Report No. 26807-TA United Republic of Tanzania Public Expenditure Review FY03 Managing Public Expenditures for Poverty Reduction Report on Fiscal Developments and Public Expenditure Management Issues June 2003 Document of the World Bank Prepared by the World Bank and Members of Tanzania PER Working Group GOVERNMENTFISCAL YEAR FY03 =2002/2003 =July 1,2002to June 30,2003 CURRENCYEQUIVALENTS CurrencyUnit =Tanzanian Shilling(T.Sh.) Interbank Market mid-rate: US$l.OO =T.Sh. 1045 (June 2,2003) ABBREVIATIONS AND ACRONYMS ACCGEN AccountantGeneral MoF Ministry of Finance AfDB Aftican DevelopmentBank MoH Ministry of Health AIDS Acquired Immune Deficiency Syndrome MSTHE Ministry of Science, Technologyand ALAT Association of Local Authorities in Higher Education Tanzania MTEF Medium TermExpenditure Framework BOP Balance of Payment NAO National Audit Office BOT Bank of Tanzania NFA Net Foreign Assets CAG Controllerand Auditor General NGO non governmentalorganization CBMS Cash Budget ManagementSystems O&E organizationaland efficiency (reviews) ccs Commitment Control System O&M operations and maintenance CFAA CountryFinancial Accountability oc other charges Assessment ODA OfficialDevelopment Assistance CG ConsultativeGroup PEDP Primary Education DevelopmentProgram CSD Civil ServiceDepartment PE personnel emoluments DANIDA Danish InternationalDevelopmentAgency PER Public Expenditure Review DFID Departmentfor International Development PHC Primary Health Care DMT District, Municipal and Town Councils PORALG President's Office -Regional Authoritiesand ESRF Economicand SocialResearch Foundation Local Government EU EuropeanUnion PRBS Poverty Reduction Budget Support FY Fiscal Year PRGF Poverty Reduction and GrowthFaciltiy GDP gross domesticproduct PRSC Poverty Reduction Support Credit GOT Governmentof Tanzania PRSP Poverty Reduction StrategyPaper HIPC Highly Indebted Poor CountriesDebt PSRP Public Sector ReformProgram Initiative RAS Regional Administration Secretariat HIV Human Immuno-deficiencyVirus REPOA Research on Poverty Alleviation IDA InternationalDevelopmentAssociation ROSC Report on Standardsand Codes IFMS IntegratedFinancial Management System SASE SelectiveAccelerated Salary Enhancement IMG IndependentMonitoringGroup SDC SwissDevelopment Cooperation IMF InternationalMonetary Fund SDP SectorDevelopmentProgram LA Local Authority SSA Sub-SaharanAftica LGBG Local GovernmentBudget Guidelines SWG SectorWorking Group LGRP Local GovernmentReformProgram TANESCO TanzaniaElect& Supply Company LPO Local PurchaseOrder TAS TanzaniaAssistance Strategy MAC Ministry of Agriculture and Cooperatives TB TreasuryBill MDG MillenniumDevelopmentGoal TRA TanzaniaRevenue Authority MoEC Ministryof Education and Culture UNDP United NationsDevelopmentProgram VAT Value Added Tax Vice President: Callisto Madavo SectorDirector: Paula Donovan CountryDirector: Judy O'Connor SectorManager: Robert Blake Task Team Leader: Robert Utz Table of Contents Acknowledgements ............................................................................................................ i ExecutiveSummary of Findings andRecommendations ............................................... i 1 INTRODUCTION . ..................................................................................................... 1 PART I:REVIEW OF FISCAL PERFORMANCE .................................................... 3 2. The MacroeconomicBackground ........................................................................... 5 Real Economic Growth............................................................................................... 5 6 Monetary and Financial Developments ...................................................................... General Price Developments....................................................................................... 7 External Sector Developments.,.................................................................................. 9 3. OverallFiscalPerformance ................................................................................... 13 Aggregate Fiscal Discipline...................................................................................... 13 14 Donor Assistance ...................................................................................................... Domestic Revenue.................................................................................................... 18 Government Expenditures ........................................................................................ 19 4. Strategic ResourceAllocation ................................................................................ 21 Are Budget Allocations a Reliable Guide to Public Expenditure Outturns? ............21 What Are Public Funds BeingUsed For?................................................................. 24 5. BudgetExecutionAssessment Basedon the Report of the Controller and . Auditor Generalfor FYOl .............................................................................................. 35 PART11: ALIGNING THE BUDGETWITH THE PRS .......................................... 39 6. Alignment of the Budgetwith Core PRSPPrinciples .......................................... 41 ResultsOrientation o fthe BudgetProcess................................................................ Country Driven .Open and Participatory Nature o f the Budgeting Process.,..........41 Partnership-Oriented ................................................................................................. Comprehensiveness -Recognizingthe Multidimensional Nature o f Poverty .........43 45 46 Long-term Perspective for Poverty Reduction ......................................................... 48 7 . Poverty ReducingExpenditures ............................................................................ 51 What are Poverty Reducing Expenditures inTanzania?........................................... 51 54 Do Public Expenditures Reflect PRS Objectives? .................................................... Can Pro-Poor Expenditures be Adequately Monitored?........................................... 56 Increased Priority Expenditures- Have Plentiful Resources UnderminedEfficient Reallocation?............................................................................................................. 59 Targeting o f Expenditures......................................................................................... Comparisonto Original PRS Projections for the Priority Sectors............................ 62 63 What Does Development Assistance Finance?......................................................... 65 8. Review of Alignment of SectoralPER work with PRS ....................................... 69 The Sector PER Process............................................................................................ 69 FullFinancing Cost Estimates andthe FinancingGap............................................. Review o f Sector PERs for FY03 ............................................................................. 71 75 Stylized Lessons and `Next Steps' ............................................................................ 78 PART 111: SELECTED ISSUES INBUDGET MANAGEMENT ............................ 80 9. Budget ManagementIssues .................................................................................... 81 Cash Budget vs Cash Flow Management ................................................................ . 81 Official Development Assistance and Fiscal Management ...................................... Implementing National and PRS Priorities Through Local Government Budgets...82 90 Update on Assessment o f Government Capacity to Track Poverty Reducing Expenditures ............................................................................................................. 95 ANNEXES ..................................................................................................................... 104 105 Annex B: Reviews of Education. Health. and HIV/AIDSPERs.................................. Annex A: Recent Exchange Rate and TradeDevelopments ....................................... 110 Annex C: Reviews of Water. Roads. andAgriculture PERs....................................... 116 Annex D: Summary Priority Sector Per Reports of Fiscal Year 20021'03 ..................124 Education................................................................................................................ 129 Health...................................................................................................................... 137 HIVIAIDS............................................................................................................... 144 Roads....................................................................................................................... Water ....................................................................................................................... 150 157 Agriculture Livestock................................................................................................................. .............................................................................................................. 164 166 Annex E: Data............................................................................................................ 172 List of Boxes 16 Box 2. Audit certificates .................................................................................................. Box 1. Fiscal incentives inthe mining sector .................................................................. 37 51 Box 4 Targeting and governance .................................................................................... Box 3. What are pro-poor expenditures? ......................................................................... . 64 Box 5. Are transfers to the districts pro-poor?................................................................. 65 ListofFigures Figure 1. Annual headline. food andnon-food inflation July 2001.Dec 2002................7 . Figure2 Annual growth rates inmonetary aggregates. M3.M2. and MO........................ . 8 Figure3. Deposits of commercial banks............................................................................ 8 Figure4. Structure o ftraditional exports, 2002............................................................... 10 . 10 15 Figure7. Change intax revenue and exemptions (% o f GDP), FY98-FYO1 .................. 17 Figure6 Tax exemptions (% of GDP), FY97-FYO1....................................................... Figure5. Structure of non-traditionalexports, 2002........................................................ 18 Figure 10. Domestic revenue and program support, level and difference betweenbudget Figure 9. The macro-economic environment................................................................... Figure8. Customs exemptions to private companies by sector 2000/01.......................... 25 25 Figure 11.Difference betweenresource estimates and actual resource availability (% of and actual (% o f GDP).............................................................................................. Figure 12. First charge expenditures, level and difference between budget and actual (% GDP) ......................................................................................................................... 26 o f GDP) ..................................................................................................................... 26 Figure 13. Deviation betweenestimated and actual expenditures on first charge items (% ..................................................................................................................... 27 Figure 14. Non wage expenditures, level and difference between budget and available of GDP) Figure 15. Development expenditure captured inthe appropriation accounts as a share o f (% o f GDP) ............................................................................................................... 28 Figure 16. Share o f project credits intotal foreign financing o f development expenditures development expenditure recordedinthe budget. FY97-FY02................................ 29 ................................................................................................................................... 29 Figure 17. Realper capita expenditures inpriority sectors doubles between FY99 and Figure 18. Spendingon education, health, and roads claims 84 percent o f priority sector FY04 ......................................................................................................................... 57 spending.................................................................................................................... 57 Figure 19 Per capita spending has more than doubled inmost priority sectors between FY99 and FY03......................................................................................................... . Figure 20. Increases indiscretionary spendingexceed increases inpriority sector 58 Figure21. Real expenditure increases inpriority sectors andpriority areas, FY00-FY06 spending .................................................................................................................... 59 60 Figure22. Percentage change inpriority and non-priority expenditures......................... ................................................................................................................................... 60 Figure23. Share o fpriority and non-priority sector expenditures intotal discretionary 61 Figure24 Change inpriority sector expenditures, FY00-FY03 ..................................... expenditures, FY99-FY06......................................................................................... . Figure 25. Spending on priority sectors exceeds PRS targets ......................................... 62 63 Figure 26 Official development assistance from all sources and spending on priority . Figure27. Increase indevelopment assistanceandpriority sector expenditurecompared sectors (as % o f GDP) ............................................................................................... 66 to levels inFY99 ....................................................................................................... 67 Figure28. Increase inflexible aid resources andrecurrentpriority sector expenditures over FY99................................................................................................................. 68 82 Figure30 Expectedand actually received external assistance....................................... Figure29. Monthly exchequer releasesand expenditure................................................. . 84 Figure 3 1. Average change inexchange rates, FY02-FY03............................................ 84 Figure 32 Capturing developmentassistance inthe budget............................................ . 86 ListofTables Table 1. Real GDP annual growth rates (at constant 1992prices). 1995-2002.................5 6 Table 3. Trade account....................................................................................................... 9 Table 2. Sector contributions to GDP (at current prices). 1995-2002 ............................... 13 Table 5. Revenue performance (% o f GDP) FY98-03 .................................................... Table 4. Financing of the fiscal deficit (% o f GDP), FY98-FY03 .................................. 14 Table 6 Composition o f external assistance (% o f GDP), FY98-03............................... 19 Table 7. Government expenditure (% o f GDP), FY98-03 ............................................... . 19 Table 8. Civil service employment FY99-FY03 (December o f each year) ..................... 20 Table 9. Composition o f public expenditures as % o f GDP, FY96-FY03 ...................... 21 Table 10. Functional allocation of recurrent expenditure (actuals, % o f GDP), FY96- Table 11. Social sector recurrent expenditures (actuals, % o f GDP), FY96-FY03 .........23 FY03 .........................................................................................................................22 Table 12. Sectoral development expenditures (actuals, % of GDP), FY96-FY03 ..........23 Table 13 Share o f votes receiving reallocations ............................................................. 30 Table 14. Earmarked and actual use o f retained "special expenditures" ......................... . 31 Table 15. Earmarked uses o f special expenditure for FY03 ............................................ Table 16. Share of votes with ovedunderspending, FYOO-FY02 ................................. 31 32 . Table 18. Ranking FY03 sector PERs ............................................................................. Table 17 Submission of accounts and production o f CAG report, FY97 -FYOl..........36 75 Table 19 Summary o f PER and MTEF full-financing estimates (Tshbillions) 2003/04 - 2005/06 ..................................................................................................................... . Table 20. Variability of revenue and aid ......................................................................... 76 83 Table 21 Reviewof progress instrengthening capacity for the tracking o f poverty . reducing spending ........................................................................................... 97 ACKNOWLEDGEMENTS Since 1997/98, Tanzania has carried out an annual Public Expenditure Review (PER) process which has the interrelated twin objectives o f supporting the budget process as well as undertaking an external review o f fiscal developments. The Tanzania PER Working Group comprising o f representatives from the Government of Tanzania, the World Bank, UN agencies, other bilateral and multilateral donors, research and academic institutions, and NGOs determines the agenda for the annual PER process, guides and finances the implementation o f the agreed work program, and reviews all outputs. It also represents an important forum for awareness raising and discussion o f public expenditure issues between government and a wide array of interested stakeholders inTanzania. This report presents the findings o f the external review o f fiscal developments. These were shared with government and members o f the PER working group as well as with a wider audience at the annual PER Consultative Meeting held in May 2003, prior to the finalization o f the Tanzanian government budget. The report i s primarily based on the findings o f a joint donor mission undertaken in February 2003, led by Robert Utz (mission leader, AFTP2) and included Benno Ndulu, Praveen Kumar, Philip Mpango, Emmanuel Mungunasi (AFTP2); Rosa Alonso ITerme (WBI), Hamisi Mwinyimvua (University o f Dar es Salaam), Ben Tarimo, Whilhelm Ngasamiaku (World Bank consultants), Helen Tilley (European Delegation), Tone Times (Norwegian Embassy), Jytte Laursen (Danish Embassy), Bertil Oden (Swedish Embassy), Amon Manyama (UNDP), and Iain Jones (DffD). Jamil Mubarak (World Bank consultant) carried out the review o f the sector PERs. The report was written under the supervision o f Fred Kilby, Sector Manager, AFTP2 and Peniel Lyimo, Deputy Permanent Secretary, Ministry o f Finance, Government o f Tanzania. Allister Moon (ECSPE) and Anand Rajaram (PRMPS) served as peer reviewers. Emmanuel Munganasi provided excellent research assistance and compiled the statistical appendix. Patrick Mamboleo and Arlette Sourou were responsible for the word processing and the physical production o f the report. I EXECUTIVESUMMARY OF FINDINGSAND RECOMMENDATIONS Introduction Over the past few years, the Poverty Reduction Strategy (PRS) has become the principal framework for public expenditure management and policy. This has important implications for the PER process, which has evolved into the principal analytic instrument to support the translation o f PRS objectives and strategies into budgets. At the same time, monitoring the alignment of the budget with the PRS objectives has become another important role for both sectoral and thematic studies carried out as part o f the PER process. In line with these developments, this external evaluation looks at budgetary developments primarilythrough the lens ofthe PRS. Review of Fiscal Performance Fiscal discipline. The financing requirements o f the PRS exceed currently available resources. To close the gap, government has thus to consider trade-offs between the impact on poverty reduction o f a slower pace o f PRS implementation if the resource envelope i s not increased and, on the other hand, the economic and social consequences o f trying to increase resources from domestic revenue, foreign aid, and borrowing to meet PRS requirements. Overall, government is clearly following a policy where macro- economic stability i s given top priority as the basis for sustained economic growth. This i s also reflected in the approach to closing the financing gap which so far relies primarily on efforts to mobilize additional foreign assistance through grants and concessional credits. Efforts to increase domestic revenue have been less successful and net-domestic borrowing has infact beennegative inrecent years. Overall fiscal policy remained very conservative in FY02 and FY03. The deficit after grants was 1.1 percent o f GDP inFY02 andprojected to increase to 2.6 percent inFY03. As in previous years, concessional foreign borrowing exceeded the deficit, allowing a reduction indomestic debt. Tanzania continues to use a cash budget system that strictly constrains spending and commitment levels to short term availability of cash. While the system was clearly useful when it was introduced in 1996 and the benefits outweighed the cost inthe form of reduced operational efficiency and regular deviations from the approved budget, improvements in budget and aid management put the continued use o f a cash budget system into question. With respect to budget management, the Integrated Financial Management System, the revised Financial Management and Procurement Act, and generally improved budgeting practices provide the government with a sophistic set o f I instruments to pursue macro-economic and fiscal stability even without the use o f the cash budget system. In addition, front loading budget support to the beginning o f the fiscal year including the availability o f PRSCl eliminates one o f the main uncertainties with respect to resource availability during the course o f budget implementation. It would be advisable to revisit options for relaxing the stringency of the cash budget by providing quarterly releases to all MDAs and giving them one fourth of their budget allocation during theJirst quarter of thefiscal year. In subsequent quarters, expenditure releases to the non-priority sectors could be adjusted in line with resource availability. Fiscal risks involved inmovingto general quarterly releases for all MDAs are likely to be small. This is especially so if for the time being the policy of providing full protection o f budget allocations only to the priority sectors i s retained. Revenue. The revenue to GDP ratio fell marginally to 12.1 percent in FY02 but i s projected to reach 12.5 percent in FY03. A series o f recent studies indicate that Tanzania's revenue effort in the short to medium term i s severely constrained by its economic structure, and point to measures such as the elimination o f exemptions which could lead to an immediate increase inrevenue. Government i s encouraged to revisit the current regime of exemptions and limit them to those where there i s a clear economic and social justification and where the potential for abuse can be kept to a minimum. With respect to new tax policy measures, it will be important to pay attention to the impact of such measures onpoverty andgrowth. Official development assistance. Foreign assistance increased inFY02 to 5.9 percent o f GDP and is projectedto increase to 7.9 percent inFY03. Most o fthis assistance is inthe form o f grants and the shift from project to program support continues. The impact of increasing levels and the changing composition o f aid on the real exchange rate and international competitiveness o f Tanzania needs to be closely monitored. Our analysis indicates a real appreciationo f the exchange rate over the 1990s coinciding with a decline o f exports as a percentage o f GDP and a virtual collapse o f exports o fmanufactures. This highlights the need to pursue policies that would support Tanzania's international competitivenessand minimize anypotential "Dutch disease" effects of aid. Micro-aspects of development assistance. Considerable progress has been made in capturing foreign assistance in the budget and in the appropriation accounts. The initiative by PO-RALG and individual sector ministries to solicit and consolidate information on donor support to LAs i s commendable. To support ongoing efforts to develop comprehensive information on both domestically and foreign Jinanced public expenditures, it would be useful to improve the scope of the information solicited from donors to correspond with the information compiled by MoF. In turn, MoF needs to routinely obtain this information from PO-RALG and incorporate/synchronize it with what already exists in the aidflow data base kept by the External Finance Department. Better information on development assistance going directly to the districts could contribute to more equitable and transparent resource allocations from the center to the districts and be a important element of efforts to integrate donor assistance into government's budgetary and financial systems. 