform over the medium term and ensure that this i s properly resourced. We would ask the government to provide a reviewo f its performance inthis area. FISCALDECENTRALIZATION We commend the government on the progress it has made in finalizing its FDS and its initial implementation. We reiterate our commitment to support these initiatives through LGDP2 and other programs and request the government to give greater priority to this issue through its MTEF allocations. As highlightedyesterday, in future, these allocations - 110- shouldbe consideredwhen discussingthe sector ceilings because they directly contribute to many ofthe service delivery activities at the LGlevel. UNFUNDED PRIORITIES We are increasingly concerned that the unfunded priorities list is growing larger each year. This year, it is almost USh 325 billion, and it is extremely difficult to prioritize from such a large list. Infuture, we hope that interministerialconsultations will provide enough space and time to discuss and reducethe unfundedpriorities suchthat the budget process through the executive and legislature can provide useful guidelines for prioritizing. OUTERYEARSOF THE MTEF Finally, the objective of an MTEF i s to try to broaden the planning cycle and make longer-termplanningeasier for sectors. However, we observe some volatility intrends in the outer years of the MTEF. We would therefore urge the government to review and present a more realistic MTEF at the next budgetworkshop in October 2003, which fully takes into account the defense review and affordability constraints, the proposed public administrationreforms, legitimate court awards, and early LTEF findings. CONCLUSION Donors urge the government to address without delay the issues raised in this statement, to implement the changes deemed essential to reestablish the confidence necessary to maintain and increase budget support. Subject to these observations and especially our concerns over defense and public administration, we endorse the government's overall MTEF and budget for next year and believe that it provides an adequate basis for donor budget support. That support may be dependent on extra assurances with respect to the strict execution of the defense budget. We would emphasize the need to ensure that the spending priorities detailed in the 2003/04 budget are respected and to restore to all stakeholdersthe confidenceinthe budget system that has beenunderminedthis year. - 111- I 2 3 I n 0 0 -xN 3 a ? E n BIBLIOGRAPHY The word "processed" describes informally reproduced works that may not be commonly available through libraries. Ablo, E , and R. Reinikka. 1998. "DOBudgets Really Matter? Evidence from Public Spending on Education and Health in Uganda." Policy Research Working Paper 1926.World Bank, Development ResearchGroup, Washington, D.C. Bank of Uganda. 2003. "Remarks onthe Macroeconomic Framework." Presentedat the Public ExpenditureReview Workshop, MFPED,May 2003, Kampala. Bethelemyand Soderling 2001), (ed) EmergingAfrica, OECD Development Center, Paris. Bevan, David, Chris Adam, John Okidi, and Fred Muhumuza. 2003. "Discussion Paper on Economic Growth, Investment and Export Promotion." Background Paper for PEAP revision, MFPED Kampala, May 14-16. Brownbridge, Martin. 2002. "Uganda's Budget Deficit." MFPED, Kampala. Processed. Development Partners, 2003a. "Remarks on Financial Management in the Justice, Law and Order Sector." MFPED, May 2003, Kampala. -2003b."ValueforMoneyinWater:DonorResponse."Paperpresentedatthe Public Expenditure Review Workshop, MFPED, May 2003,Kampala Donor Decentralization Subgroup, 2003, "Budgeting and Financial Management at Local Government Level. " MFPED, May 2003, Kampala. Dunn,David2002. GrowthAccounting for Uganda, IMFmimeo, Washington DC, 2002. DWD 2002 Government of Uganda 2002. Background to the Budget 2002/03, MFPED,Kampala. Government of Uganda. 2003a. Background to the Budget 2003/04, MFPED, Kampala. .2003b. Budget Performance for the First 3 Quarters of FY 2002/3, Kampala. .2003c. Budget Speech,June 2003, MFPED, Kampala. - 114- . 2003d. "Design of the Financial Management, Accountability and Reporting Systems under Fiscal Decentralization Strategy and Issues on Local Government Financial Management for Public Expenditure Review (2003)." Kampala. . 2003e. "Opening Speech by Dr. Speciosa Wandira Kazibwe." Presented at the PER Workshop, Government of Uganda, May, Kampala. . 2003f. "Summary of the Background to the Budget, Second PRSP Progress Report." MFPED, Kampala. Hutchinson Paul 2001 "Combating Illness" in Reinikka, R., and P. Collier, eds. 2001. Ugandas Recovery: The Role of Farms, Firms, and Government. Regional and Sectoral Studies. Washington, D.C., The World Bank. International Development Association and International Monetary Fund. 2003. "Joint StaffAssessment of PRSP ProgressReport." August 2003 Washington DC. Jayasuriya, Rawal, Margaret Miller, Elizabeth White, and Quentin Wodon. 2002. "Reaching Health and Education Targets in Argentina: A Provincial Level Analysis. December2002." World Bank, Washington, D.C. Processed. Jayasuriya, Rawal, and Quentin Wodon. 2002a. "Development Targets and Efficiency in Improving Education and Health Outcomes in Mexico's Southern States." World Bank, Washington, D.C. Processed. , 2002b. "Explaining Country Efficiency in Improving Health and Education Indicators: The Role of Urbanization." World Bank, Washington, D.C. Processed. Keefer, Phillip. 