11 * . Expenditures. Expenditures increased to 18.3 percent of GDP during FY02 with most o f this increase being spending on OC. The expenditure outturn for FY03 is projected at 20.4 percent o f GDP. With the sharply increased amount of resources available for OC, it will be important to monitor more closely the efficiency of the use of these resources and strengthen systemsfor the efJicient use and management of government assets such as vehicles and ofJicespace. Government is encouraged to sustain the implementation of a medium term wagepolicy which would increase real salariesfor civil servants and thusprovide afirmer basisfor many ongoing efforts to increase the performance of the public sector. While government has been keen to contain the wage bill, employment allowances which contain payments for overtime, per diems, travel, etc. have risen significantly over the past two years. Consequently, it would be useful to examine whether there is scope to reduce expenditures on employment allowances and use the savings to increase basic salaries. Strategic ResourceAllocation Functional allocation of resources. The main change in strategic resource allocation in FY02as the increase insocial spending from 4.4 percent o f GDP inFYOl to 4.7 percent o f GDP in FY02, budgeted to increase to 4.8 percent o f GDP in FY03. The only other sector that saw a large increase in resources was administration expanding from 2.3 percent in FYOl to 3.5 percent o f GDP in FY02 and budgeted to further increase to 3.6 percent o f GDP in FY03. However, a significant part o f the increase in administration i s due to the clearing o f arrears and cleaning up o f liabilities to staff and related to the divestment o fparastatal enterprises. Distribution of resources between the center and districts. Central MDAs were the main beneficiaries of increases in priority spending, while transfers to the local authorities increased only slightly during FY02 and are projected to remain at that level in FY03. The concerns inthis area are two-fold. First, there are still large amounts of expenditures inthe education and health sectors shown inthe budget o f ministries which more appropriately should be shown as transfers to local authorities. This would also increase transparency in the distribution o f resources across districts, which i s desirable once there i s a clear consensus that more resources should be channeled towards the poorer and less endowed districts. In order to ensure that thepoor benefitfrom increased social spending, government is encouraged to give priority to the poverty mapping exercise and use the results together with sectoral information to target increases in social expenditures to the poorest areas. The second area o f concern relates to the still low levels o f funding for basic services, despite large resource gaps identified inthe PRS and sector PERs. In some sectors such as water and agriculture, increasing transfers to the districts is contingent on the availability o f district sector development plans. To ensure the speedy development of such plans, increased technical assistance to the districtsfor thepreparation of suchplans may be needed. The budget is regaining its role as primary resource allocation mechanisms. In the past, one of the key weaknesses o f the budget process has been that budget allocations ... 111 and actual spending showed only little relation. However, over the past five years, the macro-economic environment has been relatively stable and provided a more conducive environment for realistic budgeting and our analysis shows indeed that revenue projections have been increasingly close to actual outturns. Disbursementson program support by donors on the other hand have been more difficult to project reliably. Inthis area, front loaded budget support through the PRBS/PRSC budget support should bring more predictability. The analysis also shows that during the past seven years, the difference betweenthe budgetedand actual resource envelope (excluding project support) was 0.7 percent o f GDP, with a deviation o f 1.8 percent o f GDP recorded in FY99 being the largest, This suggests that macro-economic stability could have been maintained, even without adjusting budgetary expenditurereleases on a quarterly or monthly basis to match short term availability o f funds. On the expenditure side, first claim expenditures (debt service and wages and salaries) were also generally quite close to what was budgeted. As a consequence o f this greater accuracy o f resource and first claim estimates, deviations between actual and budgeted discretionary non-wage expenditures have been less than one percent o f GDP over the past five years. Unallocated expenditures. In FY02, the Ministry o f Finance retained T.Shs. 232.4 billion in its budget for reallocation during the budget year for earmarked purposes. Out o f this unallocated expenditure, T.Shs. 202.4 billion was reallocated to specific expenditure items either withinthe Ministryo f Finance (T.Shs. 9.9 billion) or under other votes (T.Shs. 192.5 billion). A review o f the reallocation warrants for FY02 indicates that while most o f the reallocations were in line with what was originally earmarked, significant amounts had also been reallocated to uses that were clearly not part o f the original earmarking. For FY03, unallocated expenditure have declined to about T.Shs. 80 billion, representing mainly funds for salary increases and new recruitment. In this respect, it will be important to pay more attention to better integration of civil service and budget management with the objective of fully integrating decisions on stafing and wages into the budgetprocess. Report of theAuditor and Controller General Improvements infinancial accountability. The C A G reports for central government and local authorities for FYO1 suggest significant improvements in financial management at all levels. Fewer accounts received an adverse opinion and accounts were submittedon a more timely basis to the National Audit Office. Delays in publishing the CAG reports. The C A G report for central government was publishedwith a 6 months delay and that of local authorities with a 2 months delay due to capacity constraints, electricity shortages experienced at the NAO, and time consuming auditing of donor funded projects and special accounts. Since presentation of the CAG reports to Parliament in a timely manner is essential to increase the impact of these reports, government is encouraged to take measures that would allow the CAG to produce its reports without delay. Such measures could include improving the contracting of work to outside auditors. iv Computer based auditing. The NAO has still not developed sufficient capacity to undertake computer based audits o f the IFMS based government accounts. Establishing such capacity should be of a very highpriority for government to ensure the integrity of the IFMS based accounting. Moving to a computer based auditing is also likely to reduce the time neededfor thepreparation of the annual reports. Reply to audit queries. Responses to audit queries have remained low as a share o f total queries. Measures should be taken to create the right incentivesfor accounting ofJicers topay more attention to these issues. The update of the HIPC expenditure tracking assessment shows progress in addressing weaknesses in government's capacity to track poverty reducing expenditures identified in the initial assessment carried out in 2001. This includes enhanced budget classification covering transfers to the districts, clearer identification o f poverty reducing expenditures in budget documents and a strengthening of the audit function. However, progress has been slower than anticipated at the time of the assessment. It will be important that measures identiJied in the action plan are implemented speedily if Tanzania wants to retain and strengthen its classijkation as one of the best performing systems in Africa. Recent expenditure tracking work indicates virtually no leakages in the flow of capitation grants from the Ministry o f Finance to schools Linkage between Fiscal Policy and the Poverty Reduction Strategy The responsiveness o f the budget and budgeting process to the PRS was assessed by focusing on three key areas: (i)the reflection o f PRS principles (country ownership, results orientation, comprehensiveness, long-term perspective, and partnerships) in the budget process; (ii) comprehensive analysis o f pro-poor spending; (iii) the role of a and sector PERs as the analytical basis for translating the PRS into budgets. Participation. The participatory public expenditure review process and, more generally, the high level of interaction and dialogue between government ministries and between government, donors and civil society are very much in line with the spirit o f the PRS process and allow the continuation o f participation into the implementation phase. However, there are concerns as to the extent to which these participatory processes have an impact on government policy. NGOs and civil society are included in government's participatory processes. However, there appears to be greater scope for their involvement, especially in the area o f participatory monitoring and evaluation. This could complement technocratic, government led approaches to monitoring underthe PRS and PERprocesses and could be particularly useful in evaluating the impact o f specific programs and the changes that have occurred as a result of program initiatives. It is thus recommended to explore optionsfor including participatory M&E as an importantfeedback mechanism into the PERprocess. V Results orientation. The PRS provides key objectives o f public policy. However, in many areas results are not sufficiently well defined to feed directly into the budget process. DeJining a key set of monitorible indicators that can be related to budgetary inputs needs to be apriority as a basisfor evaluating the impact of budgetary decisions on desired outputs. This would enhance government's capacity to make appropriate adjustments to expenditure policies ifdesired outputs and outcomes are not achieved. Tanzania has also put in place a variety o f monitoring and performance management systems. However, the link o f most o f these systems to the budget process is only tenuous. Consequently, the allocative process shows little responsiveness to performance o f spending units. Enhancing the results orientation of the budget process requires also more attention to the development and monitoring of efficiency indicators, based on a clearer deJinition of desired outputs. To assess whether public expenditures really beneft the poor, work on benefit incidence analysis as well as specific program evaluations need to be carried out. Use of information on poverty for allocative decisions. The 2001 household budget survey, a labor force survey, the census as well as the poverty and human development report provide a rich set o f data and analysis on poverty inTanzania. It will be important that spending units and the Ministry of Finance use this information to strengthen the design of pro-poor expenditure policies. This includes work on poverty maps, which could be an important input to special resource allocations to regions and districts as well as to the design o f sectoral programs. Data on pro-poor expenditures. Monitoring o f pro poor expenditures i s constrained by changing definitions of priority expenditures, limited information on spending by the local authorities and donor funding outside the budget, and yet incomplete GFS classification o f the development budget. Work to address these weaknesses i s ongoing and it will be important that government continues efforts to make detailed data on pro- poor expenditures, preferably on a quarterly basis, widely available. Real per capita spending in the sectors and areas dejined as pro-poor has more than doubled between FY99 and FY04. However, the analysis shows that spending outside the priority sectors increased at a similar pace. The increased spending on priority sector was made possible by a expanding public sector resource envelope and not the result o f reallocations from other sectors. However, a significant amount o f spending increases outside the priority sectors was due to the settlement o f arrears and liabilities related to the privatization process. In addition, some expenditure increases aimed at enhancing service delivery in the priority sectors cannot be limited to these sectors alone. This holds inparticular for the implementation o f the public sector mediumterm wage policy. Resource gaps. Significant gaps between requirements and budget ceilings for priority sectors are cause for concern. It will be important to develop a clear medium to long term plan as to how the resource gaps will be filled. This would also require a clearer definition o f what being a priority sector means in terms o f annual resource allocations. Government i s encouraged to continue its focus on increasing domestic revenue through measures that are compatible with sustained economic growth and poverty reduction. In vi this context, it seems useful to complement recent analytic work on the revenuepotential and on measures to increase revenue with work that would look at the vertical and horizontal equity of the tax system and the poverty and growth impact of proposed changes in tax policy and administration. Spending on the priority sectors has responded positively, although with a lag, to donor financing and currently priority sector spending exceeds the total volume of ODA providedto government. With the increase inbudget support inrecent years, an important question is whether such increases can be mapped to increases in poverty reducing expenditures. Given the fungibility o f funds, the question indeed also pertains to development funds provided in the form o f project support. The analysis shows that increases inproject support as well as sector programs and basket funding arrangements can be mapped to increases in development expenditures in the priority sectors and it appears that spendinghas increased faster thanfunding from these sources. Increasesin general budget support and HIPC debt relief which are primarily used to fund recurrent expenditures, have not been matched by an equal increase in spending on the priority sectors funded from the Consolidated Fund. This implies that increases in program support have also been used to finance expenditure increases outside the priority sectors. As noted above, for some expenditures such as wages and salaries it i s not feasible to limit increases only to the priority sectors. However, it is recommended that government make efforts to ensure that increases in budget support are used to enhance spending in thepriority sectors and to cover the signijkant resource gaps existing in these sectors, if PRSP and MDG targets are to be met. Developing a clear strategy on how budget support will be used to fill resource gaps in the priority sectors would be useful. Sector PERprocess InFY03, government reestablished sector working groups to enhance the PERprocess at the sectoral levels, which plays a critical role in translating poverty reduction objectives into budget proposals. There appear to be significant differences across sectors with respect to the role o f the sector working groups inthe sectoral budget process. While in some sectors sectoral PERs and sector working groups seem to be well integrated into the regular budget process, in other sectors there appear to be parallel processes. Recommendations to strengthen the relevance o f the participatory processes and especially sector PER work cover the following areas: Define more clearly the role o f sector working groups identify steps inthe budget preparation process where ministerial submissions to the Ministry o f Finance are discussed inthe sector working group. Strengthened participation by the Ministry o f Finance inthe Sector Working Groups. Restructure the content o f sector PER work where annual updates are standardized clearly based on the information requirements o f the Ministry o f Finance. Develop a multi-year schedule o f indepth strategic sector PERs. vii A reviewofthis year's sector PERs indicates that expenditure management andexecution have significantly improved since the commencement o f the PER process. The improvements, however, are uneven, as reflected by the varying quality o f the strategies, plans and performance, costing and full-financing estimations o f this year's sector PERs. A common theme of the sector PERs are weaknesses in financial management and implementation capacity at the local level which constrain the accelerated implementation o f the PRS. Based on these findings, there is a need to adequately support weaker sector PERs, strengthen the identijkation, costing, and providing comprehensiveforecasting of priority and sector activities, revise PRS targetsfor sector that slipped behind, develop monitorable performance indicators and PRS targets for HIVAIDS and Agriculture sectors, build capacity particularly at the LAs, and link the rate of decentralization with that of capacity improvements at LAs. Local GovernmentBudget Issues Harmonization of fiscal years. Government plans to align the fiscal year o f the local authorities with that of the central government starting in July 2005. As part of the harmonization process, it will be important to harmonize thepreparation process and the content of budget guidelinesfor central and local governments. Local government budget guidelines. The local government budget guidelines play an important role in linking the central and local budget process. However, it needs to be recognized that there i s an inherent tension betweentheir role o f providing guidance to local authorities on central (PRS) expenditure priorities and the objectives o f the local government reform program to give local authorities more independence in making decisions on the use o f resources. The local government budget guidelines for FY03 contain very detailed instructions on the process o f budget preparation, forms to be used, and reporting requirements. Aside from general references to the Vision 2025, the CCM election manifesto, and the priority sectors under the PRS, they contain no specific sectoral guidance. Sector ministries provide guidance to the districts directly outside the budget guidelines process, but some o f the LAs visited indicated that they often do not have the required documents on sectoral objectives, policies, and standards. Most of the information contained in the local government budget guidelines could be presented in a standard budget manual which remains valid over a longer period. However, it would be useful if local government budget guidelines contained guidance on changes in sectoral policies, objectives, and standards which affect budgetary planning by the districts. The preparation o f the local government budget guidelines is the responsibility o f the President's Office - Regional Administration and Local Authorities. Efforts to include other stakeholders, including sector ministries, local authorities, and ALAT are mainly limited to commenting on guidelines rather than providing inputs and actually participating in the preparation of budget guidelines. There appears to be a frequent disconnect between local authorities and central sectoral ministries in the preparation o f both the central and local government budget guidelines processes. It is recommended that both central and local budget processes are reviewed with the objective to ensure Vlll ... better communication and information sharing between the local authorities and the sectoral ministries throughout these processes. Establishment of a local government budget guidelines committee with representationfrom the key sectors, local authorities, the central ministries, andALAT should be considered. Review of Progress in Addressing Issues Identijied in Previous PERs Significant progress has beenmade inaddressing most o f the issues identified inprevious PERs, notably regarding improving debt contracting and management, including revision o f the Loans, Guarantees and Grants Act No. 30 o f 1974. Similarly, a format for presenting disaggregated poverty reducing expenditures has been presented in the 2003/04 - 2005/06 Budget Guidelines to act as a framework for informing the PRS progress report and tracking o f pro-poor expenditure. However, government is encouraged to operationalize specijk recommendations regarding the institutional roles for debt contracting and management and synchronization of debt data bases in the ACGENS ofJice and BOT.The government is also urged to expedite theplanned work to review existing financial and service delivery tracking systems. The following table provides detailed information on progress with the implementation o f recommendations from previous external evaluations. ix X n .3 X -0 9 L L E E .3 X i f a: N rd C 2 0 3x D Bm P e! BM B f 3 e, 23 9e, 2 3 se,e e, aB b4 ae, s K T Y e, a .-e, $ v) .-c -.-e e N Ye -4 0 M G '5 2 gu q 0 % Summary of Key Recommendations IIssues I Recommendation - FISl LLPERFORMANCE 1. Need for better revenue Strengthen tax administration through implementation of performance recommendationsof Tax AdministrationProjectmid-termreview. Consider findings of recent studies on government revenue for implementation. With respect to new tax policy measures, it will be important to pay attentionto the impact on povertyandgrowth of suchmeasures. 2. Tax exemptions and voucher Extend the voucher scheme to other categories of import tax exemptions - scheme inthe next fiscal year. 3. The current regime of Revisit the current regime of exemptionsand limit them to those where exemptions there is aclear economic and socialjustification while the potentialfor - abuse can bekept to aminimum. STB rEGIC RESOUCEALLOCA ON ._ 4. Expenditure increases appear Target budgetary resources more aggressively to front line service primarily in central spending delivery unit budgets rather than in Enhance budget classification to ensure that all transfers to districts are transfers to the districts, who indeedshown as such in the budget. are responsible for basic Strengthen capacity of local authorities to develop plans and programs servicedelivery necessary to access funds under the Agriculture Sector Development and other sectoral program.. 5. Distribution of resources and Give priority to the poverty mapping exercise and use the results together targetingthe poor inthe social with sectoral informationto target increasesin social expendituresto the spending poorest areas. Analysis of the benefit incidenceof government programs should also provide important information for the allocation of public - resources. 6. Expenditure/Wagepolicy Sustain the implementation of the public sector medium term wage policy. Review expenditures classified as "employment allowances" (which cover payments for overtime, per diems, travel, etc.) which have risen significantly over the pasttwo years. Examine whether there is scope to reduce expenditures on employment - allowancesanduse the savingsto increasebasic salaries. - BUI ET EXECUTIONASSESSM IT BASED ON THE CAGREPORT 7. Timelinesof CAG audit Take measuresthat would allowthe CAG to produce audit reports reports without delay. Suchmeasurescouldincludecontractingan increased - amount of work to outside auditors. 8. Computerbasedaudit Build capacity o f NAO to undertakecomputer based audits to ensure the integrity of the IFMS based accounting. Moving to a computer based auditing is also likely to reduce the time neededfor preparing the annual reports. 9. Responses to audit queries Measures should be taken to create the right incentives for accounting have remained low as a share officersto pay more attentionto these issues andfollow-up. o f total queries. &I;NMENTOFTHEBUDGET XOCESSWITHPRSCOREPRINCIPLES 10. CSO participation in the More active involvementof CSOs inthe externalreview phase of the External Reviewof the PER. PERprocess-perhaps throughtheNGO policy forum- shouldbe encouragedas part of a long-termstrategy to progressivelydevolvethe exerciseof externalaccountabilityof government to domestic actors. The Tanzanian government and civil society may wish to consider extending the PRS monitoring master plan to include a participatory monitoring and evaluation system in which civil society could play a leadingrole. xvi Issues Recommendation 11.II Limited feedback from Explore avenues to coordinate the information provided by the output- multiple monitoring systems basedbudgeting system inthe context of the MTEF and the performance- to budget process based budgeting exercise led by the Civil Service Department so as to allow efficiencyjudgments to influence allocative decisions at the MoF. More broadly, the links between findings of the poverty monitoring master plan and policy-making processes, including on the fiscal side, I should be strengthened. - PO7 RTY REDUCING EXPEND1 JRES 12. Dataon pro-poor expenditures The government scores high with regard to data production, availability, and dissemination, regarding both fiscal and poverty data. More effort is needed, however, in the use of poverty data in the fiscal policy-making process, the definition and consistent identification of "priority" - expenditures, and the collection of routine data by local authorities. 13. Resource Gap Significant gaps between requirements and budget ceilings for priority sectors and the planned decline in the share of government-hnded allocations to these sectors are cause for concern. It will be important to develop a clear medium to long term plan as to how the resource gaps - will be filled. 14. Definition of `priority' Government is encouraged to define the meaning of "priority" sectors regarding the allocation of domestic funds. This definition should, at the least, include a commitment to the maintenance of the share of priority expenditures in domestically-funded expenditure and of their absolute allocations in real terms. ii)Government i s also encouraged to examine the areas not defined as priority inthe PRS for savings to be re-allocated - to the priority sectors, given the funding shortfalls inthe latter. 15. Seemingly inconsistenciesin To enhance transparency, it will be important that the PRS progress report the data of PRS andPRS contain a precise definition of what is included in the priority sector progress reports. expenditures. In addition, it will also be important that sector PERs pay greater attention to monitor PRS expenditures and try to ensure - consistencyincoverage, definition, and source of data. 16. Use of `flexible' donor support The government should monitor closely the use of "flexible" donor support to ensure that it i s usedto enhance spending inthe priority sectors and to cover the significant resource gaps existing in these sectors, if the - PRSP andMDGtargets are to be met. RE\ IW OF THE ALIGN.1IENT0 SECTORALPERWORK WITH PRS 17. Participation ~~ ~ Clearer definition of the role of participation and setting of an agenda identifying steps in the budget preparation process where ministerial submissions to the MoF are discussed in the Sector Working Group (SWG). - Strengthened participationby the MoF inthe SWGs. 18. The role of the Sector In addition to the role of defining and overseeing the work on sector Working Groups PERs, the role of the SWGs should also include at a minimum: Review and discussion of sectoral submissions prior to their submission to the budget guidelines committee Review and discussion of sectoral budget proposals prior to their submission to the Ministry of Finance Review and discussion of the sectoral Medium Term Expenditure Framework Designation of contact person at the MoF for the priority sectors to participate in SWG and facilitate a more fluid communication with line ministries and sectoralworking groups. xvii - Issues Recommendation 19. Scope for enhancing the Consider reducing the frequency of in depth sector PERs and adopt multi usefulness of sector PER work year cycle through better integration with Develop minimal TORS for other than in depth PERs to fulfill annual government processes and budget and MTEF requirements and develop timetable that inputs are rationalization of scope and available at the right time of the budget cycle to feed into sectoral budget frequency submissions to the budgetguidelines, MTEF, and annual budgets. Carry out occasional in-depth PERs for non-priority sectors to enhance allocative and operational efficiency. In the medium-term, the government may consider extending the PER process to non-priority sectors in order to extend the benefits of the process in terms of openness, transparency and enhanced planning and - review to those sectors. 20. Review of sector PER work Pay particular attention to sectors with weaker PERs, notably, HIVIAIDS, shows large differences in but also Agriculture, Roads, and Education. Inmany cases weaknesses in their coverage and quality the PERs appear to be reflective of weaknesses in sectoral policy - frameworks. 21. Strengthening and There is an urgent need to strengthen the costing of sector and PRS ~ ~~ identification and costing of priority activities so that it can be a more useful tool for MTEF sector andpriority activities. preparation and priority targeting. After each annual cycle, the PRS, MTEF, and priority sector PER task teams or working groups should regularly meet and compare notes on how to improve their respective - roles and inputs into the following cycle. 22. Revising and/or Developing For sectors that haveno sector-specific targets that are consistent with the PRS Targets overall PRS targets (HIVIAIDS and Agriculture), developing targets, intermediateor proxy, if necessary, should be priority. For sectors that appear to be far from reaching their targets either due to financing constraints or absorptive and capacity constraints, it may be wise to review PRS targets for those sectors that have fallen behind (e.g., - water). 23. Capacity building at the LAs All sector PERs indicate that they face most serious implementation and level absorption capacity constraints at the LAs level. These capacity constraints include institutional management, financial management, financial control and accountability, and procurement skills. There is thus - a needto strengthen capacity building at the LA levels. 24. The rate of decentralization Given the capacity and the implementation constraints of LAs described and absorptive capacity at above, there is a need to review and orchestrate the pace of LAs. decentralization across the priority sectors. The combined transfer of fiscal responsibilities and resources to LAs will exert tremendous burden on LAs to manage resources and carry out programs. For. Since there is only so much human capital available in the short-run, it may be wise to link the rate of decentralization with the rate of capacity building at the - LA levels. 25. Follow-up of Sector PER Future sector PERs should be required to follow up on the recommendations recommendations of past PERs. This does not mean that all recommendations proposed by any of the sector PERs must be adopted. Rather, it means that recommendations need to be seriously considered - BUT ETMANAGEMENTISSUES and any actionstaken upon them should be reported. 26. Cash flow management IIRelax cash flow budgeting by providing quarterly budget releases to all snendinp units and make anv necessarv adiiirtm~ntrnt a niinrtprlv rnthpr thanmonthly basis. 27. Managing variability and Government could also attempt making budget design more flexible in mpredictability of donor cash order to better manage resource variability. One option could be to flows develop a conservative baseline budget by discounting donor commitments by, say, 10 percent and identifying items that would be financed only if disbursements exceededthe baseline xviii - Issues Recommendation 28. Information flows on donor Progress in enhancinginformation flows on development assistanceat the assistance to the central central government level needs to be sustained, with scope for government and especially harmonization of formats between departments in the Ministry of Fiance, local authorities need to be the ACCGEN, andthe sector ministries. further enhanced Efforts at the central government level should be extended to capture information on direct development assistance to local authorities. MoF needs to routinely get this information from PO-RALG and incorporate/synchronize it with what already exists in the aids flow data - base kept by the ExternalFinanceDepartment. 29. Effectiveness and value for Analytic work to assess the efficiency and effectiveness of technical money o f technical assistance assistanceand capacity building efforts should be a priority to provide the in enhancing government basis for developing a coherent strategy for the use of technical assistance - capacity and capacitybuilding efforts. 30. PRSand LA budget guidelines The guidelines could be improved further by: (i)putting more emphasis on national and PRS priorities (ii)clearly stating sector policies, priorities and strategies, albeit briefly, as mentioned earlier, requiring that LAs review past performance (covering revenue collection, expenditure, and inventory of projects and resources).and status of implementation of national priorities (iii)taking a longer term perspective (say 3 years) to budget guideline issuance, with only resource ceilings being provided on a yearly basis. Much of the present content of the budget guidelines concerningthe budget process and forms to be used could be codified in a - budget manual. 31. Creating a formal process of It i s recommended that both central and local budget processes are consultation or forum between reviewed with the objective to ensure better communication and sector ministries and LAs information sharing between the local authorities and the sectoral ministries throughout these processes. The establishment of a forum that brings together LAs, regions and ministries is thus key to enhance the budget guidelines for local authorities. Also ensure appropriate participation by LA representatives in the central government budget - process. 32. Synchronization of Central To ensure that the realignment of fiscal years yield the expected benefits, Government and LAs fiscal it will be important to: years. Carefully plan the transition for local authorities, who will have to prepare a budget covering the period January - June 2004 before they switch to the synchronized budget year with a budget for July - June 2005. Inorder to provide sufficient time for the preparation oflocal government budgets, the central government guidelines will need to have a firm indications of resources available to the local authorities. The synchronization of the budget year will also require a synchronization of central and local government budget guidelines, creating an opportunity for improved involvement of local government stakeholders in both processes. The MTEF will need to pay more attention to allocations to local authorities. xix 1. INTRODUCTION 1.1 As an integral part o f the Public Expenditure Review Process in Tanzania, an external review o f fiscal developments i s carried out every year. The external review, where external means external to government, is carried out by members of the PER working group under the leadership o f the World Bank. It focuses primarily on overall fiscal discipline, cross-sectoral budget allocations, and budget management issues, while sectoral PERs carried out by the sectors focus on allocative and efficiency issues within the priority sectors. Since this is an annual exercise, the report does not intendto provide a comprehensive analysis o f public expenditure issues in Tanzania. Its intent i s to provide an update on key development and to analyze a few selected issues identified by government and the PER working group inmore detail. 1.2 The theme o fthis year's external evaluation is "Managing Public Expenditurefor Poverty Reduction" and the report looks at public expenditure allocations and management issues primarily from the angle o f fiscal management as a key instrumentto implementTanzania's poverty reduction strategy. 1.3 The report i s divided into three parts. The first part o f the report provides information on macroeconomic developments and analyzes overall fiscal performance and strategic resource allocation in FY02 and the first half of FY03, and summarizes financial management issues arising from execution based on the report o f the Controller and Auditor General. The review o f aggregate fiscal policy concludes that fiscal policy remained consistent with the objective o f maintaining macro-economic stability. However, it also points to the need to keep employment levels inthe public sector under review and take measures to increase domestic revenue so as to generate sufficient funding for the implementation o f the PRS. The analysis o f strategic resource allocation points towards the continued focus on social expenditures. The section also highlights the fact that fiscal decentralization i s progressing only slowly and that most o f the resources are still controlled by central government. The latest reports o f the Controller and Auditor General show some improvement inpublic sector financial management. 1.4 Part I1o f this report analyzes the alignment o f the budget process with the PRS. Tanzania prepared its poverty reduction strategy in 2000, and has since prepared two annual progress reports. The budget and, more broadly, fiscal issues, lie at the heart o f PRSP implementation. The budget articulates more clearly than any other policy instrument a country's priorities and few policy reforms can be enacted without proper budgetarymanagement. Therefore, analyzing the budget management processes is a 1 good lens through which to assess progress toward PRSP implementation. The report contains thus detailed analysis o f the extent to which PRS principles and substance are reflected ingovernment's budgets and expenditure patterns. 1.5 An important element inbringing the PRS and the budgetprocess closer together was the re-establishment o f sector working groups in FY03 and the external evaluation undertook a review of both the functioning o f the sector working groups as well as the quality and relevance o f sector PERs. 1.6 Chapter 6 o f the report examines whether key principles o f the PRS process - country driven, results orientation, comprehensiveness, partnership oriented, and long term perspective - are also reflected in the budget process. Chapter 7 provides a very detailed analysis of priority sector expenditures in Tanzania. It documents a significant increase in priority sector expenditures during recent years. However, it also points out that these increases are driven by a general increase in the resource envelope and not so much by reprioritization o f public expenditures. Weaknesses in the information base on priority sector expenditures are documented and greater attention to consistency o f expenditures figures in various government documents i s encouraged. As sector PER work contributes the analytic foundation to translating PRS objectives and targets into budgetary allocations, chapter 8 reviews whether they indeed fulfill this function and highlightsstrengths and weaknesses ofwork carriedout inFY03. 1.7 The third part o f the report covers selected issues inbudget management. Issues discussed include the case for maintaining a cash budget system, the management o f foreign aid, and the role o f the budget guidelines as a link between central and local government budgeting. With respect to the cash budget system, the section makes the case that recent budget management reforms provide government with more sophisticated means to achieve macro-fiscal objectives and that the cash-budget system could be further relaxed without runningmajor fiscal risks. On external assistance, we commend the continuing efforts by donors and government to integrate donor resources into the budget. One o f the key challenges remains inthe area of technical assistance, both with respect to better capturing inthe government accounts, but also with respect to enhancing the efficiency and effectiveness o ftechnical assistanceand capacity buildingefforts. The section also summarizes work on the macro-economic impact o f aid 2 PART I:REVIEW OF FISCAL PERFORMANCE 2. THE MACROECONOMIC BACKGROUND RealEconomicGrowth 2.1 Tanzania's macroeconomic performance continued to improve in2001 and 2002. The annual growth rate reached 5.6 percent in 2001 and rose to 6.2 percent in 2002 despite a substantial deterioration inthe terms of trade. (Table 1) Table 1. Real GDP annual growth rates (at constant 1992 prices), 1995-2002 source: Economic Survey, 2002 2.2 The rise in real GDP growth in 2001 was the result of a relatively strong performances inagriculture, mining, wholesale, retail trade, as well as the manufacturing sector. The agriculture sector grew by 5.5 percent in2001, up from 3.4 percent in 2000, with most of the growth emanating from crop productionand fishing. The sector, which accounts for about 44 percent of the GDP, grew by 5.0 percent in 2002. The growth of the mining sector also remained strong at 15 percent in 2002 and increased its contribution to the GDP to 1.8 percent from a level of 1.2 percent in 1997. Wholesale and retail trade (including tourism) grew by 7 percent in 2002 compared to 6.7 percent in 2001. Industry -incorporating manufacturing, utilities, construction, transport and 5 communication, grew at an average o f 6.7 percent in 2001 and 11.O percent in 2002. Increased government expenditure in the priority sectors enhanced the growth o f public administration and other services from 2.7 percent in 1997 to 4.1 percent in2002. Table 2 below summarizes sector contributions to GDP inTanzania. EconomicActivity 1995 1996 1997 1998 1999 2000 2001 2002 Agriculture 47.1 48.0 46.8 44.8 45.1 45.0 44.8 44.1 Miningand Quarrying 1.3 1.1 1.2 1.5 1.4 1.5 1.7 1.8 Manufacturing 7.2 7.4 6.9 7.4 7.3 7.5 7.4 7.4 Financial and Business Services 12.6 13.1 13.3 14.4 13.7 13.7 14.2 14.4 Public Administration and Other 9.1 8.7 10.6 11.0 10.9 10.6 10.3 10.4 Services Less: Financial Services Indirectly -4.0 -4.0 -3.2 -2.8 -2.4 -2.3 -2.1 -2.0 Measured GDP at Factor Cost 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 2.3 While overall investment as a percentage o f GDP did not increase significantly between2000 and 2001, foreign direct investment (FDI) rose by 16.6 percent from US$ 192.8 million to US$ 224.8 million. GeneralPriceDevelopments 2.4 Tanzania has been quite successful in containing inflationary pressures. The inflation rate declined from 27.5 percent in 1995 to 4.5 percent in 2002. During this period, although non-food inflation remained on an upwardtrend, this was outweighed by declining food inflation resulting in a drop o f the general price level. This good performance in price movements has mainly been due to sound macroeconomic policy, especially sustained fiscal discipline and a tight monetary policy stance. Giventhat food has a big weight inthe National Consumer Price Index (NCPI), the favorable food supply situation inthe country played a significant role inthis general price performance. Figure 1 summarizes the trend o fthe inflation rate inTanzania. 6 Figure 1. Annual headline, food and non-food inflation July 2001- Dec. 2002 10.0 - 9.0 8.0 - 2e 5 7 . 0 - * - - - 6.0 >-* - - I - , 5.0 dh 4.0 - % - - - - - - - - - 3.0 - - - I - - - . 0.0 - 7 - --- Headline Inflation Food Inflation -Non-food Inflation Source: BOT Monetary and Financial Developments 2.5 Duringthe periodunder review, there was a general increase inReserve Money (MO). The increase was mainly due to an increase ingovernment deposits following an increase in revenue collection. Although Net Foreign Assets (NFA) increased slightly, this was outweighed by the drop in Net Domestic Assets (NDA). By December 2002, Gross Reserves o f the Bank o f Tanzania had increased from US$ 1.4 billion during the previous period to US$l.5 billion -- equivalent to 8 months o f imports o f goods and non- factor services. Broad money (M2) went up slightly during the year mainly on account of the increase o f the NFA of the banking system. Although credit to the government declined, that to the private sector increased quite substantially. The extended broad money (M3) showed an increasing trend possibly reflecting portfolio diversification in favour o f foreign assets. The Tanzanian Shilling has been gradually depreciating (in nominal terms) from Tshs 575 per US$ in 1995 to Tshs 979 per US$by the end o f December, 2002. Ingeneral, during the period under review, all the monetary aggregates were on an up-ward trend. Broad money supply grew by 23.2 percent down from 24.9 percent o f the previous period, while annual extended broad money grew by 24.0 percent. Figure 2 below summarizes the picture. 7 Figure 2. Annual growth ratesin monetaryaggregates,M3, M2, and MO 45.0 40.0 35.0 30.0 25.0 20.0 15.0 10.0 5.0 I &'Q' +&'Q$ Source: BOT 2.6 Since 2000, total commercial bank deposits, comprised of demand, time, savings and foreign currency deposits have also experienced an up-ward trend (see Figure 3). Foreign currency deposits constituted the largest proportion o f the deposits, accounting for 34.8 percent of total deposits by December 2002. Figure 3. Depositsof commercialbanks 1600.00 , 1400.00 1200.00 1000.00 800.00 600.00 0.00 t - Foreign currency deposits - - Total deposits TZS denominated deposits Source: BOT 8 2.7 On average for the period 2001-02, the overall structure o f interest rates on domestic currency exhibited a marginal upwardtrend. The overall time deposit rate rose marginally from 3.2 percent. The savings deposit rate remained fairly constant at around 3.2 percent. Lending rates for 1-2 year, 2-3 year, and 3-5 year loans rose from 15.6 percent, 12.1 percent and 16.3 percent to 16.2 percent, 13.0 percent and 17.5 percent, respectively. Consequently, the overall lending rate rose from 15.8 percent to 17.0 percent by the close o f 2002. It should be noted that for the past three years, the average lendingrate was as highas 20 percent. The current developments, however, have slightly increased the interest rate spread betweenthe lendingrate and the deposit rate. ExternalSector Developments Trade Account 2.8 For the year ending in December 2002, the trade account improved by 20.3 percent to a deficit o f USD 598.4 million from a deficit o f USD 751.3 million. This was mainly due to an increase inexports coupled with a decrease inimports (Table 3). Table 3. Trade account Value m Mdl.of USD 2002 % December % January-December % November December(p) Change 2001 2002p Change 2001 2002p Change Exports 166.2 169.5 2.0 132.6 169.5 27.8 1,455.8 1,541.1 5.9 Goods 107.8 99.8 -7.4 82.3 99.8 21.4 776.6 877.1 12.9 Services 58.4 69.6 19.3 50.3 69.6 38.4 679.3 664.0 -2.3 Imports 159.4 168.4 5.6 158.4 168.4 6.3 2,207.2 2,139.5 -3.1 Goods 114.5 123.6 7.9 112.2 123.6 10.2 1,558.3 1,504.3 -3.5 Services 44.9 44.8 -0.4 46.2 44.8 -3.1 648.9 635.2 -2.1 Trade Balance 6.8 1.1 -83.7 -25.8 1.1 -104.3 -751.3 -598.4 -20.3 * Year ending December Note: p =Provisionaldata Source: Bank o f Tanzania, TRA 2.9 For 2001-02, overall exports o f goods and services increased by 5.9 percent. The biggest increase was recorded on the exports o f goods which increased by 12.9 percent, Exports o f non-factor services, however, declined slightly by 2.1 percent. The main contributors to the traditional exports are tobacco (25 percent), cashewnuts (23 percent), coffee (17 percent), Tea (15 percent), Cotton (14 percent) and Sisal (3 percent) (Figure 4). 9 Figure 4. Structure of traditional exports, 2002 1 Cloves Sisal Tobacco 25% 15% 2.10 Non-traditional exports accounted for 75 percent o f the total exports. Gold alone i s the leading non-traditional export accounting for about a half o f all the total non- traditional exports (Figure 5). Figure 5. Structure of non-traditional exports, 2002 Other exports Fish, and fish Gold P* 49% Manufactured goods 10% r Horticultural \Other Minerals 2% 6% 2.11 Imports during 2002 declined by 3.1 percent to USD 2,139.5 million compared with the corresponding period in 2001. All categories of imports: capital goods, intermediate goods, and consumer goods registered a fall o f 2.7 percent, 4.4 percent and 3.7 percent, respectively. The 4.4 percent decline in intermediate goods imports i s largely explained by the decline in oil prices in the world market from USD 223.4 per ton in 10 2001 to USD 220.2 per ton in 2002. The decline in consumer goods imports is partly attributed to a 13.7 percent decline in food and foodstuffs imports following improvement inthe domestic food supply situation. Debt Developments 2.12 By end-December 2002, Tanzania's debt stock reached USD 8,178.0 million out o f which extemal debt was USD 7,384.7 million or 89.2 percent and domestic debt was USD 891.6 million or 10.8 percent. O f the total committed extemal debt, the disbursed outstanding debt (DOD) was USD 6,4 13.4 million and committed un-disbursed debt (CUB) was USD 1,616.5 million. Interest arrears alone reached USD 929.2 million at end-December 2002. Overall, the total debt stock during the year increased slightly by 1.2 percent mainly due to the reconciliation exercise with the African Development Bank Group coupled with exchange rate fluctuations. The profile o f disbursed outstanding debt by creditor category shows that 59.7 percent and 32.8 percent of the total debt is owed to multilateral and bilateral creditors, respectively. The proportion o f total debt owed to commercial sources and export creditors was 4.7 percent and 2.7 percent, respectively. The composition o f disbursed outstanding debt by borrower category shows that the Central Government i s the largest borrower accounting for as muchas 92.0 percent o f the total debt. The proportion o f parastatal andprivate companies inthe total debt stock i s 2.4 percent and 5.6 percent, respectively. 2.13 Regarding domestic debt, the total debt stock owed by the government o f Tanzania at end-December 2002 stood at TZS 870.5 billion. Analysis o f the domestic debt stock by instruments shows that government bonds constitute 49.4 percent of the total domestic debt. When Treasury Bills are taken into account, total government securities constitute 94.6 percent o f the debt stock, while other government debts and interest arrears constitute 5.3 percent and 0.1 percent of the debt stock, respectively. Commercial banks are the largest creditors holding 42.3 percent o f the debt stock. Pension funds come in second with 25.5 percent o f the debt. Government debt to the Bank of Tanzania accounts for 23.2 percent of the total domestic debt while private and nonbank financial institutions account for 3.5 percent and 3.0 percent of domestic debt, respectively. The rest o f the creditors: insurance companies, other official entities and the BOTspecial funds altogether account for 2.5 percent ofthe debt. 11 12 3. OVERALL FISCAL PERFORMANCE Aggregate FiscalDiscipline 3.1 Overall fiscal performance remained solid in FY02 and FY03. Government continued to implement a tight fiscal policy during FY02 and overall fiscal performance was in-linewith the budgettargets. The recurrent deficit was 1.5 percent of GDP against 2.5 percent projected in the fiscal framework at the beginning of FY02. Despite a shortfall of 66 billion in external grants (mainly project grants), government was able to limitthe overall deficit after grantsto 1.1percent of GDP against abudget estimateof 1.2 percent of GDP. The key operative policy target was zero net domestic borrowing, enforced by the continued use of a cash budget system, which limits spending each month to available resources. Table 4 details how the fiscal deficit was financed. Government was able to clear off all domestic arrears amounting to T.Shs. 59 billion. Net domestic borrowings were negative as inprevious years. Table 4. Financingof the fiscal deficit(O hof GDP), FY98-FY03 Central GovernmentOperations FY98 FY99 FYOO FYOl FY02 FY03 FY03 Actual Actual Actual Actual Actual Budget Proj. Balancebefore Grants (checksissued or -3.7 -5.6 -7.8 -5.4 -5.6 -10.0 -7.9 commitment basis) Grants (checks issued or commitment basis) 3.0 4.0 4.5 3.8 4.5 5.7 5.3 Program (import support) 0.7 1.2 1.4 1.5 2.2 1.9 3.2 Project 2.3 2.8 3.O 1.6 1.6 3.0 1.6 HIPC Relief (IMF, WE3 and AfDB) 0.1 0.6 0.7 0.8 0.5 Balanceafter Grants (checks issuedor -0.7 -1.6 -3.3 -1.6 -1.1 -4.3 -2.6 commitment basis) Foreign Loans (net) 1.o 0.4 1.5 1.2 1.4 3.7 2.9 Program loans (import support) 1.3 0.6 0.9 0.6 1.o 1.2 1.7 Development project loans 0.8 1.1 1.9 1.7 1.2 3.6 2.1 Amortization -1.1 -1.2 -1.3 -1.1 -0.8 -1.1 -0.9 BalanceAfter Grants and Foreign 0.3 -1.2 -1.8 -0.4 0.4 -0.6 0.3 financing (checks cleared or cash basis) Domestic (net) -0.3 1.3 1.8 0.4 -0.3 0.5 -0.2 Bank -0.9 0.0 -0.4 -0.2 -0.7 0.2 -0.2 Nonbank (net o f amortization) 0.5 -0.1 0.5 0.2 0.4 0.1 0.0 Privatization Funds 0.1 0.2 0.2 0.4 0.0 0.2 0.0 Parastatalrecapitalization 0.0 1.2 1.5 0.0 0.0 0.0 0.0 Source: IMF and Tanzanian authorities 13 3.2 The FY03 budgetframework provides for a more relaxedfiscal policy and a more than 110percent increase inthe development expenditure over FY02.' The overall deficit after grants, is budgeted to increase to 4.3 percent o f the GDP at the end o f the year. Duringthe first six months of FY03,expenditures, particularly development expenditure, have beenvery slow while revenuereceipts and grants have kept pace. Duringthe second half o f FY03 recurrent expenditures accelerated, however, development expenditure remained significantly below the budgeted amounts. Thus, the deficit after grants is projectedto be only 2.6 percent o fthe GDP for FY03 (Table 4). DomesticRevenue 3.3 Tax revenuewas on-target but is stagnant at around 12 percent of the GDP. The government revenue target o f T.Shs. 1043 billionwas marginally exceeded inFY02. Highpetroleum prices and streamlining of tax administration, particularly the creation of a Large Taxpayers Department, helped to improve efficiency o f collection. While revenue has grown in the last two years in nominal and real terms, the performance appears flat interms o f GDP (about 12 percent inboth FYOl and FY02, Table 5), which means that it has grown at the same rate as the GDP. The revenue target for FY03 i s T.Shs. 1172billion, about 12.3 percent o f the estimated GDP. Actual revenue outturn for FY03 is projected to reach 12.5 percent of GDP, exceeding the budgetedamount mainly on account o f significant improvements incollection from large tax payers through better compliance. Table 5. Revenue performance (YOof GDP) FY98-03 Item FY98 FY99 FYOO FYOl FY02 FY03 FY03 Actual Actual Actual Actual Actual Budget Proj. Total Revenue 12.0 11.5 11.3 12.2 12.3 12.3 12.5 Tax Revenue 11.0 10.3 10.1 10.9 11.0 11.2 11.4 Taxes on Imports andExports 3.5 3.6 3.2 4.8 4.7 5.0 5.O VAT andExciseonLocal Goods 2.7 2.7 2.6 2.5 2.5 2.5 2.4 Refimds 0.0 0.2 0.1 0.1 0.4 0.0 0.0 IncomeTaxes 2.9 2.7 3.0 2.6 2.7 2.7 2.9 Othertaxes 1.9 1.2 1.3 1.1 1.1 1.1 1.2 Non-tax Revenue 1.o 1.2 1.2 1.3 1.2 1.1 1.1 Source: IMF andTanzanianauthorities 3.4 Need for better revenue performance. Current expenditure on priority sectors i s determined by constrained availability o f resources rather than the demand placed by PRSP objectives.2 Inorder to achieve PRSP objectives within the stipulated time frame, it i s imperative that government expenditure on priority sectors grow at a faster rate. While external aid has been on the increase, revenue performance also must improve much beyond the current level o f around 12 percent o f GDP. Better revenue performance is ' This reflects in part, the Songo Songo gas project, the PDEP, and a better recording of in-kindproject grants providedby donors. Reliable cost estimates of achieving PRSP targets are not available. A study carried out by Bevan and EME in2001 approximatedthe 'financing gap' i.e., the excess of resource requirement for meetingPRSP targets inthe plannedtime periodover availableresources, at 3 percentof GDPper year. 14 also necessary from the point o f view o f reducing risks from inherent volatility o f aid flows and ultimatelyreducing aid dependency. 