2000. "Growth and Poverty Reduction in Uganda: The Role of InstitutionalReform." mimeo, DevelopmentEconomic ResearchGroup, Washington, D.C. Local Government Finance Commission, 2002, "LGFC Revenue Sharing, Phase I1 Report", July 2002, Kampala, MOLG. Ministry of Education and Sports 2002 Eighth Education Sector Review, Final Report, 2002, Kampala. Ministry of Education and Sports (MOES), Republic of Uganda. 2003a. Education Sector Six Monthly Report (ESSMR) October 2002-April 2003. Kampala. . 2003b. Final Aide Memoire for the Ninth Education Sector Review (ESR). Kampala. .2003c. Report on Recommendations of TrackingStudies.Kampala. - 115- ,2003d. Semi-Annual Budget Performance Report FY 2002/03. Kampala. Ministry of Education and Sports, 2003, Draft Tracking Study Report on Teacher Recruitment,Deployment andPayrollManagement, Kampala. Ministry of Education and Sports, EighthEducation Sector Review, FinalReport, 2002, Ministry o f Finance, Planning and Economic Development (MFPED), Republic of Uganda. 2002a. "Is the Water Sector Delivering?" Kampala. . 2002b. Second Participatory Poverty Assessment Report: Deepening the Understanding of Poverty. Kampala. .2003a. Background to the Budget Financial Year 2003/2004. Kampala. . 2003b. Infant and Maternal Mortality in Uganda: Causes, Interventions and Strategy for the Way Forward. Kampala: Task Force on Infant and Maternal Mortality. . 2003c. "Overview of Public Expenditure Issues." Presented at the Public Expenditure Review Workshop,, May 2003, Kampala. . 2003d. "The Macroeconomic Framework and MTEF FY 2003/04-05/06.'' Presentedat the Public Expenditure Review Workshop, May 2003,Kampala. , 2003e. Semi-Annual Budget Performance Report, FY 2002/2003. Kampala: Budget Policy and Evaluation Department. . May 2003f. "Strategies to Promote Economic Growth." Presented at the Consultative Group Meeting Government of Uganda, May 2003,Kampala. , 2003g. "The Macroeconomic Framework and the Resource Envelope." Presentedat the Public Expenditure Review Workshop, May 2003, Kampala. . 2003h. The Macroeconomic Plan and Indicative Budget Framework for FY 2003/04-2005/06, Submission to Parliament under the Budget Act 2001 (section 4), March2003. Kampala. Ministry of Health (MOH), Republic of Uganda. 2002. Annual Health Sector Performance Report: Financial Year 2001/02. Kampala. 2002 Annual HealthSector Policy Review ,2001-02 data. . 2003. "Budget Efficiency and Equity in Incidence of Health Service Benefits." Presentedat the Public Expenditure Review Workshop, May 2003, Kampala. - 116- HealthSector Strategic PlanMid Term Review Report, 2003. Ministry of Health, Health Sector Strategic Program (HSSP) 2003 (Midterm Review Report), May 2003, Kampala. MinistryofJustice.2002. "Justice, Law andOrder Sector (JLOS)." InBaseline Study on Financial Management Systems. Deloitte and Touche, May 2003, Kampala. Ministry of Justice, Law and Order, Republic of Uganda. 2003. "Baseline Study on Financial Management Systems." Presentedat the Public ExpenditureReview Workshop, May 2003, Kampala. Ministry of Local Government, Republic of Uganda. 2003. Budget and Financial Management at Local GovernmentLevel. May 2003,Kampala. Ministry of Water, Lands and Environment, Republic of Uganda. 2002a. "Value for Money in Water Sector." Presented at the Public Expenditure Review Workshop, MFPED, May 2003, Kampala.. .2002b. WaterSector Performance. Kampala. . 2003a. "Commentary on Value for Money inthe Rural Water Sector." Presented at the Public Expenditure Review Workshop MFPED, May 2003, Kampala. . 2003b. "Value for Money in the Rural Water Sector." Presented at the Public Expenditure Review Workshop, MFPED, May 2003, Kampala. Moller, Lars. 2003. "Uganda's Progress in Attaining the PEAP targets inthe Context o f the MillenniumDevelopment Goals." Background Paper for the Consultative Group MeetingMFPED, Kampala, May 14-16. Reinikka,R., andP. Collier, eds. 2001. Ugandas Recovery: TheRole of Farms, Firms, and Government. Regional and Sectoral Studies. Washington, D.C.: World Bank. Reinikka, Ritva, and Jakob Svensson. 2003. "Working for God? Evaluating Service Delivery of Religious Not-for-Profit Health Care Providers in Uganda." Policy Research Working Paper 3058. World Bank, Development Research Group, Washington, D.C. Reinikka, Ritva, Jacob Svensson, and Magnus Lindelow. 2002. "Health Care on the Frontline: Survey Evidence on Public and Private Providers in Uganda." mimeo, World Bank, Development ResearchGroup, Washington, D.C. Uganda Bureau of Statistics. 2001. Demographic and Health Survey 2000/01. Kampala. UgandaBureauof Statistics. 2002. National Service Delivery Survey 2001, Kampala. - 117- . 2002. StatisticalAbstract. Kampala. Uganda Debt Network. 2003a. "Discussion of Financial Management at Local Government Level." Presented at the Public Expenditure Review Workshop, MFPED, May 2003, Kampala. . 2003b. Participatory Community Based Budget Performance Monitoring and Evaluation. Abridged report, Tororo District Pilot. Kampala World Bank. 2002. Report on the Progress and Challenges of Budget Reforms. PER Report No 24882-UGYWashington, D.C. . 2003a. "Budget Strategy and Performance, 2002/03." Statement presented on behalf of all development partners at the Public Expenditure Review Workshop, MFPED, May 2003, Kampala. . 2003b. "Medium Term Expenditure Framework 2002/03-2004/05 ," Statement presented on behalf of all development partners at the Public Expenditure Review Workshop, MFPED, May 2003, Kampala . 2003c. "Poverty Reduction Support Credit 2 and 3." Various aides-me`moires. Washington DC. World Health Organization "Report of the Commission for Macroeconomics and Health", Geneva (2002). World Health Organization, MOH Study, 2000,2001 & 2002 data. - 118 -