3.5 Tax Exemptions. Tax exemptions have been attracting active attention of the government. Measured as a percentage o f GDP, they increased sharply from 0.9 percent o f GDP inFY97 to 4.5 percent o f GDP inFY99 and declined subsequently to 2.8 percent by FYOl (Figure 6). Much o f the increase can be explained by the granting of tax exemptions to the mining sector in 1997 and the enactment o f the mining act, which fostered significant investments in the mining sector in Tanzania. After the initial investments for gold mining were put in place in the late 1990s, investment activities tapered out which also let to a decline inthe amount of tax exemptions. Figure6. Tax exemptions(O hof GDP), FY97-FY01 5.0 4.5 4.0 3.5 E 3.0 2.5 0 $ 2.0 1.5 1.o 0.5 0.0 I FY97 FY98 FY99 FYOO FYOI Source:TRA 3.6 As figure 7 illustrates, changes in the value o f exemptions were only marginally reflected in changes in tax revenue. This i s the consequence o f the fact that much o f the exemptions are concentrated in the mining sector, with changes in the value o f tax exemptions beingdetermined by the activity inthe mining sector. Prior to the granting o f tax exemptions to the mining sector activity has been low and subsequently mining activity had no impact on tax revenue due to the tax exemptions. It is also difficult to establish what mining activity would be in the absence o f the tax exemptions and it i s thus not possible to state unequivocally that inthe absence o f exemptions revenue would increase by the amount o f exemptions removed. Further analysis will also be required to ascertain whether a higher contribution of the mining sector to tax revenue can be expected once initial investments have beenfully written off againstprofits. 15 Box 1. Fiscalincentivesinthe miningsector Input Taxes Import Duty, Sales Tax and Excise Duty Exemptionfor miningequipment and supplies directlyrelatedto the operations of the projectgranted up to one year after the start of production. A cap limit of 5 percent ImportDuty and 5 percentSales Tax on importsof equipmentandsupplies shall apply thereafter. FullVAT Exemptionfor purchasesof inputsandsupplies (wherethe product is exported). The Mineral Rights Holder shall be exempt from Domestic Withholding Tax on goods and services suppliedby him. However,he shall be obligedto withholding tax on domestic goods or servicespurchased by him. Withholding Tax or Technical Tax on Technical Service Payments to Subcontractors both resident and non-residentand management fees will be 3 percent of gross payment (unless management fees exceed 2 percentof operatingcosts whentax will be 20 percent), the tax withheld will satisfy the incometax liability o f the subcontractorinrespect of the technicalservice. Output Taxes Royalty. 3 percent on netbackvalue at the mine of the minerals sold(5 percentfor diamonds). Gemstones -3 percent on exports of rough stones (to discourage exports of rough stones). Nil for cut and polished stones (to promote value adding and mitigate smuggling). The aim is to develop Tanzania as a gemstone center. No ExportTax or Stamp Duty onreceiptsapplyto the sales of minerals. Profit Taxes "Specified Minerals" provision in the Income Tax Act abolished and all minerals extracted under the MiningAct treatedalike. Corporation Tax on income from mining activities at the standard rate determined for all types of investmentsinTanzaniainthe context of any revisionofthe InvestmentAct. The expenditureupona taxpayer's interest inone mine may be deductedfromthat taxpayer's incomefrom an interest in another mine, but miningexpenditure may not be deductedfrom income from other kinds of business (and vice versa). In other words, there will be a ring-fencingaroundthe mining sector but none within it. Depreciation Allowances of 100 percent will be available on all mining capital expenditure. All types of assetswill bepooledanddepreciatedat the same rate. At the end of eachtax year the balanceof UnrecoveredCapitalExpenditure inrespect of a Mining License (or other right to develop a mine) will receive an addition o f 15 percent capital allowance each year; no other investmentallowance is to apply. Lossesmay be carriedforward for recoverywithout limit. Accounts for tax purposesmay be kept inU S dollars. Withholding Tax onPaid-Out dividends and Tax on DistributionofBranchProfitsto non-residentswill be 10percent. Interest Withholding Tax will not apply to interest payments by mining rights holders on foreign borrowingsat arm's length. The Statewill not seek concessionalstate participation. 16 Figure 7. Change in tax revenue and exemptions(YOof GDP), FY98-FY01 3.5% 3.0% 2.5% 2.0% a 1.5% 1.0% % 0.5% 8 0.0% -0.5% -1.0% -1.5% -2.0% FY98 FY99 FYOO FYOI Source: TRA 3.7 Following an analytical review o f tax exemptions in the years 1999/2000 and 2000/2001carried out by the TRA, the Finance Minister referred to the need for reviewing customs, excise duty and VAT exemptions in his Budget Speech for 2002/03. The introduction o f a Treasury Voucher System starting January 2003 for indirect tax exemptions to non-religious NGOs and import of motor vehicles by government officials will allow the government to treat tax exemptions as tax expenditures and extend them the same level of scrutiny during the budget process. Publishing the names o f individuals, companies, and institutions that have been exempted from paying taxes, along with the amount exempted, will introduce transparency in the operation o f exemptions. 3.8 We would encourage the government to consider extending the voucher scheme to other categories of import tax exemptions in the next fiscal year, as it gains more experience with its implementation. A comprehensive evaluation o f the costs and bene$ts o f various exemptions, particularly to the mining sector would also be in line with the findings ofthe review carried out bythe TRA (Figure 8). 17 Figure 8. Customs exemptions to private companies by sector 2000/01 I Manufacturing ning 1% 6% Source: Analysis of tax exemptions for the years 199912000 and200012001, TRA, 2002 3.9 During FY03, the government has taken some steps towards reducing revenue leakage, curbing tax evasion and expanding the tax base. The Budget Guidelines also mention specific medium-term measures such as encouraging small enterprises to shift from the informal to the formal sector, streamlining motor vehicle registration and improving fiscal reporting. It i s not clear what the impact of each of these measures on revenue will be. Three recent studies done by Massawe, Emst and Young, IMF, and REPIM, make several suggestions on improving the tax performance. Government is encouraged to consider the suggestions in a systematic manner and spell out a medium- term strategy for reforming the tax system, and making other policy decisions for improving domestic resource mobilization, in consultation with stakeholders. Early insights from such an exercise could feed into the budget for the next fiscal year. Another area related to tax efforts that needs better understanding is the equitability ofthe existing tax policies and their impact on growth. DonorAssistance 3.10 Table 6 summarizes extemal assistance inthe recent years. A few pattems stand out: 0 Extemal assistanceis on the rise. 0 Both grants and concessionalborrowing are increasing. 0 A greater share of grant support is provided as program assistance. 3.11 External assistance (including HIPC relief from multilateral donors) increasedto 5.9 percentof GDP inFY02 and is projected to reach 7.9 percent of GDP inFY03, an all- time high inthe recent years. About half of it is expected to be in the mode of program support. 18 Table 6. Composition of external assistance (YOof GDP), FY98-03 FY98 FY99 FYOO FYOl FY02 FY03 FY03 Actual Actual Actual Actual Actual Budget Proj, Grants and Loans 4.0 4.4 6.0 5.0 5.9 9.4 7.9 Grants 3.0 4.0 4.5 3.8 4.5 5.7 5.3 Program 0.7 1.2 1.4 1.5 2.2 1.9 3.2 Project 2.3 2.8 3.0 1.6 1.6 3.0 1.6 HIPC grant relief 0.0 0.0 0.1 0.6 0.7 0.8 0.5 Loans (net) 1.o 0.4 1.5 1.2 1.4 3.7 2.6 Loans 2.1 1.6 2.8 2.3 2.2 4.8 2.9 Program 1.3 0.6 0.9 0.6 1.o 1.2 1.7 Project 0.8 1.1 1.9 1.7 1.2 3.6 2.1 Amortization -1.1 -1.2 -1.3 -1.1 -0.8 -1.1 -0.9 Source: IMF and Tanzanian authorities GovernmentExpenditures 3.12 Governmentexpenditurewas 18.3 percent o f GDP (against abudgettarget o f 18.8 percent) in FY02 compared to 17.6 percent of GDP in FYO1. Lower-than-planned spending reflected a tight control over recurrent expenditure, but also a shortfall in receiptsof foreignproject assistance. There were also some delays inprocurementdue to new regulations, exacerbated by the late availability of cash releases from the MoF, towards the end o f the year. The increase was mainly on goods and services, while development expenditure fell slightly, and expenditure on the wage bill remained at around4 percentof GDP. Total expenditureis budgetedto grow to more than 22 percent o f GDP in FY03, driven mainly by a large increase in the foreign-financeddevelopment expenditure, a higher wage bill, and increasedspending on goods, services and transfers. Actual expenditureoutturn for FY03 is projectedto be below budget, mainly because of less developmentexpenditurethan budgeted. Spending on wages and salaries i s likely to be inline with the budgetedamountswhile the outturn on interest expenditureis likely to be belowandthat for spendingon other goods and servicesabove budget. Table 7. Government expenditure (YOof GDP), FY98-03 FY98 FY99 FYOO FYOl FY02 FY03 FY03 Actual Actual Actual Actual Actual Budget Proj. Total expenditure and net lending 15.7 17.0 18.9 17.6 18.3 22.3 20.4 Recurrentexpenditure 11.0 11.0 11.8 13.0 13.8 14.6 14.9 Wages and salaries 4.3 3.7 4.2 4.1 4.0 4.3 4.3 Interestpayments 2.3 1.6 1.9 1.7 1.4 1.4 1.2 Domestic 1.o 0.6 1.2 1.o 0.8 0.6 0.5 Foreign 1.3 1.o 0.7 0.7 0.7 0.8 0.7 Other goods, services& transfers 4.5 5.7 5.8 7.2 8.3 8.9 9.4 Clearanceof domestic arrears 0.8 0.8 0.1 0.4 0.7 0.1 0.0 Bank andparastatalrecapitalization 0.0 1.2 1.5 0.1 0.0 0.1 0.0 Dev.expenditureand net lending 3.8 4.1 5.3 3.8 3.4 7.6 4.6 Domesticallyfinanced 0.5 0.3 0.3 0.5 0.6 1.o 1.0 Foreign-financed 3.4 3.8 5.0 3.3 2.8 6.6 3.6 Net lending 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Expenditurefloat 0.0 0.0 0.3 0.3 0.4 0.0 0.8 Source: IMF and Tanzanian authorities 19 3.13 The wage bill stayed close to 4 percent o f the GDP inFY02 despite an increase in both civil service employment and average wage. After falling for several years the civil service employment grew by more than 3 percent, by the end o f first half year in FY03, driven mainly by a large increase inthe number o f teachers (Table 8). This increase i s as expected inview o f the education sector objectives inPRSP. Civil service employment i s now 6 percent higher than the targeted number in the medium-termpay reform strategy. The size of the TGOS needs to be kept under review as it has grown by more than 11 percent in the lastfour years. Table 8. Civil service employmentFY99-FY03 (December of each year) q Salary Scale FY99 FYOO FYOl FY02 FY03 FYOO FYOl FY02 FY03 Percenta e change TGOS 35651 34806 35001 37530 38829 -2.4 0.6 7.2 3.5 TGS 63787 63310 59994 56838 57095 -0.7 -5.2 -5.3 0.5 TGTS TPSW + TGPSW 122215 118868 119566 122148 130158 -2.7 0.6 2.2 6.6 36190 34821 36448 39823 38208 -3.8 4.7 9.3 -4.1 OTHERS 12785 11381 8837 772 775 -11.0 -22.4 -91.3 0.4 TOTAL 270628 263186 259846 257111 264965 -2.7 -1.3 -1.1 3.1 Target under Medium-term Pay 258543 249,000 Reform Strategy Notes: TGOS -Tanzania Govemment Operational Service, TGS -Tanzania Govemment Service, TGTS - Tanzania Government Teachers' Service, TPS -Tanzania Protective Service, Tanzania Government Protective Service Source:CSD 20 4. STRATEGIC RESOURCEALLOCATION What Are PublicFundsBeingUsedFor? 4.1 This section reviews developments in the allocation o f resources in FY02 and FY03 as well as actual expendituresduringFY02 and the first half o f FY03. Table 9. Compositionof publicexpendituresas YOof GDP, FY96-FY03 VOTE HOLDER I FY96 FY98 FYOO FYOl FY02 FY02 FY03 FY03* Actual Actual Actual Actual Budget Actual Budget Actual Recurrent Expenditures 12.5 13.1 12.8 13.6 16.6 14.9 15.8 5.8 Debt Service 3.7 4.9 4.2 3.4 4.0 3.2 3.4 0.8 Supply Votes 8.8 8.3 8.6 10.1 12.7 11.7 12.4 5.0 Recurrent Central 6.5 6 6.3 7.4 10.0 8.8 9.5 3.7 Recurrent Regions and Districts** 2.3 2.3 2.3 2.7 2.6 2.9 2.9 1.3 DevelopmentExpenditure 0.5 1.6 1.5 2.1 3.5 2.7 7.5 0.5 Total Expenditure 13 14.7 14.3 15.7 20.1 17.6 23.3 6.3 ***July -December 2002 Transfers from the central government to the regions and local authorities. Source:Appropriation Account (FY96 - FY02), Expenditure Release Reports (FY03) 4.2 The appropriation accounts show an increase in total expenditure (including debt service) from 15.7 percent of GDP in FYOl to 17.6 percent o f GDP in FY02. Most o f this increase (1.3 percentage points of GDP) was inthe form o f recurrent expenditures, while development expenditure captured in the appropriation accounts increased by 0.6 percentage points o f GDP. The budget for FY03 projects a further increase in overall spendingto 23.3 percent o f GDP with development spending increasing from 2.7 to 7.5 percent of GDP and recurrent expenditure increasing from 14.9 to 15.8 percent o f GDP. The massive increase in development expenditure reflects both improved capture o f ongoing assistance in the budget but also the scaling up o f activities, especially in the infrastructure and energy sectors. 4.3 The share o f recurrent expenditures used for debt service payments declinedfrom about 3.4 percent o f GDP in FYOl to 3.2 percent in FY02, reflecting debt relief received under the HIPC initiative. As in previous years, most of the increase in recurrent spending (1.4 percent o f GDP) accrued to the central MDAs, while centrally fimded The definition o f recurrent expenditure in this section uses the government o f Tanzania classification, which includes total debt service payments as part of Consolidated Fund Services, while the classification used in the previous section includes only interest payments but not amortization as part o f recurrent expenditures. There is are also significant differences between the estimates of development expenditure in the CGO and development expenditure recorded in the appropriation accounts. While the figure for development expenditure shown in the CGO is an estimate based on commitments, the appropriation accounts only show development assistance which i s actually recorded inthe government books. 21 4.4 expenditures by the regions and districts increased only by 0.2 percentage points from 2.7 percent to 2.9 percent o f GDP. The budgeted increase inrecurrent expenditure for FY03 i s focused on central government expenditures, while allocations to regions and districts remain stagnant as a percentage o f GDP. Since transfers to local authorities provide resources for the delivery o f decentralized services in the priority sectors covering education, health, water, and agriculture, the fact that these expenditures have increased at a much slower pace than expenditures by the central government i s o f concern. However, it should also be noted that not all central government inputs to service delivery at the local level are captured as transfers to local authorities. Some recurrent expenditure funding services under the responsibility o f local authorities, such as that for the provision o f drugs or funds earmarked for the implementation o f the Primary Education Development Program, i s still maintained by the sectoral ministries. Donor financed expenditures for service delivery at the local level are, if at all, typically captured in the development budget for the sectoral ministry. Government is urged to pay close attention to ensuring that all transfers to districts are indeed shown as such in the budget. This would send appropriate signals as to the progress in fiscal decentralization and would allow for greater transparency inthe distribution o f resources across districts. ! I Table 10. Functional allocation of recurrent expenditure (actuals, YOof GDP), FY96-FY03 Sector I F?Z: 1 F:: I FYT"p- FYOl FY02 FY02 FY03 FY03 * Actual Actual Actual Actual Budget Actual Budget Actual Administration 2.3 5.0 3.5 3.6 1.3 Defence and Security 2.4 2.2 2.0 2.2 2.1 2.1 2.3 ~ 1.o Social Services 3.5 3.6 3.6 4.4 4.5 4.7 4.8 2.1 Economic Services 0.2 1 0.4 0.7 1.o 0.8 1.o 1.2 0.4 Productive Services 0.5 0.3 0.3 0.4 0.4 0.4 0.5 0.2 Supply Votes 8.7 8.2 8.6 10.1 12.7 11.7 12.4 5.0 Consolidated Fund Services 3.5 4.0 3.2 3.4 0.8 Total Recurrent 3.7 4.9 4.2 )Expenditures II II II * 12.5 13.1 12.8 13.6 16.6 14.9 15.8 5.8 Expenditures for period July - December 2002 Souice: Appropriation Accounts (FY96-FY02), Expenditure ReleaseReports (FY03) 4.5 Table 10 presents the functional classification o f recurrent expenditures funded from the central government budget for the period FY96-FY03. Expenditures on the social sectors, which had remained fairly constant at around 3.6 percent o f GDP over the period FY96 to FYOO increased to 4.4 percent of GDP during FYOl in response to enhanced HIPC debt relief and the introduction o f the PRSP. They increased further to 4.7 percent o f GDP during FY02 and are budgeted to increase to 4.8 percent o f GDP in FY03. The increase in spending on the social sectors was facilitated by the overall increase in recurrent expenditure. Its share in recurrent spending on supply votes has remained at 40 percent between FY96and FY02. 4.6 Expenditures on administration increased from 2.3 percent o f GDP inFYOl to 3.5 percent inFY02. A significant part (about T. Shs. 90 billion shown under the votes o f the Accountant General and the Ministry o f Finance) o f this increase i s related to the settling parastatal debts and contingent liabilities and the clearance o f arrears to suppliers 22 incurred by all ministries. Administration also includes the vote for the President's Office - Regional Administration and Local Authorities which in FY02 and FY03 contains large amounts (T.Shs. 14 and 20 billion, respectively) for transfers to the local authorities for the funding o f capitation grants in primary education. In addition, administration also includes expendituresfor priority areas related to accountability inthe public sector such as the OCAG, the Judiciary, the Civil Service Reform Department, and the Ministry o f Lands. Expenditures on defense and security have declined from 2.4 percent o f GDP in FY96 to 2.1 percent in FYOI. Expenditures on economic and productive services remained constant as a percentage o f GDP between FYOl and FY02. However, the budgetfor FY03 foresees increased spendingfor these two areas. Table 11. Social sector recurrent expenditures (actuals, 'YOof GDP), FY96-FY03 1 1 1 Sector FY96 FY98 FYOO Actual Actual Actual Education 0.3 0.4 0.3 Health 0.5 0.6 0.6 0.6 0.2 Water 0.1 0.1 0.1 0.2 0.1 IScience,Technology & HigherEducation 1; i: 0.6 0.6 0.6 0.6 0.3 Regions 2.3 2.3 2.7 2.6 2.9 2.9 1.3 Total Social Services 3.5 3.6 3.6 4.4 4.5 4.7 4.8 2.1 * July-December2002 Source: AppropriationAccounts (FY96-FY02), ExpenditureReleaseReports (FY03) 4.7 Recurrent expenditures in the social sectors have increased from 4.4 percent o f GDP in FYOl to 4.7 percent o f GDP in FY02. The budget for FY03 provides for an increase in social spending in line with real economic growth. Most of the increase in social spending occurs at the regional and district level, where the responsibility for the delivery o f basic services such as primary education and health care lies. In order to ensure that the poor are able to benefit from increased social spending, government is encouraged to give priority to the poverty mapping exercise and use the results together with sectoral information to target increases in social expenditures to the poorest areas. Analysis of the benefit incidence of government programs should also provide important information for the allocation ofpublic resources. Table 12. Sectoral development expenditures (actuals, YOof GDP), FY96-FY03 Sector FY96 FY98 FYOO Actual Actual Actual Administration 0.1 0.4 0.3 Defenceand Security 0.0 0.0 0.0 Social Services 0.2 0.4 0.5 Economic Services 0.1 0.6 0.6 ProductiveServices 0.1 0.2 0.2 ITotalDevelopmentExpenditures1 0.5 I 1.6I 1.5 2.1 2.7 ' Source: AppropriationAccounts (FY96-FY02) 4.8 Recorded development expenditures show a continuous upward trend increasing from 0.5 percent o f GDP in FY96 to 2.7 percent o f GDP in FY02. This increase in development expenditures i s partly due to a greater share o f donor funded expenditures being captured in the appropriation accounts, but also to improvements in project 23 implementation performance. The budget for FY03 projects a significant jump in development spending to 7.5 percent o f GDP. Important contributors to this increase in development expenditure are higher spending on education (increase by T.Shs. 64 billion), roads (increase by T.Shs. 84 billion), water and lifestock (increase by T.Shs. 28 billion), PORALG (increase by T.Shs. 55 billion to fund implementation o f the primary education development program (PEDP) inthe districts) and energy and minerals (T.Shs. 122 billion for Songo Songo Gas field development). Are BudgetAllocations a ReliableGuideto Public ExpenditureOutturns? 4.9 In order for a budget to fulfill its role as being the key mechanism for public sector resource allocations subject to public and parliamentary scrutiny, it i s important that there i s a close correspondence between budget estimates and actual expenditures. If deviations are large on a regular basis, the budget process loses credibility and typically more informal and less transparent resource allocation mechanisms are employed to fill the void created by unrealistic budgets. There are three principal reasons for differences between budgeted and actual expenditures: 0 differences betweenprojected and actual resources available which under a cash budget can lead to expenditure releases by the Ministry o f Finance which are less than the budgetedamounts; 0 reallocations between spending units and in particular reallocations from the contingency account at the Ministry o f Finance in the course o f the budget year; and 0 under/overspending by Ministriescompared to the approved budget. Variations in the Resource Envelope 4.10 This section analyzes the impact of variations inresource availability on spending units. The analysis is based on three key aggregates, flexible budgetary resources, first claim expenditures, and non-wage discretionary expenditures. Under the cash budget system, non-wage discretionary expenditures are determined as a residual o f available resources and expenditureson first-charge items. 24 Figure 9. The macro-economic environment 7 0 - . --__ - ---- ------ I ___c -7 6 0 5 0 b o 1 5 0 15000 3 0 100 10000 2 0 5 0 500 0 10 1 0 0 1996 1997 1998 1999 2000 2001 2002 2003 1 1 0 1995 19S6 1997 1998 1999 2000 2001 2002 I 4.11 A key factor influencing government's ability to estimate budgetary resources adequately is the predictability o f the macro economic environment. Key macro- economic variables that have a bearing on budget estimates are the rates o f economic growth and inflation and the performance o f the export and import sectors. As figure 9 shows, most o f these variables have been on fairly smooth trajectories during the past eight years. Economic growth has steadily increased from below four percent in 1997 to about six percent in 2003 and the rate o f inflation has fallen from above 20 percent to below five percent. Similarly, exports have increased to above US$ 1000 million by 2003 after an initial decline to around US$ 500 million in 1998. Imports have also increased from around US$1,400 million in 1998 to about US$ 2,200 million by 2003, although on a slightly more erratic trajectory than exports. Overall, the macro-economic environment in which budgets were prepared over the past eight years was stable and thus conducive to reliable revenue and budget estimates. In addition to a benign economic environment, the cash budget removedincentives for unrealistically optimistic revenue estimates. Finally, efforts to strengthen government's capacity to undertake economic and budgetary projections and a more consultative approach to the preparation o f budget estimates in the context o f the PEWMTEF process were also intended to support realistic budgeting and resource estimates. Figure 10. Domesticrevenue and program support, leveland difference between budgetand actual (Yo of GDP) Rerourcer (% 01 GW) 1E 3h . 5 x `;i ----.I.- --- I 14 `2% 12 c* 1 C% . . . , c 1 3 C I 0 5). i! 6C% C C % P 63% - C 5 % 4 C% 2 3% .1 3% 3 3% .1 5% h 9 7 FY96 FV99 h O ( I FY31 h(I2 h 0 3 h 9 7 -98 FY98 F Y C O mol Fvc2 h C 3 DOnmltlC Revenue 0 Budget Support Comestic Revenue Budget Suppott 25 Figure 11.Difference between resource estimates and 4.12 The analysis o f resources actual resource availabilitv (YOof GDP) covers domestic revenue and donor I budget support including program grants and credits, HIPC debt relief from multilateral sources, and basket funding. Project support i s not included since variations in project support are matched exactly by variations in project support funded expenditures. Domestic revenue declined from 13.5 percent o f GDP in FY97 FY98 FYQQ FYOO -01 FY02 FY03 FY97 to 11.3 percent in FYOO. In subsequent years, revenue has increased marginally to reach 12.5 percent inFY03. The analysis o f the difference betweenbudgetedand actual amounts of domestic revenue reveals that during the period when the revenue to GDP ratio declined, revenue targets did not adjust which led to under collection o f revenue between 0.5 and one percent between FY98 and FYOI. Revenue estimates were adjusted downward in FYOI, when revenue actually started to recover slightly and as a consequence revenue performance was better than projected. In FY02 and FY03, actually revenue collection was slightly above target. The second source o f budgetary flexible resources, program aid, declined slightly between FY97 and FY99, but more than doubled. Subsequently, in most years, actual program support has been considerably below the budgeted amounts, which reflects over-optimism on both the government and donor side with respect to government's capacity to implement actions that would trigger the release o f budget support. It i s also noteworthy that the profile o f actual disbursementso f program support appears to be fairly smooth, while budget estimates of program support display much more variability. One interesting facet o fthese observations i s that ifthe budget had used the previous year's performance (or even the year prior to that) as the basis for projections, budget deviations would have been smaller than the ones actually observed. Insummary, inrecent years resource shortfalls havebeen on average0.7 percent o fGDP, but didnot exceed 1.8percent o fGDP. Figure 12. First charge expenditures, level and difference between budget and actual (YOof GDP) Firstcharge expendlures (% of GDP) 6 CA 5 ?A 40L 3c 3 9% 20% 1 0 % " "p. M97 M 9 8 FY99 M O O M01 M 0 2 FY03 Debt S e r v ~ eand Arrears Wage6 and Sslerlel 4.13 First charge expenditures include payments for debt service and wages and salaries. Debt service payments have declined significantly from 4.8 percent o f GDP in 26 FY97 to about 2 percent o f GDP in FY03, both on account o f prudent management o f domestic debt and falling interest rates, but also on account o f HIPC debt relief and Paris Club debt reductions. Initially, debt service payments were under-funded in the budget. However, in more recent years budgetary estimates have actually exceeded debt service payments, as a result o funderestimating the amount o f available debt relief. 4.14 Expenditures on wages and salaries declined from 4.7 percent in FY97 to 3.8 percent in FY99 and fluctuated thereafter around the target o f 4.2 percent o f GDP. The figures on deviations reveal that the initial decline inthe wage bill to GDP ratio was not anticipated in the budget estimates and thus resulted in significant under spending. In subsequent years, the budgetouttum on wages and salaries has been quite close to budget estimates, with only minor variations o f between0.1 and 0.2 percent o f GDP. Figure 13. Deviation betweenestimatedand actualexpenditureson first charge items ( O hof GDP) Ddference between estimates and actualexpendtures on first charge item (Ohof GDP) 10% 0 8% 0 6% 0 4% 8 ; 02% 00% s -02% -0 4% -0 6% -0 8% -1 0% FY97 FY98 FY99 WOO F f O 1 FYO2 Ff03 4.15 Overall, deviations between budgeted and actual expenditures on first charge items varied between -0.8 and plus 0.8 percent of GDP. Although in most years deviations have been small, it appears that even first charge items respond, to a certain extent, to resource availability. For example, this appears to have been the case inFY99, when resource mobilization was significantly below budget estimates and both expenditures on wages and salaries and on debt service were also below the budget estimates. Marginal adjustments o f expenditures on wages and salaries to resource availability during the fiscal year are made possible by the practice o f retaining funds for salary increases inthe budget allocation o f the Ministryo f Finance. Incase revenue and aid outtum i s below the estimates, salary increases and additional hiring have been delayed in the past. In case that towards the end o f the fiscal year there i s significant under-spending, the Ministry o f Finance has in the past used these funds for early retirement o f debt. 27 Figure 14. Non wage expenditures, level and difference between budget and available ( O h of GDP) AValIable resources and non-wage expendnure 1% of OW) Wference behYeen budget estimate and actual 1% of GDP] 2 cm '5% 1 3 4 = c 54 ? 0 3 6 4 5 % # -1C% .1 5 4 .2 0% .2 5% .? 3 4 -97 FY98 FY99 FVCO FY31 N O 2 h 0 3 Cy97 h 5 9 FV99 F Y O I Q?' h 3 2 hX Resources for "on-w age expendnurer 0 Non-wage expendnure$ ~ R B I O U I C ~for "on-w age expenddurer ~ N o n - w a g eexpenddures S 4.16 The difference between flexible resources and first charge expenditures is the amount o f resources available to government for spending on other charges and domestically funded development expenditure (non-wage discretionary spending). Since FY98, this amount has increased steadily from 5.5 percent to reach 10.8 percent inFY03, mainly a consequence o f increased program support and decreased debt service payments. Alongside with the increase in available resources, there has been a drastic change in the predictability o f resource availability. While until FYO1, available resources for non-wage discretionary spendingtypically were at least one percent o f GDP below the budget estimates, available resources have exceeded budgeted amounts in FYOl and FY03 and were only slightly below the budget estimates in FY02. This indicates that improvements in budget management practices have indeed led to improved budget estimates and implies that the budget management system i s now robust enough to ensure aggregate fiscal discipline, even without the application o f a cash budget system. 4.17 The working o f the cash budget system is evident inthe comparison o f resources available for non-wage discretionary expenditure and actual non-wage discretionary expenditures. Until FYOO, expenditures had always been below residual resource availability. Since FYOl, coinciding with the adoption o f the PRS, expenditures actually exceeded resource availability by a small margin. However, as the analysis o f the difference between budgeted and actual expenditures shows, compared to budgetary targets, budget implementation has consistently been even more conservative than budgetingleadingto smaller budgetdeficits than targeted. 4.18 As for the recurrent budget, similar improvements are also observable for the development budget. As figure 15 shows, the difference between budgeted development expenditures and the outturn recorded in the appropriation accounts has declined significantly. This i s primarily the result o f joint government donor efforts to channel more development assistance through the exchequer system and to enhance the flow o f information on project spending between donors and government. Another factor contributing to the improved capture of development expenditure i s the fact that the share o f credit financing o f the development budget has increased in recent years. (Figure 16). Credit financed development expenditures are fully captured in the budget and the appropriation accounts, as such development activities are implemented by government usinggovernment systems. 28 Figure 15. Development expenditurecapturedin the appropriation accounts as a share of development expenditurerecorded in the budget. FY97-FY02 80% 70% 60% 50% 40% 30% 20% 10% 0% FY97 FY98 FY99 FY00 FYOI FY02 Figure 16. Share of projectcredits in total foreign financing of development expenditures 60% 50% 40% 30% 20% 10% 0% FY97 FY98 FY99 FYOO FYOI FY02 4.19 To summarize the foregoing discussion, the difference between budget estimates and expenditure outturns for all main aggregates has become quite small in recent years. This has created a situation where budget estimates appear to be appropriately accurate and conservative so as to make disruptive expenditure cuts duringbudget implementation unnecessary. It might thus be appropriate tofurther relax the cash budget system while keeping a close watch on the realism of budget estimates. 29 Reallocations 4.20 First we analyze the reallocations that took place during FY02. As table 13 shows, 28 percent o f all votes received reallocations that were more than 15 percent of their original allocation, 20 percent of votes received reallocations between 5 and 15 percent. Comparing these `shares to those of previous years, no apparent trend i s recognizable. Reallocations are authorized by the Ministry of Finance and submittedto Parliament semi-annually ex-post in the form o f a Statement o f Reallocations - Realloation Warrants. Table 13. Share of votes receiving reallocations Size o f Reallocation (?? o f original budget allocation) FYOO FYOl FY02 (inpercent) (inpercent) (inpercent) >+15 32 20 28 +5 - +15 27 18 20 -5 -+5 39 51 50 -15 -5 - 0 4 0 <-15 2 7 2 Source: Author's calculations based on budget and appropriation account data 4.21 Almost all reallocations that took place during FY02 were funded from the "special expenditure" allocation retained at the Ministry o f Finance. The budget for FY02 contained T.Shs. 232.4 billion of unallocated expenditure which were retained in the budget o f the Ministry o f Finance, representing about one third o f non-wage recurrent expenditures inFY02. Out o f this unallocated expenditure, T.Shs. 202.4 was reallocated to specific expenditure items either within the Ministry of Finance (T.Shs. 9.9 billion) or under other votes (T.Shs. 192.5). The FY02 PER report expressed concern about the transparency o f the allocative process with such a large unallocated share of budgetary special expenditure at the end of thefiscal year which is submitted to Parliament.)' resources and recommendedthat the Ministry o f Finance prepare "a report on the use of Reporting to parliament took place in the form of the semi-annual reallocation warrants. However, that reporting format makes it difficult to reconcile actual uses with the earmarked purpose. Table 14 compares intendedand actual uses of`special expenditure" inFY02. While it is apparent that the largest share of the "special expenditure was used in line with the earmarked purpose, a not insignificant amount of the "special expenditure" was also reallocated to uses that are clearly not related to any o f the earmarked purposes. 30 Table 14. Earmarked and actual use of retained "specialexpenditures" IntendedUse T.Shs. T.Shs. (billion) Actual Use (billion) IVAT refinds I 56.0 I contingent liabilities parastatals 44.0 payment arrears 41.9 salary adjustments 42.7 primary school fees 11.0 Subscriptionto Other International Organizations I I I 1.o TanzaniaAirport Authority 1.6 I I I CooperativesAudit and SupervisionFund 1.7 ForeignAffairs 9.6 RoadFund Share 3.2 Other 23.O 4.22 For FY03, the Ministry o f Finance has made efforts to reduce the need for the retention o f the special expenditure allocation and only retained T.Shs. 77.9 billion. The items to be covered from the special allocations relate mainly to salary increases in government and the parastatal sector, tax expenditures for exemptions for non-religious NGOs, debt swaps, HIPC debt relief adjustment, the anti-corruption campaign, and contingencies proper. Table 15. Earmarked uses of special expenditure for FY03 (in T.Shs.) SPECIAL EXPENDITURE I2002103 Budge1 Parastatalwage increase 6,000,000,000 Contigency proper 8,000,000,000 Debt Swaps 4,000,000,000 Voucher systemfor non religious NGO's 13,150,000,000 Salary Adjustment 41,311,652,800 Anti-Corruption Campaign 800,000,000 HIPC Relief Adjustment 1,721,000,000 Total 74,982,652,800 Source: MoF 31 4.23 The key remaining budget management issue arising in the context of special expenditures i s the budgeting for wages and salaries, which i s not yet fully incorporated into the budget process. Although the Civil Service Department has developed a functioning Integrated Human Resource and Payroll System, the Ministry o f Finance maintains its own database for budgeting and payment o f salaries. At present, the Ministry o f Finance determines the overall wage bill, typically as a target wage bill to GDP ratio, while the CSD makes decisions on employment and salary increases within the financial frame provided by the Ministry of Finance. While this division of responsibilities i s appropriate, there i s clearly scope for the Ministry o f Finance and CSD to work closer together in the preparation o f the budget and eventually include salary adjustments in the allocations to the spending units rather than retaining them under special expenditure. Inparticular since salary increases are typically announced on May 1, allocation o f salary increases to spending units in the budget should be feasible. Another challenge i s the integration o f budgetingfor wage and non-wage expenditures at the sectoral level, decisions that are currently takenthrough entirely separateprocesses. Deviation between budgeted and actual spending by MDAs 4.24 With respect to deviations between actual spending and budgeted (including reallocations) amounts, Table 16 indicates that seven percent o f all votes overspent by between 5-15 percent (Local Government Service Commission, Ethics Secretariat, and Law Reform Commission), twenty percent o f votes underspent by 5- 15 percent o f their allocation, and 20 percent o f votes underspent by more than 15 percent o f their allocation. Table 16. Share of votes with overlunderspending, FYOO- FY02 FYOO FYOl FY02 >+I5 2 2 0 +5 - +15 0 0 7 -5 52 89 65 -15 -5 --+5 18 7 20 <-15 27 2 9 Source: Author's calculations based on budget and appropriation account data 4.25 Overall, the patterns o f deviations between budgeted and actual spending reflect the quality o f expenditure and commitment control systems, which facilitate budget execution in accordance with the approved budget and prevents overspending. On the other hand, the cash budget still leads to under-spendingby about 30 percent o f all votes. In that context it is interesting to note that in FYO1, when overall available resources exceeded budgeted amounts, deviations between budgeted and actual expenditures were quite small. This attests to the ability o f the government to implement programs as budgeted and that under-spending i s indeed primarily the consequence o f lack of funds rather thano f implementation capacity constraints. 4.26 Since FYO1, vote holders are required to seek Ministry o f Finance approval for intra-vote reallocation. Indiscussions with various ministries this measure was generally considered to be useful. First o f all, it helpedto increase attention to budgeting. While 32 poor ministerial budgeting practices in FYO1 led to numerous reallocation requests, apparently improved budgeting during FY02 has reduced the number o f required reallocations. Vote holders also indicated that the requirement o f MoF approval also protects vote holders from political interference during budget execution. Nonetheless, in the medium term the objective should be to strengthen managerial accountability of spending units and give them sufficient scope to manage their budgets to reach agreed results. 33 34 5. BUDGET EXECUTIONASSESSMENT BASED ONTHE REPORT OF THE CONTROLLERAND AUDITOR GENERAL FOR FYOl 5.1 The National Audit Office (NAO), formerly the Office o f the Controller and Auditor General (OCAG), i s mandated to audit the accounts o f both central and local governments. The N A O produces two separate audit reports for (a) ministries, departmentsand regional directorates, and (b) for local authorities. As noted inprevious PER reports, the current audits carried out by the N A O are essentially compliance audits. Under the new Public Finance Act No. 6 o f 2001, the Controller and Auditor General (CAG) i s also empowered to carry out value for money audits. The C A G audit reports have proved to be useful since they provide important information on the efficiency o f budget execution, the integrity o f the government accounting system and the extent of resource leakages. 5.2 This section analyses government compliance to prudential budget management procedures and regulations, based on the report o f C A G for FYOl. The analysis updates the previous analyses done for PERs for FY99 through FYO14. The aim o fthis section is to establish the prevalence of qualified or adverse audit opinions issued by CAG; identify the main causes o f the unfavorable audit opinions; assess the extent o f responsiveness to audit queries and follow-up actions by vote holders; and analyze and explain the patterns for excess spendingor under-funding of votes.5 5.3 Delays in Publishing the CAG Reports. There were continued delays in the publication o f C A G reports for both central and local governments, in violation o f the requirementso fthe Public Finance Act No. 6 o f2001. The central government report was publishedwith a 6 months delay andthat of local authorities with a 2 month delay (Table 17). The reasons for this include capacity constraints, electricity shortages experienced at the NAO, and time consuming auditing o f donor funded projects and special accounts. The other reason is late submission o f the accounts to CAG by the accounting officers - the practice which, however, has shown an improvement over time with less accounts being submitted late by the accounting officers. Sincepresentation of the CAG reports to The current analysis, therefore, avoids background information, to minimize unnecessary repetitive details o f issues that were covered inthe previous reports. 5 The Country Financial Accountability Assessment (CFAA) carried out in 2001 provides a detailed analysis o f the strengths o f financial accountability processes in Tanzania, covering both the public and the private sectors, including an assessment o fthe government accounting and audit system. 35 Parliament in a timely manner is essential to increase the impact of these reports, it would be important that the government take measures that would allow the CAG to produce audit reports without delay. Such measures could include contracting an increased amount of work to outside auditors. Other options include building capacity of NAO to undertake computer based audits of the IFMS based government accounts. Establishing such capacity should be of very high priority for government to ensure the integrity of the IFMS system. Moving to a computer based auditing is also likely to reduce the time neededfor preparing these annual reports. Table 17. Submission of accountsand productionof CAG report, FY97-FYOl Source: CAG Reports (various) 5.4 Improvements in Financial Accountability. The reviews o f the CAG reports for central government and local authorities for FYO1 suggest significant improvements in financial management at all levels as fewer accounts received an adverse opinion. Unlike FYOO, in FYOl the majority of audit certificates of Central Government (ministries & regions) accounts and statements issued by C A G were predominantly qualified (37 percent) and clean (30 percent). Only 33 percent o f all accounts received adverse certificates, which i s lower than the 58 percent received inFYOO. Among ministries, the share o f votes which received a clean certificate increased from 31percent inFYOO to 34 percent in FYO1; those which received a qualified certificate also increased from 26 percent to 53 percent; while those that received an adverse opinion decreased substantially from 43 percent to 14 percent. For regions, there was also a substantial decrease in the percentage o f accounts which received an adverse opinion from 82 percent in FYOO to 67 percent in FYOI. Regional accounts which received clean certificates increased from 13 percent to 23 percent, while accounts which received qualified opinions rose from 5 percent inFYOOto 10percent inFYO1. 5.5 During the year 2001, financial accounts of 114 local authorities were audited. The state o f the LA accounts improved slightly in 2001 compared to 2000. The share o f councils that received clean certificates decreased from 14 percent in 2000 to 11 percent in2001, while the share of councils which received qualified certificates increased from 36 22 percent to 54 percent and those that received adverse opinions decreased from 64 percent to 35 percent. 5.6 During FYOl, response rates to CAG's audit queries declined for both ministries and regions. Only a very small proportion of the total queries issued received replies by vote holders. Overall only about 7 percent o f queries issued to ministries and regions were repliedto inFYO1, indicating a decrease over the period FY96 FY99when - overall an average o f about 11percent o f issued queries were repliedto, and FYOO when about 26 percent of the queries were replied to. This i s not a healthy situation and an indication that less action seems to be undertaken by accounting officers to rectify the recurring weaknesses in expenditure control and other compliance issues. As noted in previous PER reports, the implication o f this situation i s that in many cases the proper utilization of public funds cannot be ascertained, and unless the offenders are sanctioned, little or no improvement will be recorded. The major cause o f non-responsiveness to the CAG's queries was observed to be basically the absence o f the will to enforce accountability and a lack o f appropriate legal instruments for enforcement. At present, there i s no mechanism inplace to sanction non-responsiveness to audit queries. As such, measures ought to be taken to create the right incentives for accounting oflcers to pay more attention to these issues. expenditures that are not vouched or improperly vouched continued Box 2. Audit certificates during the FYoly but at a lower The Controller and Auditor General issues for each audited ProPodion compared to the Past few vote an audit certificate which indicates whether the supply and development vote, revenue statement, miscellaneous years. Improperly vouched deposit accounts, and statement o f assets and liabilities (a) expenditures as a share of total present fairly the fmancial position o f the respective vote expendituredecreasedfrom about 6 (clean audit certificate), (b) present fairly the financial percentin FYooto about percent position o f the respective vote subject to comments made in FYOl and the dmre of by the Controller and Auditor General (qualified audit certificate), or (c) do not present fairly the fmancial position unvouched expenditures to total o f the respective vote (adverse audit certificate). 5.8 Like inFYOO, there was a notable rise inincidences o f embezzlement o f cash and stores in FYO1. The magnitude o f cash and stores losses increased from 0.16 percent o f total expenditure in FYOO to 1.62 percent in FYO1. Embezzlement of cash and stores remain highmainly on account o f poor procurement and stores management. 5.9 The CAG FYOl report shows a decline in excess spending and savings with respect to recurrent expenditures. The level o f excess spending on the development account also decreased, while savings increased. Excess spending occurs in specific votes either due to additional spending on a line item inexcess o f the budget allocation or because of unauthorized reallocations across specific line items. The relative large amount o f excess spending on the recurrent account under the cash budget system, where 37 overall spending levels are tightly controlled, implies relatively large deviation o f actual expenditure patterns from the budget. With the introduction o f the requirementthat intra- vote reallocations need to be pre-approved by the Ministry o f Finance, excess spending should decline. The increase o f excess spending on the development account i s likely to represent donor funding o f projects where actual funding exceeded the funding levels indicated inthe budget. 5.10 Savings occur if expenditure on a certain budget item i s below the budgeted amount. On the recurrent account, savings decreasedfrom T.Shs. 66.3 billion inFYOO to T.Shs. 30.2 billion in FYOl. This reflects the fact that under the cash budget system expenditurereleases inFYO1were significantly below budget. Poor integration of donor project finance into the government accounting has made savings on the development budgetto be a perennial problem. DuringFYO1, for example, savings increased to 130.5 billion from T.Shs. 116.2billion recorded inFYOO. Savings on the developmentaccount occur mainly due to unrealistic estimates; unrecorded expenditures; non-recording or non-charging o f aid expenditurefor some projects to the accounts hence making it appear as savings against the appropriate estimated provision; and failure to fully account for aid received in the form o f goods, materials and equipment. Poor certification and losses are mainly the result of weak controls, extra budgetary expenditures and irregularities. In particular, the immediate and underlying causes o f the most frequent audit queries include incidences o f excess votes; inadequate supporting documentation o f expenditures; failure to follow approved procurement procedures; and failure of purchasing units to account for purchases, and cash and store losses. Dishonesty and weak technical manpower capacity, especially in the ranks of accountants and auditors, will for sometime continue to contribute to audit queries, although improvements in financial management are expected to reduce some o f the above-mentioned causes o f audit queries. 38 PART 11: ALIGNING THE BUDGETWITH THE PRS 39 40 6. ALIGNMENT OF THE BUDGETWITH CORE PRSP PRINCIPLES 6.1 There are five core principles underlyingthe development and implementation o f poverty reduction strategies. The strategies should be: 0 country-driven - involving broad-based participation by civil society and the private sector inall operational steps; 0 results-oriented -focusing on outcomes that would benefitthe poor; 0 comprehensive inrecognizing the multidimensional nature o f poverty; 0 partnership-oriented - involving coordinated participation o f development partners (bilateral, multilateral, andnon-governmental); and 0 based on a long-term perspective for poverty reduction. 6.2 With the budget being one o f the key instruments for implementing the PRS, these principles are also directly relevant for the budget process. If public expenditure management practices deviate significantly from these principles, a disconnect between the PRS and the budget process becomes more likely. Ofcourse, most o fthese principles are also relevant and important for sound budget practices by themselves. This section thus examines to what extent these principles are reflected in public expenditure management inTanzania. CountryDriven Open and ParticipatoryNatureof the BudgetingProcess - 6.3 Having a country driven PRSP and local ownership enhances the prospects that the PRS is translated into concrete programs funded by the budget. However, there are three specific aspects that need to be considered in assessing the extent to which the budget i s country driven. First i s the degree o f broad based participation by local stakeholders, including parliament, civil society, and the private sector in the budget process. The second concems the role o f IFIs and donors inthe budget process. The last aspect concems the role o f donor support inthe budget. 6.4 The adoption o f the government-led, participatory PER process in 1997 had as one o f its key objectives the opening up o f the budget process to wider participation. Several evaluations that have been carried out recently attest that indeed much progress has beenmade inthis area. Over the past five years, the dialogue between line ministries, donors, civil society and the Ministry o f Finance during the budget formulation process has been significantly enhanced. The PER process provides a forum for exchange of views among key stakeholders at the sectoral as well as the macro levels. 41 6.5 The participation o f civil society and the donor community inthe PER process is an important channel through which the participatory nature o f the PRS process i s carried through into the policy implementation stage. In this regard, the openness to and active encouragement o f the engagement o f civil society in the PER process by the Tanzanian government i s to be praised. Civil society organizations (CSOs) in Tanzania are strong and have become increasingly actively engaged in the PER process over the past few years. Constituted by a combination o f service delivery, advocacy and mixed NGOs, Tanzanian CSOs act as a conduit for grass-roots concerns to reach the policy level. 6.6 CSO participation provides a much welcome complement to the perspectives provided by government and the donor community and plays a crucial role in enhancing the accountability o f overall policy-making as well as o f the PRS and the PER processes inTanzania. CSOs provide valuable contributions to the sectoral working groups intheir areas o f expertise and, progressively, to overall macroeconomic and structural issues (See NGO Statements. Consultative Group Meeting. December 2-5,2002). CSO effectiveness will further be enhanced by the recent creation o f a "broker" organization-the NGO policy forum, which will be led by the various member NGOs inturn. 6.7 To further strengthen the participatory nature o f the budget process i)more active involvement o f CSOs inthe external review phase o f the PER process -perhaps through the NGO policy forum-- should be encouraged as part o f a long-term strategy to progressively devolve the exercise o f external accountability o f government to domestic actors; ii)the Tanzanian government and civil society may wish to consider extendinghe- structuring the PRS monitoring master plan to include a participatory monitoring and evaluation system inwhich civil society could play a leading role. 6.8 The involvement o f the private sector in the PER process has been limited, but government has instituted pre-budget consultations with the private sector that allow the private sector to bring its specific concerns to the table. 6.9 Finally, the role o f Parliament in the budget process is well defined relating to approval o f the budget and oversight related to follow-up to the Controller and Auditor General's report and hearings o f the public accounts committee. One o f the key constraints to effective parliamentary oversight has been the fact that reports by the Auditor and Controller General are typically submitted to Parliament and discussed with significant delays, which precludes Parliament to take C A G findings into account in its discussion o f the budget. The Public Accounts Committee o f Parliament conducts hearings with the Accounting Officers (usually Permanent Secretaries) on the findings of the Auditor and Controller General's Report. However, there is no reporting on the follow-up to C A G findings andParliament has now means to ensure follow-up. 6.10 To conclude, while through the PER process government has displayed a great willingness to open up the budget process to greater stakeholder involvement, there is still considerable scope for enhancing the process. In addition to creating avld sharpening opportunities for participation, equal attention needs to be paid to stakeholders capacity to participate. This will involve the streamlining o f processes and clear definition o f entry points for and the role o f participation, but also greater attention 42 to strengthening the organization and capacity o f stakeholder groups to participate effectively inthe budgetprocess. ResultsOrientation of the Budget Process 6.1 1 The PRS defines key objectives o f public policy and lays out the strategies for achieving these objectives6 This has given rise to a shift from defining priorities along administrative or sectoral lines to defining priorities in terms o f desired outcomes. One o f the key challenges for the budget process i s to translate these objectives and desired outcomes into (preferably quantifiable) outputs which represent the contribution o f government to the achievement o f the PRS objectives and to estimate the resource requirementfor the "production" o fthese outputs. At the same time, mapping budgetary resources to desired outputs and outcomes i s also a key requirement to properly assess whether the use o f public resources contributes indeed to the achievement o f the objectives. 6.12 The PRS provides the framework for budgetary decision making, but in many areas results are not sufficiently well defined to feed directly into the budget process. A pre-condition for establishing a stronger link betweenPRS objectives and the budget process i s that the PRS targets are well defined and have uantitative indicators and benchmarks against which progress can be measured. The 2" PRS progress report 1 contained a list o f indicators. However, work on establishing baseline values and targets for all o f these indicators i s still ongoing. For those indicators where targets have been established, the period over which the target should be achieved differs by indicator. Relatively little effort has been made to facilitate monitoring o f these indicators by setting targets that would coincide with the budget cycle or the MTEFperiod. While for some indicators monitoring on such a short term basis i s not possible due to the (in)frequency o f data collection exercises, all effort should be made to set annual targets for the M E F period which would allow better alignment of the MTEF strategy with the PRS as well as better monitoring of the impact of expenditure allocations on poverty reduction indicators. 6.13 A large number of processes to monitor results are in place, with a variety of objectives. These include: The Poverty Monitoring System, which provides the overall framework for assessingprogress inmeeting the objectives o f the PRS and which comprises both data collection and analysis and also assesses the impact o f policies and programs onpoverty reduction; the PER process, which has as one o f its tasks the costing o f achieving specific PRS objectives as well as the broader tasks o f supportingbetter budgetingthrough the analysis o foutput basedresource requirements; the MTEF, which inaddition to providing a mediumterm framework for budget allocation also includes the quarterly monitoring o f budget execution and progress indeliveringplannedoutputs; At present, efforts are underway to integrateMDGtargets into the PRS. 43 0 monitoring by the Prime Minister's Office, which monitors the implementation o f the CCM's election manifesto and which reports annually on implementation progress to parliament; 0 "performance budgeting," which also aims to link budgetary inputs to specific outputs, however, while the PER process i s led by the Ministry o f Finance, performance budgetingi s championed by the Civil Service Department; 0 sector development programs for health and primary education which provide a well definedframework for linking resources to desired outputs andoutcomes and have a regular reviewmechanismfor the achievement o f outcomes; and 0 public sector performance management system, which focuses on performance assessment of the public sector and linking organizational goals to individual performance. 6.14 Most o f these processes were launched before the PRS and poverty reduction objectives are thus rarely explicitly articulated. However, even before the PRS was put in place, government's development efforts were focused on sectors which are likely to have the biggest impact on development and poverty reduction. These sectors are the same that were identified as priority under the PRS and most o f the above processeshave been initially piloted and implemented for the priority sectors and are only gradually being rolled out to other sectors. Efforts are still ongoing to fully align sectoral objectives with those o f the PRS. 6.15 There is a fairly large number of monitoring processes in place, however their linkage to budgeting is tenuous. Existing monitoring systems were set up with different objectives and information is often not adequately channeled to budgetary decision makers. A recent review o f results oriented public expenditure management in Tanzania7concludes that with all these processes there is a proliferation of indicators and targets in all PRS relevant sectors, they lack consistency and a well structured hierarchy in relation to the intended use of the indicator and the institutional responsibilities for deciding targets and monitoring achievements. An important aspect of enhancing the results orientation of the budget is thus to harmonize bothprocesses and indicators. 6.16 The budget process itself shows little responsiveness to results. Both the MTEF and especially the performance budgeting approach are aimed at using feed back on performance in decisions on budget allocations. However, there are at least two sources o f tension between the objective o f creating a stronger link between performance and allocations. The first i s the inherent political nature o f the budget process. The second i s more subtle and derives from the identification o f priority sectors inthe PRS. There is pressure from both domestic and international constituencies to increase spendinginthe priority sectors. Given the complexities inassessingperformance andthe fact that budgetary allocations are the easiest to monitor indicator o f government's commitment to poverty reduction, there i s little scope for punishingpoor performance in the priority sectors with lower budget allocations. Another important constraint of the results orientation o f the budget process i s the often complex relationship between 'Frans Ronsholt et al. Results-OrientedExpenditure Management Country Study Tanzania. Preparedfor OverseasDevelopmentInsititute.London, UK.FinalDraftReport. 18" December2002. 44 budgetary inputs and outputs and outcomes and only very limited knowledge on the production function for achieving certain targets. However, greater use o f well defined and objective performance indicators (incidence effects, measures o f efficiency, etc. ) could inform and constrain pressures to spend on ex-ante notions o f priority and likely impact and assist to improve the design of programs. 6.17 Monitoring efficiency is underdeveloped. An important element in achieving greater results orientation in the use o f public resources would be greater attention to efficiency. So far, the main focus in budget management has been on budgetary allocations and, to a lesser extent, on outcomes. Efficiency inthe use o f resources, on the other hand, has received rather less attention. In moving forward, it will thus be important to identifi eflciency indicators and monitor efJiciencyover time. 6.18 Monitoring impact of public expenditures on the poor needs to be strengthened. Another aspect o f results orientation concerns the direct impact of public expenditures on the poor. As discussed inthe section on pro-poor expenditures, Tanzania uses a system o f broad targeting with spendingallocated to priority sectors that have been identified though the PRS process as being o f direct relevance to poverty reduction. In order to evaluate whether such spending has indeed the desired impact on poverty reduction objectives, it will be important to monitor more closely who the beneficiaries of such spending are. Inaddition to the benefit incidence analysis o f the main social sector programs which i s being carried out based on the HBS data, it will also be important to carry out specific program evaluations inthe priority sectors to assess whether the poor benefit from public expenditures. Similarly, expenditure tracking and service delivery surveys could also be designed in a manner that would provide information as to the poverty impact o f the studiedprograms. Comprehensiveness-Recognizingthe Multidimensional Nature of Poverty 6.19 Inaddition to the general requiremento f comprehensiveness of the budget as one o f the principles of good budget management, comprehensiveness inrelationto the PRSP also relates to the extent to which the budget fully represents the objectives set out inthe PRS and the extent to which information and analysis on poverty i s used in the budget process. 6.20 As discussed in more detail inthe analysis of poverty reducing expenditures, the budget process fully reflects the focus areas defined in the PRS. However, the budget typically defines additional expenditure items as priority, some o f which are the consequence o f operational imperatives such as the financing o f elections. Insome cases, the budgetprocess also reflects the evolving discussion onpoverty reduction inTanzania, even before it i s formally captured inPRS documents. For example, as the discussion o f poverty reduction has recently given more weight to economic growth, the budget also has given greater attention to growth enhancing expenditures, while maintaining the areas o fpriority expenditure definedinthe PRS. 6.21 Another aspect o f comprehensiveness i s the extent to which poverty reduction objectives are fully integrated into the budget process or confined to a "special project" 45 status in the budget. The least comprehensive approach would be the set up of an institutional poverty reduction fund, with the rest o f the budget process unaffected by the PRS. On the other end o f the extreme, all aspects o f fiscal policy and expenditure management are aligned with the PRS. Tanzania clearly has adopted the latter model, where poverty reduction objectives are fully streamlined into sectoral budgets and the designo f fiscal policy. 6.22 Another important dimension i s whether the budgetprocess i s guided by available data and analysis on poverty. Government has established a comprehensive poverty monitoring framework and established a poverty monitoring group with subgroups responsible for (i)surveys and censuses, (ii)administrative data, (iii)research and analysis, and (iv) dissemination. The poverty monitoring system has already greatly expandedthe information o fpoverty inTanzania, inparticular information available from the household budget and other surveys. To date, this information has not yet beenused extensively to guide budget allocations and to ensure that sectoral expenditure programs indeed benefit the poor. Government is encouraged, to fully utilize that information in the budget process. Benefit incidence analysis and poverty maps are examples of how available data could be used to inform budgetary decisions. The PER process could support a review o f expenditure programs in light o f recent information and analysis on poverty. Partnership-Oriented 6.23 Tanzania receives a large amount of development assistance from its developmentpartners, which is an expressionof the existing partnership. About 90 percent o f the development budget is financed by donors. Through sectoral and general budgetsupport, donors are also increasingly contributing to the financing o f the recurrent budget. 6.24 The PRS is the framework for this partnership. Donors are increasingly aligning their support with the PRS and individual donor strategies typically take the PRS as the strategic framework within which donor assistance i s situated. 6.25 The Tanzania Assistance Strategy (TAS) defines principles for the partnership between Tanzania and her development partners. The importance o f a well fbnctioning partnership between government and development partners had been established before the PRS was developed, though the PRS played an important role in further focusing and developing this partnership. The TAS spells out principles for the collaboration o f government and donors intended to enhance ownership, efficiency, and effectiveness o f foreign aid. In particular, it encourages the integration o f aid into the government's budget process and the use o f government systems, rather than the establishment o f parallel donor financed structures. 6.26 Mechanism for monitoring the partnership. To keep the evolution of the government donor relationship and the adherence to agreed principles under review, an Independent Monitoring Group (IMG) has been tasked to conduct annual reviews. The IMG consists of local and international experts and represents the institutionalization of 46 previous ad hoc reviews o f the partnership. The Development Assistance Committee, a donor coordination mechanism, also serves as a peer pressure mechanism for working withinthe agreed principles ofpartnership. 6.27 With respect to the budget processes, this partnership between government and development partners has been systematically developed through the PER/MTEF process. This process has become a platform for dialogue on budgetaryand aid strategy. At the same time, the PER process also had as its objective to foster greater openness o f the budget process to local stakeholders. This has ledto greater openness in the relationship between government and the donor community to other local stakeholders, including NGOs, parliament, and the private sector. Main consultative fora such as the annual PER consultative meeting and the Consultative Group meeting are beingheldinTanzania and open to civil society. 6.28 Enhanced partnership through budget support and sector programs. The partnership between government and donors has seen significant changes in recent years. Inaddition to overall increased donor support to Tanzania on the basis of successful and sustained efforts to improve the institutional and policy environment, there has also been a move towards aid modalities that support domestic ownership. This includes the move from primarily donor managed and implemented projects to support of sectoral programs as well as general budget support for the implementation o f the PRS. Sector development programs are in place for primary education, health and agriculture. In addition to sector programs, there are also several thematic basket funding arrangements in place such as those to support local government reform program or the tax administration program. 6.29 Effortsto integrateforeign aid intothe budget. As the budget is government's primary resource allocation mechanism, enhanced partnership also requires integration o f donor support into the budget. As documented elsewhere in this report, in recent years the PER process has been the platform for discussion and alignment o f budgetary and donor strategies. In addition, the PER process has also facilitated better exchange o f information on donor commitments and expected and actual disbursements and utilization o f the government exchequer system for aid disbursements by a greater numbero fdonors. 6.30 Greater scope for partnership with the private sector. The PRS recognizes private sector development as one o f the key focus areas to foster sustainable economic growth. Strategies for some o f the priority sectors, such as rural roads, also draw a clear distinction between the government as financier o f services and the private sector as implementer. For other sectors where private provision also plays an important role, current budgeting barely acknowledges the role o f the private sector. For example inthe health sector, private dispensaries are important providers o f services. Nonetheless, budgetary planning for the health sector focuses on public facilities in meeting PRS and MDGtargets. 6.31 Partnershipswith NGOs and Civil Society Organizations. NGOs, including religious organizations, make considerable contributions to pro-poor service provision in 47 some o f the key priority sectors for poverty reduction such as education, health, and water. However, information on private and non-governmental financing and service provision i s generally scant and not available in consolidated form at the national level. This precludes central government from taking such service provision systematically and transparently into account in its decision making process, although apparently some such information does feed into decision on allocations to the districts. In some sectors such as health and water, efforts are underway to obtain a more comprehensive picture o f finance and service delivery through private and non-governmental organizations to arrive at a better allocation o f resources. In order to focus scarce public resources on improving service delivery to thepoor, more systematic mapping of the role of theprivate and the public sector could promote greater efficiency and prevent duplication in the sector. Long-termPerspectivefor PovertyReduction 6.32 The PRS adopts an appropriate long term perspective, with targets and objectivesdefinedover varying medium and longterm horizons. The MDG targets, which are being integrated into the PRS, also cover a time horizon o f ten to fifteen years. However, to create a stronger link between the PRS and the MDG targets, it will be important to develop short and medium term targets which would correspond to the budget and MTEF horizon. As the review o f sector PERs indicates, this breakdown o f targets i s still incomplete and there i s thus scope to enhance the link between the long term perspective o f the PRS and fiscal policy. 6.33 Tanzania's budget process uses a Medium Term Expenditure Framework with a three year horizon, which is considered to be adequate for budgetary planning. A key objective o fthe mediumterm expenditure framework inTanzania is to strengthen the estimates o f available resources in the medium term and enhance predictability. The main uncertainty arose in connection with donor assistance, where government typically only had firm commitments for one year. To the extent that donor assistance is more fully integrated into the budget, there is now increasing need for better information on medium term plans and for longer term commitments to ensure that fluctuations in donor assistance do not derail budgetary strategy. While the M E F horizon is adequate for budgetary planning, it will need to be supplemented by occasionalfiscal analysis of speciJic issues over a longer time horizon. Inthe following paragraphs we identify several issues which requirethe adoption o f a longer time horizon inbudgetary planning. 6.34 Assessment of the feasibility of achieving MDG and PRS targets requires a long-term perspective. At present, costing exercises and sector PERs frequently indicate significant financing gaps over the MTEF horizon. In order to assess the prospects for meeting MDG and PRS financing requirements inthe mediumto long term, it would be desirable to develop and review scenarios on the likely development of revenue from all sources (domestic, foreign, and borrowing) to assess whether the chosen targets are realistic or whether these targets need to be revised or measures undertaken to bringresources inline with requirements. 48 6.35 Inthis context, the assessment of aid dependency needs to beconsidered over a longer time horizon. Iti s clear that inthe short to medium term there is little scope for reducing levels o f aid and aid dependence without a direct impact on progress towards poverty reduction targets. Appropriately, the current discussion and in-country efforts focus on how to minimize negative effects o f aid dependency by stressing the need for government ownership and integration of aid into government systems. However, the assessment o f prospects for being able to fund the government program and PRS to a greater extent from domestic sources should be done with a long term perspective to assess the levels o f economic growth and domestic revenue effort necessary to reduce the level o f aid dependency inthe medium to long term. 6.36 The initial debt sustainability analysis carried out under HIPC adopted a 20 year time frame, which is necessary given the long repayment periods of lending. Regular updates o f the debt sustainability analysis will also require a review o f the fiscal projections over the same time horizon. 6.37 Finally, the assessment of recurrent cost implications of development programs also requires the adoption of a long term perspective. This is particularly important for projects where recurrent cost are likely to increase over time or which might involve high rehabilitation and maintenance requirements in the medium to long term. 49 7. POVERTYREDUCINGEXPENDITURES What are Poverty ReducingExpenditures in Tanzania? 7.1 The Tanzania PRS identifies specific poverty reduction targets covering both Box 3. What are pro-poorexpenditures? income and non-income Pro-poor growth and pro-poor expenditures are central dimensions o f poverty. It also elements in the discussion of poverty reduction strategies. spells out strategies o f how Analysts of pro-poor growth debate whether growth is pro- these objectives can be poor simply if it helps to enhance the income and social achieved and public indicatorsfor the poor or whether they also needto leadto a expenditures are an important change in income distribution towards the poor. With respect to public expenditures, similar questions can be instrument to achieve these raised. Are expenditures only pro-poor if they benefit targets. The PRS focuses on disproportionally the poor or is it sufficient that public three main areas o f outcomes expenditurescontributeto increasingthe welfare ofthe poor and actions for realizing them. to label them pro-poor. Indeed, assessing whether The first i s reduction in the expenditures are pro-poor is even more complex since the breadth and depth o f income time horizonover which public expenditure programs have poverty. The outcome targets an impact on poverty differs from program to program. include reduction o f basic Currently, poverty reducing expenditures are usually needs poverty and food assessed with respect to their immediate impact on poverty, with a particular focus enhancing access to public services for the poor. However, on rural areas, where poverty i s other public expenditures and public investmentmighthave most prevalent. The second a more significant and long lasting impact on poverty by enhancingeconomic growthinthe mediumto longterm. area i s that o f improving the quality o f life and social well- being. This entails improved human capabilities, enhanced longevity and survival, social inclusion and personal security, improved nutrition, and containment o f extreme vulnerability (mainly through social safety nets). The third broad area concerns sustaining an environment that i s conducive to development, which encompasses macroeconomic stability and good governance. 7.2 With respect to the public expenditure implications, the PRSP states that "Finally, the government, guided by the findings from the stakeholders' workshops, will confine its financial interventions mostly to (i)education (notably at primary school level), (ii) health (primary health care), (iii) agriculture (research and extension), (iv) roads (in the rural areas), (v) water, (vi) judiciary, and (vii) HIV/AIDS." In addition, the PRS also indicates that "expenditure allocations for the `core' sectors are expected to increase significantly when compared to the allocations for other sectors. In particular, total allocation for "other charges" for all priority sectors i s projected to increase 51 (as a proportion o f discretionary expenditure) over the medium term from about 21 percent in FY 2000/01 to over 22 percent and nearly 26 percent in FY 2001/02 and FY 2002/03, respectively. Simultaneously, the allocation for "other charges" for the core areas i s expected to increase, respectively, from 11.3 percent to 13.4 percent and 16 percent." Definitioninthe budget guidelines 7.3 The definition o f priority sectors inthe budget guidelines differs from those inthe PRS in that anti-corruption i s also included as a priority sector. In addition, the FY03 priorities. The FY04 budget guidelines expand the set o f priority expenditures as ".. for budget guidelines also added energy as another "important sector" to the spending the need o f effective poverty reduction, consolidating and sustaining macro-economic stability, realizing high economic growth, strengthening law and order, anti-corruption reforms, attracting investment, protecting environment, stimulating rural based agro- industries, and enhancing foreign exchange earnings, the following sectors are included inthe list ofpriority areas: energy, lands,police, andprisons." 7.4 The alignment o f this year's budget guidelines with the attainment o f the goals established in the PRS presents some problems. The 2002 Poverty and Human Development Report shows how, despite strong progress in some areas such as primary education enrollment rates and vaccination, the attainment o f most targets contained in the PRS will pose an enormous challenge. Namely, despite significant progress in implementing sector-wide reforms in priority sectors like health and education, as evidenced by the second PRS progress report, resource availability constitutes a binding constraint to the achievement o f their targets. According to the budget guidelines themselves, the shortage in funding between "requirements" and "ceilings" for priority sectors in 2003 alone amounts to T.Shs. 150,067 million, with similar gaps in the outer years.' Given this shortage o f funds, the addition o f new "priority" areas in the 2003 budget guidelines raises important questions. 7.5 Moreover, the share o f "priority" in overall expenditure i s becoming fairly large, thereby endangering the intrinsically relative meaning o f the concept o f prioritization. Before addition o f new areas, priority expenditures amounted to 57.8 percent of total non-debt expenditure. The additional priority sectors account for 11.8 percent o f total non-debt expenditure, leading to a total o f 69.6 percent considered "priority" by the 2003 budgetguidelines. Finally, when some original PRS priority sectors likejustice only get 7 percent o f their requirements, it makes it difficult to substantially continue to include them among the "priority" areas. 'This does nottake account ofthe fact that some ofthe figures includedas requirementsinthe budget guidelinesare far smaller thanthose containedinsector programs.Inthe case o f primary education, for instance,the figure containedinthe PEDPas requirementis 323,049 Tshmillionwhilethe figure contained inthe budgetguidelinesis only 243,970, with a resultinggap of 79 Tshmillion. Similarly,inthe water sector, the requirement figure includedinthe financial submissionofthe PERworkinggroup is 16.6 Tsh millionlower thanwhat appears as requirement inthe budgetguidelines. 52 7.6 The PRS should be a living document. However, due process and the use o f data and analysis inits modification are fundamental to maintaining the credibility o fthe PRS, o f its participatory and transparentnature and o f the commitment o f the government to its implementation. Therefore, the addition to or amendment o f priority areas as defined in the PRS should follow the PRS process and, preferably, be conducted in the context of the three-yearly PRS review. 7.7 The precise implications o f being a priority sector are not clearly defined. In practice, being a priority sector has the following implications: 0 protection o f budgetallocations from cuts duringbudgetexecution; 0 quarterly instead o f monthly expenditure releases; 0 priority inresource allocation duringthe budget process; 0 support for annual sector PERs; 0 usually among the first to be included inpublic sector reform initiatives; and 0 greater focus o f donor resources on priority sectors. 7.8 While the prioritization o f expenditure allocations will be analyzed indetail inthe subsequent sections, it i s important to note that at present there i s no general rule as to what prioritization inthe allocation process implies. Government is encouraged to define the meaning of "priority" sectors regarding the allocation of domestic funds. This definition should, at the least, include a commitment to the maintenance o f the share o f priority expenditures indomestically-funded expenditure and o f their absolute allocations in real terms. It would also add transparency if there were a clear commitment that any real increases inavailable funds will be applied preferentially to the priority sectors. Real expenditure increases outside the priority sectors should require a clear justification in terms o ftheir impact on poverty reduction objectives. 7.9 Tanzania has indeed fully protected budget items funded from the consolidated account from cuts during budget execution. The protection o f allocation does not extend to donor funded items in the development budget and as discussed elsewhere, there are indeed fairly large discrepancies between budgeted amounts and reported expenditures. These discrepancies are the result o f delays in project implementation as well as o f incomplete capture o f donor expenditures in the government accounts. Given the importance of donorfunded expenditures in these sectors, every effort should be made to ensure realism of budget estimates of development expenditures and strengthened monitoring of development expenditures in thepriority sectors. 7.10 The system o f quarterly releases to the priority sector ministries is very much welcome. The beneficiary MDAs report that quarterly releases remove the more severe strictures o f the cash budget system and allow them to improve their planning and strategic resource allocation, hence supporting their achievement o f PRS goals. Non priority MDAs, on the other hand, report extreme difficulties in their ability to plan and to run their operations in a rational and efficient manner. Government is encouraged to extend the system of quarterly releases to all MDAs given the limited risk posed by the extension and the dfficulties the cash budgeting system imposes on the planning and operations of non-priority MDAs subject to monthly releases. 53 7.11 One o f the potentialpitfalls of the strong focus on priority sectors inthe PER and public sector reform process is that service delivery and financial performance o f non- priority sectors receive relatively less scrutiny. This isparticular truefor the PERprocess and it would be important to pay greater attention to non-priority sectors and ministries by undertaking every year sector PERs for one or two non-priority ministries, starting with the largest ones. The PER process could then be used to monitor follow up on recommendations and implementation o f these sector PERs. Government is also encouraged to examine the areas not defined as priority in the PRSfor savings to be re- allocated to thepriority sectors, given thefunding shortfalls in the latter. Can Pro-Poor Expendituresbe AdequatelyMonitored? 7.12 A fiscal transparency review of standards and codes (ROSC)' carried out by the IMF highlights progress in enhancing fiscal transparencythrough the implementation of various reforms in recent years. These include the implementation o f the Integrated Financial Management System, the government-led, participatory PEWMTEF process, the enactment o f a new legal framework for fiscal management and procurement, enhanced controls on commitments and expenditures, and the quarterly publication o f the amounts and timing o f budget transfers to spendingagencies and local governments. The assessment also acknowledges recent efforts to institute quarterly reporting o f revenue and expenditures by subnational governments and to make these reports public, and the requirement o f the national anti-corruption strategy that each ministry i s required to report and publish corruption cases. 7.13 To further enhance fiscal transparency, the ROSC recommends focusing on the establishment o f a sound legal framework, rules and regulations governing the relationship between different levels of government for fiscal management in light o f the increasing devolution o f political, fiscal, and administrative authority to subnational governments. Budget documents should also provide information on public debt, financial assets, contingent liabilities, and tax expenditure. 7.14 Tanzania ranks highon the scale o f public availability o f fiscal data. The Ministry o f Finance publishes quarterly fiscal data on its website, World Bank and IMF reports are public and available through the internet and local authorities are mandated to post their budgets up for scrutiny by the population. Moreover, the participatory nature of public expenditure reviews provides a forum for government representatives, the donor community and civil society to engage in the analysis and discussion o f available data, including the government's budget guidelines. 7.15 With respect to information on poverty reducing expenditures, the PRS contained projections for expenditures in the seven priority sectors, broken down into expenditure on personal emoluments, other charges, and development. It also contained projections for the priority sub-sectors o f primary education, primary health, and agricultural research and extension and a disaggregated list of priority expenditures which defines core expenditures and provides expenditure projections at the item level. http:llwww.imf.orglextemal/pubsiftlscrl2002lcr0259.pdf 54 7.16 The first and the second PRS progress reports provide information on expenditures inthe priority sectors and priority areas. However, the PRS progress reports do not contain a detailed breakdown o f expenditures similar to the list o f priority items provided inthe PRS. 7.17 A review o f the data reported in the PRS and the PRS progress reports indicates inconsistencies across PRS reports, but also between the PRS and PRS progress report data and data presented in the budget books, the budget guidelines for FY04 and sector PERs. Since the PRS progress reports do not contain the precise definition o f what i s included in the sectoral expenditure data, it i s difficult to ascertain the source o f the discrepancies. In addition, there are several problems in capturing adequately priority expenditureswhich are discussed below. 7.18 One cause o f inconsistencies in the data may be found in classification problems o f expenditure. This seems to be particularly relevant for the education sector, where spending on basic education appears to be only partially captured. This is a consequence o f the fact that financial support under the primary education development program (PEDP) i s allocated to the sector through various ministries and many o f the relevant budget codes were not known at the time o f the preparation o f the PRS. Thus, expenditure on basic education appears to be under-reported inthe PRS progress report. To enhance transparency, it will be important that the PRSprogress report contain a precise dejnition of what is included in the priority sector expenditures. In addition, it will also be important that sector PERs pay greater attention to monitoring PRS expendituresand try to ensure consistency in coverage, definition, and source of data. 7.19 For several priority sectors including primary education, primary health, rural roads, and water, service delivery i s the responsibility o f the local authorities. Information on expenditures from local authorities i s thus key to assessing developments inpriority expenditures. At present, where most local government spending is financed through conditional grants from central government, it i s possible to ensure that local government expenditures conform with PRS priorities. However, as fiscal decentralization reforms progress and local authorities are given more scope to decide on the use o f central government grants, targeting expenditures on the priority sectors will become more difficult. In Tanzania, revenue collected by local authorities represents only a relatively small share o f their overall resources and lack o f information on local government expenditures i s unlikely to present a significant omission. However, some local authorities that have relatively high local revenuemobilization also use some o fthat revenue to supplement central government funding for the priority sectors. 7.20 Adequately capturing development expenditures for the priority sectors i s another problem. Although the capture o f donor-funded development expenditures inthe budget has improved, a significant share is still not adequately captured. This i s likely to lead to an underestimation o f spending in the priority areas, although the PRS seems to make some adjustment for some o f the development expenditure not captured in the budget. For example, for FY02 the PRS estimates actual development expenditure for the road sector at T.Sh. 113.9 billion while the appropriation accounts only indicate spending o f T.Sh. 50.1 billion. Although it is well known that the appropriation accounts typically 55 underestimate development spending, we consider it prudentto work with the figure from the appropriation accounts rather thanwith other estimates o f spending. 7.21 Another problem inthe assessment o f expenditure performance inrelation to PRS objectives i s the fact that data are limited to government expenditures in the priority sectors. In cases where there i s service delivery by NGOs, faith-based organizations, or the private sector, focusing only on public expendituresmay give a misleadingpicture of actual resources devoted to address specific issues. This i s particularly important for decisions on resource allocations and the targeting o f resources, where efforts by the private sector andNGOs need to be taken into account. 7.22 Finally, data on priority sector and area spending provide only very limited and conjectural information as to their impact on poverty reduction. To ascertain that the identiped priority sector expenditures contribute indeed to poverty reduction, detailed evaluations of their impact onpoverty will have to be carried out. For example, based on the household budget survey, marginal benefit incidence analysis could be undertaken to assess whether priority expendituresindeedbenefitthe poor. DoPublicExpendituresReflectPRS Objectives? 7.23 The PRS presents information on priority sector expenditures in nominal terms. The primary data source for the subsequent analysis are the PRS progress reports, which provide data on actual spending in the priority sectors up to FY02 and the budget estimates for FY03. Projections o f expenditures for the period FY04 to FY06 are obtained from the budget guidelines for FY04-FY06. This information has been complemented by information from the budget books, appropriation accounts, and sector PERs. For our analysis, we have adjusted the spending figures shown in the PRS for basic education and for spending on roads in line with the discussion in the previous section. The data annex presents the adjusted nominal expenditure data. We convert nominal expenditure figures into real per-capita expenditures at constant FY03 prices. This not only permits useful comparisons over time, but also provides a more direct indicator o f the potential increase in service delivery to individual Tanzanians, as it takes into account population growth. 56 Figure 17. Real per capita expendituresin priority sectors doubles between FY99 and FY04 3 5 , 0 0 0 . 0 0 3 0 , 0 0 0 . 0 0 2 5 , 0 0 0 . 0 0 2 0 , 0 0 0 . 0 0 P r i o r i t y B r o a d 1 5 , 0 0 0 . 0 0 a p r i o r i t y N a r r o w 1 0 , 0 0 0 . 0 0 5 , 0 0 0 . 0 0 7.24 We start the analysisby consideringoveralltrends inspendingon priority sectors and more narrowly defined core priority areas. As figure 17 shows, per capita spending on the priority sectors increasedby 53 percentfrom about T.Sh. 14,500 inFY99 to T.Sh. 22,100 in FY02. The budget for FY03 and the budget guidelines for FY04-FY06 project per capita spendingto increase to about T.Sh. 30,000 by FY04 and remain at that level thereafter. 7.25 The picture for more narrowly defined priority expenditures conveys a similar picture. They increasedfrom about T.Sh. 7,400 inFY99 to T.Sh. 13,100 inFY02 and are projectedto reach T.Sh. 20,600 by 2005. Indeed,narrowly definedpriority expenditures increased faster than broadly defined priority expenditures, i.e. by 77 percent over the periodFY99-FY02 comparedto the 53 percentincrease inpriority sector expenditures. Sectoral Developmentsin Priority Expenditure Figure 18. Spending on education, health, and roads claims 84 percent of priority sector spending /BBasic education II J U " " J ~ " ~ "2%~ ' " 4% 3% Pnmary health 4 0Water Education 11% 45% OAgncultura (Research 47% and Extension) mLands 11' Rural roads Health 17% Judiciary 18% 7.26 Figure 18 shows the distribution of expenditure across the priority sectors. Individual priority sectors receive very unequal shares of total priority expenditures. Educationis by far the biggest and claims 47 percentfollowed by roads (20 percent) and health (17 percent). Together, these three sectors claim about 84 percent of all priority 57 expenditure. Water and agricultural research and extension each claim six percent while judiciary and HIV/AIDS each only account for two percent o f priority expenditures. While these relative allocations within priority spending reflect to a certain extent resource needs of the priority sectors, it has nonetheless potential policy implications. In particular, to the extent that sector performance can be boosted by increasing expenditures, the marginal cost o f increasing performance inthe smaller sectors might be much less than in the larger sectors. This should be taken into account in making resource allocations. 7.27 Considering the narrow definition o f the priority sectors, the key difference arises from the exclusion o f trunk and regional roads from expenditures in the roads sector, whose share shrinks from 20 percent o f broadly defined priority sector expenditures to only four percent o f the more narrowly definedpriority sector expenditures. Figure 19. Per capita spending has more than doubled in most priority sectors between FY99 and FY03 16000 14000 12000 10000 8000 6000 4000 2000 0 7.28 As figure 19 shows, all the priority sectors have experienced significant increases inrealper capita spendingover the past four years. Itis important to note that the smaller sectors show higher percentual increases than the larger sectors, although the larger sectors have received the bulk o f additional resources in absolute terms. However, the figure also indicates that for sectors, for which international spending norms exist, Tanzania's expenditures are still significantly below these norms. For example, the cost o f a minimum health care package has been estimated to be between US$ 9 to 12, while spending inTanzania i s only around US$ 5. However, lack o f trained medical personnel makes it unlikely that the minimum health care package could be implemented successfully in the short to medium term. The human resource constraints in the sector are aggravated by the HIV/AIDS pandemic, which i s causing a further depletion o f human resources. Absorptive constraints in other sectors are similar. For example, work carried out previously indicates a severe shortage o f qualified road engineers at the district level, which constrains the implementation o f rural road development and maintenance programs, even at current levels o f funding. As the sector PERs document, 58 further increases inspending onthe priority sectors needto go hand inhandwith relaxing existing absorptive constraints, such as lack of trained personnel in the health sector or limitedplanning capacity inthe water and agriculture sectors at the district level. IncreasedPriorityExpenditures-HavePlentifulResourcesUnderminedEfficient Reallocation? 7.29 In addition to documenting the substantial increases in spending on the priority sectors and the core priority areas, it i s also o f interest to examine the source o f funding o f these increases. Are they a result o f overall increased spendingby government or do they reflect a re-prioritization of government expenditures? A related question i s whether the increases inpriority sector spending are mainly financed by donors, driven largely by decisions that are made outside the budget process, or whether the increases reflect a genuine prioritization by government o f budgetary resources through the budget process. Figure 20. Increases in discretionary spendingexceed increasesin priority sector spending Real Expenditure Increases 600000 500000 400000 300000 200000 100000 0 -100000 &qDiscretionary Expenditure Priority Sector (broad) 7.30 Figure 20 shows that, except for FYOO, increases indiscretionary spending" have exceeded increases in priority spending. This implies that the observed increases in priority sector spending could be financed from additional resources available to government and did not require a reduction inreal spendinginnon-priority sectors. loDiscretionary spending is defined as total expenditures (development and recurrent) minus statutory payments (mainly debt service andpensions.) 59 Figure 21. Real expenditure increasesin priority sectors and priority areas, FY00-FY06 Real Expenditure Increases FY99 FYOO F Y O l FY02 FY03 FY04 FY05 FY06 Riority Sector (broad) 0 Riority Sector (narrow) 7.31 Duringthe period FY00-FY03, increases in broad priority sector spending have generally beenhigher than increases in narrow priority expenditures, i.e., real increases on narrow priority expenditures were funded from additional resources available to the priority sector without requiring reprioritization within the sector. The projections for FY04-FY05 indicate higher absolute increases innarrow priority sector expenditures than in overall priority sector expenditures. This will require a shift of resources from non- priority to priority areas within the priority sectors. Figure 22. Percentage change in priority and non-priority expenditures % change in priority and non priority sector expenditure 60.0% 50.0% 40.0% 30.0% 20.0% 10.0% 0.0% -10.0% -20.0% -30.0% 7.32 Inadditionto comparing absolute increases inspending, prioritization can also be assessed by comparing the relative increases invarious expenditure categories. Figure22 60 reveals that priority sectors apparently have not received consistent priority in budget allocations, as the non-priority sectors experienced higher relative increases than the priority sectors in FYOl and FY02. Similarly, percentage increases for the core priority expenditure areas exceed the increases in broadly defined priority sector expenditures only in FYOO and FYOl. InFY02 and FY03, the situation was the reverse, implying that within the priority sectors, expenditures outside the core areas increased actually faster than those inthe core areas duringthose years. Figure 23. Share of priority and non-priority sector expenditures in total discretionary expenditures,FY99-FY06 100% 90% 80% 70% 60% 50% 11 ca Priority Non-Riority Broad 40% 30% 20% 10% 0% FY99 WOO F Y O l FY02 FY03 FY04 FY05 NO6 7.33 Analyzing priority sector expenditures as a percentage o f total discretionary expenditures gives the same picture (Figure 23). After an initial increase inthe share o f priority sector expenditures in total discretionary expenditures from 52 percent in FY99 to 59 percent in FYOO, the share subsequently declines to reach about 49 percent in FY06. Given severe financing gaps in the priority sectors as indicated in the budget guidelines, there appears thus to be scope for sharper prioritization of expenditures, even when taking into account that some expenditure components such as the pay levels for civil servants, can only be increased across the public sector and thus lead to an increase inexpendituresbothinthe priority andnon-priority sectors. 61 Figure 24. Change in priority sector expenditures, FY00-FY03 Change in Riority Sector Ependitures (TSh. million at constant Fy03 prices) 250,000 200,000 c 5E 150,000 100,000 k 50,000 I woo F Y O I FY02 FY03 Change in Recurrent Riority Change in Development Riority 7.34 Figure 24 decomposes increases in priority sector expenditures into increases in recurrent and development expenditure. BetweenFYOO and FY02, most of the increase in priority sector spending is attributable to increases in recurrent expenditure. This situation arises partly from the current scope o f priority sectors, which are heavily concentrated on social service delivery and were the initial priority was to enhance service delivery capacity through funding o f recurrent items such as drugs or text books. However, this picture also points towards a potential weakness inthe current definition o f priority sectors. Development and maintenance o f capital intensive infrastructureinareas such as transport and energy i s largely outside the PRS priority areas. If funding for these areas were to be neglected since they are currently not defined as PRS priorities, the prospects for long term economic growth could be affected. Comparisonto OriginalPRS Projectionsfor the Priority Sectors 7.35 Expenditureson priority sectors exceed PRS projections. The PRS contained projections for expenditures in the priority sectors, disaggregated into spending on PE, OC, and development for the period FYOl - FY03. Comparing these projections to actual spending in FYOl and FY02 and budgeted spending for FY03 shows that actual performance typically exceeded the PRS projections. Total priority sector spending was 3 percent below the target in FYO1, but exceeded the PRS projections by 9 and 26 percent inFY02 and FY03, respectively (Figure 25). The slight shortfall inFYOl was on account o f unders ending on the development account in the health, water, and agriculture sectors. IP l1Donor funded development expenditures is typically only partially capturedinthe government accounts, although the situation has improved in recent years. This may explain the observed under-spending in FYOl on the developmentaccount. 62 Figure 25. Spending on priority sectors exceeds PRStargets 1,000,000 900,000 800,000 700,000 600,000 500,000 400,000 300,000 200,000 100,000 I FYOl FY02 FY03 7.36 Comparing PRS projections and expenditurehudget data sector by sector indicates that there was initial underspendinginthe health, water, and agriculture sectors as well as on HIVIAIDS. Underspending in these sectors during FYOl and FY02 was compensated by spending in the education sector which significantly exceeded the PRS target. The budget for FY03, however, allocates sufficient resources to all sectors that match or exceed the original PRS projections for that year. Only the figures for HIV/AIDS related spending are still below the projections, although this may be partly due to persistent problems inadequately capturing HIV/AIDSrelated spending. 7.37 Finally, we consider the match o f PRS projections and actual spending by expenditure categories, i.e., personnel emoluments, other charges, and development. Spending on wages and salaries (PE) matched most closely the PRS projections. Spending on OC exceeded PRS projections with an ever increasing margin, reflecting the government's commitment to the financing o f the priority sectors. Finally, development spendingwas below projections inFYOl, and above projections inFY02 and FY03. The improvement reflects both better information on donor-financed development projects and a bettertargeting o f donor resources on the priority sectors. Targeting of Expenditures 7.38 There are two principal types of targeting. The first is referred to as "broad targeting." Inthis case, government tries to achieve its targeting objectives by allocating resources to sectors or areas that are likely to benefit the poor, without however making an effort to identifjr individuals who are most needy. Most poverty programs now stress reallocating expenditures to basic social services. But supplyingmore basic services does not necessarily go hand in hand with more poor people using them. In addition to reallocating expenditures, efforts should concentrate on ensuring that the poor make use of the services made available to them. 7.39 The second approach, known as "narrow targeting," entails identifying groups according to their income or geographic location. Narrow targeting can take three forms: individual/household assessment, categorical targeting, and self selection. 63 Individual/household assessment is a method in Box 4. Targetingand governance which an Official "If the poor lack power, the benefits of poverty programmes are assesses, household by unlikely to reachthem - or, ifthey do, to makealastingdifference. household, or individual by Effective targeting follows from empowerment, not the other way individual, whether the around. The very term targeting probably clouds the issue: betterto applicant is eligible for the talk more generally about focusing poverty reduction resources. program. Categorical The poor have to be organizedto advance their interests - to stand targetingmethod refers to a a chance of being heard and taken seriously. Once organized, the in which in a poor will find that politicians are more interestedin their fate and specific category - say a governments more responsive to their demands, Other popular forces will show increasedinterest injoining them in alliances for particular age group or change. And more anti-poverty resources will be directed their region - are eligible to way. Generally it is safe to assume that resources will not flow to receiveUnderself benefits. the powerless, despite the many rhetorical flourishes to the selection, a program i s contrary." (UNDP,2000). open to all but the design Recent research by Coary et al. (2002) also indicates that targeting dimensions that performance appears to be better in countries with higher income are thought to encourage and greater "voice and accountability". 7.40 Tanzania's approach i s almost exclusively "broad targeting" with the focus on priority sectors and priority expenditures. There are barely any "narrowly," targeted programs inplace. Given Tanzania's tight resource constraints, it might be worthwhile to consider methods o f targeting public expenditures and access and usage o f public services more directly to the poor. Inconsidering such targeting, it also will be necessary to take into account the administrative, private, incentive, social and political cost o f targeting.l2 7.41 Inparticular, the optionof geographictargeting basedon recent sectoraland household budget survey information and related poverty analysis should be consideredto ensure a greater impact of limited public resources on the poor. Given the gaps that the Poverty and Human Development Report reflects in almost all social indicators between rural and urban areas, it i s essential to re-direct expenditure toward rural areas andthe poorer regions and districts. 12 administrativecost - mainly the cost of collecting information, for example, for example conducting means testing for a householdor conductinga survey on which to base apovertymap); private cost - costto household of participating ina programor getting access to benefits; incentive changes inhousehold behavior in order to become eligible to benefit from targeted programs; social - social stigma attachedto be seen as padcipating in a targetedprogram; and political cost- reduced broadbase political support for targetedprograms. d 64 Box 5. Are transfers to the districts pro-poor? Districts have the responsibility for service delivery in most of the key PRS priority areas - i.e., primary education, primary health, agricultural extension, rural roads, and water. Financing for these services is provided through transfers &om the central government to the districts. At present, the size of the transfers to districts i s primarily determinedby historical endowments with facilities (primary schools, health clinics, etc.) to provide for the staff cost as well as some non-wage inputs for these facilities. It i s commonly held that this system of transfers benefits the better-off districts, which are likely to have greater facility endowments. A reform of the system of fiscal decentralization in currently underway and likely to replace the current system for determining transfers with a formula based system, where the poverty status of a district would explicitly enter into the transfer allocation formula. In preparation of these reforms, a detailed studyI3 of the current systems was carried out, which yielded several interesting andpartly unexpectedfindings with respectto the pro-poor bias ofthe transfer system: Larger (more populous) districts receive slightly fewer central government grants inper capitaterms; Rural local governments(those with lower population density) receive slightlylarger grantsper person; Wealthier local governments (as indicated by higher monthly consumption expenditures) receive substantiallyhigher per capitagrants; and Higher poverty rates also results insystematically greater grants. The study concludes that there is a substantial pro-poor tendency in the allocation of local government resources. However, even if districts with higher poverty rates receive larger per capita grants, this does not automatically imply that the poor in these districts benefit. The pro-poor orientation of spending at the district level and more specifically, designing mechanisms that would encourage such an orientation inthe public expenditure management of local authorities has received relatively little attention inthe discussion of pro-poor spending and fiscal decentralization. This is an issue that will have to be tackled in a reformed system of intergovernmentaltransfers. What Does DevelopmentAssistance Finance? 7.42 One of the key motivations for the HIPC debt initiative was to allow countries to redeploy resources that were previously used for debt service for increasing expenditure on poverty-reducing expenditures. Given the fungibility of resources and annual fluctuations in the overall resource envelope of government, earmarking of HIPC resources for specific purposes is neither desirable nor analytically meaningful. What i s of interest i s government's overall allocation of resources for priority expenditures. 7.43 However, just as it analytically and practically not meaningful to ring-fence certain expenditures as being funded by HIPC debt relief, it i s similarly of limited use only to compare amounts of HIPC debt reliefto changes inpoverty targeted expenditures. For example, since several priority sectors benefit from significant donor financing, attributing such a donor financed increase in priority expenditures to the availability of HIPC debt relief sources would imply double counting. What i s thus requiredi s to relate donor financing from all sources to increases in priority spending in order to be able to assess whether the totality of donor financing supports the increase in priority expenditures or whether due to the fungibility of donor support it also finances l3 Roy Bahl et. al. Developing a System of Intergovernmental Grants in Tanzania. 2003. http:l/isp- aysps,gsu.edu/projectsltanzanialindex2.html 65 expenditures outside the priority sectors. Doing this type o f analysis i s o f particular importance as more support i s provided trough general and sector specific budget support. 7.44 Figure 26 compares development assistance from all sources14(grants, loans, and HIPC debt relief) to government spending on the priority sectors. There are several important observations that can be made. First o f all, HIPC debt relief available to Tanzania since FYOO has not led to a significant increase in total development assistance to Tanzania. In fact, increased access to HIPC debt relief in FYOl and FY02 was more than offset by a decline in other forms o f development assistance. Figure 26 also shows that initially development assistance available to the government o f Tanzania exceeded its spending on the priority sectors. However, as part o f the implementation o f the PRSP government continuously increased its spending on the priority sectors and in FY02 government spending on the priority sectors in fact exceeded development assistance available to Tanzania. Using data on the more narrow definition o f priority expenditures indicates that donor funding significantly exceeds government spending on core priority areas in the priority sectors. The ratio o f narrow priority sector spending to total ODA has increased from 50 percent to 70 percent over FY99 - FY02. So has the ratio o f broad priority sector spendingto total ODA which i s now over 110 percent. Figure 26. Official developmentassistance from all sources and spending on priority sectors (as % of GDP) I 9 0% 8 0% 7 0% 6.0% 5 0% 4 0% 3 0% 2.0% 1 0% 0 0% 1 FY99 FYOO FYOl FY02 MTotal Official DevelopmentAssistance HPriority Sector (Broad) 0Priority Sector (Narrow) 7.45 Aside from looking at the overall relationship between donor assistance and priority sector expenditures, it i s also of interest to examine whether increases in development expenditures since FYOO in the form o f HIPC debt relief and other ODA have been matched by an increase in priority sector spending. Figure 27 indicates that l4Table 6 only shows HIPC assistance f?om multilateral institutions, since bi-lateral HIPC debt relief is captured in a reduction o f debt service. The estimate o f ODA in figure 26 also includes an estimate o f the value o f bilateral debt relief. 66 increases in priority expenditures were initially below increases in development assistance. However, in FY02 the increase in priority expenditure (using both a narrow and broad definition o f priority sector expenditures) over the FY99 level has exceeded the increase inODA. Figure 27. Increasein developmentassistance and priority sector expenditure compared to levels in FY99 3.0% 2.5% 2.0% 1.5% 1.O% 0.5% 0.0% I FYOO FYOl FY02 7.46 In addition to the overall relationship between donor funding and priority expenditure, it i s also useful to examine the use o f "flexible donor resources." Flexible donor resources include general and sectoral budget support as well as HIPC debt relief." Figure 28 shows the increase in flexible resources over the level attained in FY99 and compares it to increases in recurrent priority sector expenditures. This analysis reveals that recurrent expenditures on the priority sectors have only gradually increased over the past three years. The increase in flexible donor resources is still higher than the increase inpriority expenditures. This impliesthat flexible donor resources also finance increases in recurrent expenditures outside the priority sectors. Figure 28 also shows resources available from HIPC support and it appears that in FYOO and FYOl the increase in priority spending was less than resources in HIPC debt relief. Only in FY02 has HIPC debt relief beenmatched by an equal increase inpriority sector expenditures. *'We calculate "flexible resources" as the difference betweenoverall developmentassistance shown inthe government accounts and foreign-funded expenditureon development.Choosing this method of calculating "flexible" resources i s necessary because of earmarked sector development and basket funding arrangements, which contain elements ofbothproject and budget support. 67 Figure 28. Increase in flexible aid resources and recurrent priority sector expenditures over FY99. 3.0% 2.5% 2.0% B 1.5% 8 1.O% 0.0% w o o P l O l FY02 Increase in Flex. Financing Increasein RecurrentRiority Sector Spending 1 7.47 In conclusion, the above analysis indicates that the increase in overall priority sector spending between FY99 and FY02 now exceeds the increase in donor resources over the same period. The increase in spending on the priority sectors has been facilitated by a greater focus o f donor assistance on the priority sectors in the development budget. At the same time, government has used flexible donor resources to increase recurrent spending for priority and non-priority sectors. The government might want to consider ways that would ensure that increases inflexible donor support are used to enhance spending in the priority sectors and to cover the signijkant resource gaps existing in these sectors, ifthe PRSP and MDG targets are to be met. Inparticular, it will be important to consider the use o f flexible resources for development expenditures beyond minimumrequirements for counterpart funds. 68 8. REVIEW OFALIGNMENT OF SECTORALPERWORK WITH PRS 8.1 Sectoral PER work i s one o f the key mechanisms for ensuring the alignment o f sector budgets with the PRS. They are being carried out annually as part o f the overall budget process to enhance its quality and to provide a forum for stakeholder involvement at the sector level. This chapter reviews the sectoral PER process, which during FY03 saw important reforms. It also reviews the alignment o f sector PERs with the PRS and the extent to which sector PERsmeet the objective of strengtheningthe analytic base for sectoral and inter-sectoral budget allocations. The Sector PER Process 8.2 When the participatory PERprocess was launched in 1997, sector working groups (SWGs) were established as the platform for dialogue among stakeholders on sectoral policy and expenditure issues and to guide the sectoral PER processes. Membership in the SWGs included representatives of the concerned ministries, donors, and NGOs. In subsequent years, SWGs ceased to exist in most sectors although sector ministries with the support o f consultants continued to prepare sector PERs. At the PER retreat held in Bagamoyo in August 2002, one o f the main issues raised was the limited ownership by the sectors o f the PER process and on o f the key recommendations was to re-establish SWGs for the priority sectors. Subsequently, in FY03, SWGs were formed comprising representatives o f the concerned ministries, NGOs, research institutes, and donor agencies. These SWGs have brought more openness to the budget process at the sectoral level. However, experiences differed widely across sectors, as the role o f the SWGs in the ministerial budgetprocess was not clearly defined. While for some sectors, the PER work seems to be well integrated into the budget process, in others it appears to be carried out in parallel to the budget process with only limited interaction between the two. In this context, we would recommend clearer understanding and agreement on the role o f the SWGs in the budget process. While it goes without saying that budgetary decisions are the responsibility o f the ministries and government, SWGs should play a well-defined role in the budget preparation and review process to enhance participation and transparency. In addition to the role o f defining and overseeing the work on sector PERs, the role o f the SWGs should also include at a minimum. 69 0 review and discussion o f sectoral submissions prior to their submission to the budget guidelines committee; 0 reviewand discussion of sectoral budgetproposals prior to their submission to the MinistryofFinance; and 0 review and discussion o f the sectoral MediumTerm ExpenditureFramework. 8.3 Another issue that surfaced was the relationship between various budget related processes at the sector level. These include the PRS, the annual budget, the PER, the MTEF, the performance budget, as well as sector development programs for primary education, health, and agriculture. The multiplicity o f processes appears to result in parallel bureaucratic processes and hightransaction cost. Inmost sectors there appears to be considerable scope for streamliningandharmonizing these processes. 8.4 The usefulness o f sector PERs lies intwo areas. The first is to provide a better information basis for the Ministry o f Finance to make cross-sectoral allocations. The second i s to enhance allocative efficiency within the sectors. In order to make sector PERs more useful to the Ministry o f Finance, it should, inconsultation with sectoral PER working groups, consider providing a GFS/IFMS-consistent standard form for all ministries to specify their financing requirements as input into the budget guidelines. In addition, the designation o f contact people at the Ministry o f Finance for the priority sectors would allow a more fluid communication with line ministries and sectoral working groups. These contact people would regularly attend sectoral working groups, provide the interface between the sector and the Ministry o f Finance, and participate in the decision-making process of inter-sectoral resource allocation. They would also be the conduit for information feedback to the PER working group regarding eventual allocations to the sector inthe budget guidelines and the actual budget. 8.5 Another problem that has hamperedthe usefulness o f the sectoral PER processes was timing. The analytic PER work was often delayed and not available at key stages o f the budget process, in particular at the time of the preparation o f the budget guidelines which define fiscal strategy and priorities for the coming year. The timing o f the PER process needs to be fully aligned with the budget process to ensure that analytic inputs are provided to the relevant line ministries in time for their consideration when elaborating financing requests to the Ministryo f Finance for the elaboration o f the budget guidelines, the MTEF, and the annual budget. Although the current process has been designed with this alignment in mind, in practice there have been frequent slippages in the delivery o f inputs and the establishment o f realistic time tables for PER work i s thus key. 8.6 The results-orientation o f sectoral PER work could be further enhanced. Inthis regard, the PER working groups, line ministries and the Ministry o f Finance are encouraged to use information from the output-based budgeting reports and performance- based budgeting information from the Civil Service Department to increase the focus o f their work onefficiency and increase accountability. 8.7 Inthe medium-term, the government may consider extending the PER process to non-priority sectors to enhance openness, transparency andplanning andreview. A more 70 comprehensive sectoral coverage o f the PER would also enhance its operation as an open process o f competition for funding across sectors. 8.8 Some participants inPER sectoral working groups have expressed the view that a comprehensive yearly review of the sector may be too demanding and not always necessary. In their view, one or two sectors per year could be selected on a needs basis for comprehensive reviews. For other sectors, while keepingthe broad policy framework inmind,the working group would focus on a reduced number o fkey emerging issues in need o f further study. In all cases, particular attention would be paid to puttingtogether the financing requests to be submittedto the relevant ministries for eventual use in the elaboration o f the budgetguidelines by the Ministry o f Finance. Review of Sector PERs for FY03 8.9 This section reviews this year's sector PERs. The purpose o f this review in the context o f the external evaluation i s to monitor the quality o f sector PER work and to identify areas for improvement. The more rigorous the sector PER analysis of performance and the more sound its recommendations, the better informed will be budget and policy decisions, and sector expenditure will likely perform better in the following fiscal year. Through this process, sector expenditures increasingly become an effective tool o f development for poverty reductionpolicies and strategies. 8.10 The primary quality assurance mechanism for sectoral PERs are the reviews by the SWGs and by the PER working group. This review draws on comments and assessments that were provided through these mechanisms. However, in addition it i s also envisaged to add a cross-sectoral perspective to the review. As this is the first attempt at rating sectoral PERs, both the framework and the process for carrying out the assessment o f sectoral PERs will need to be further developed as part o f the PER process. It is hoped that this review o f sectoral PER work will be adopted as a regular feature of the external evaluation process. As such, it will also serve as a summary and "institutional memory" o f the discussions o f sector PER work that take place during the PER process. 8.11 The focus o fthe review is on the comprehensiveness and adequacy of the analysis contained in the sector PERs as the analytical basis for the preparation o f the MTEF and various sectoral budget submissions. Generally, the information and the analysis requiredinclude an analysis o f recent expenditure trends, particularly o f PRSP priority activitiedareas; an assessment of the output performance against PRSP and sector targets; a review o f policies and plans; and an identification and/or estimation o f PRSP/sector priority activities and their full financing budgetfor the medium-term(2003/4-2005/6). 8.12 This year, PERs for Education, Health, HIV/AIDS,Water, Roads, and Agriculture were prepared. Generic terms o f reference for sectoral PER work were produced by the PER secretariat and approved by the PER working group. SWGs then adapted the generic terms of reference to reflect the specific situation an analytic needs o f each sector. Subsequently, the sector PER work was undertaken by the sector ministrieswith the support of consultants andinputsbythe SWG. Drafts ofthe sector PERreports were 71 reviewedby the SWGs and finally cleared by the PER working group intime to feed into the preparation ofthe MTEFandthe budget. 8.13 The following criteria, which are an expanded version o f those developed in the FYOl PER16,were usedto revieweach sector PER: 0 quality o f the analysis o f the consistency o f sectoral objectives and targets with the PRSP; 0 quality o f the analysis of the adequacy o f the sectoral policy framework to address and achieve the statedtargets; 0 sector coverage, expenditure coverage and tracking, and degree o f definition o f the activity basis for achievingthe targets; 0 treatment o f past sector PER recommendations; 0 output performance against PRS and MTEF targets; 0 the presence of implementation or absorptive capacity constraints which limit the effective use o f allocated funds; 0 development o f costings; 0 expenditure forecast and full financing; and 0 prospects for sector-based financing (user charges, insurance funds, efficiency gains, intra-sectoralreallocations, etc). 8.14 The results o f the review for each sector PER are presented in Annex B (Education, Health, and HIV/AIDS) and Annex C (Water, Roads, and Agriculture). Overall, expenditure management and execution have significantly improved since the commencement o f the PER process. The improvements, however, are uneven, as reflected by the varying quality o f sector strategies, plans and performance, costing and full-financing estimations o f this year's sector PERs. The results can be summed up as follows: Plans,PolicyFrameworks,Priorities,Targets 0 In some cases there is a divergence and/or ambiguity between the sectoral ministries' objectives (as set out inthe sectoral strategy document) and the targets set out inthe PRSP (HIV/AIDS, Roads, Agriculture). 0 The policies and strategies o f the sector ministry are sometimes poorly defined, without clearly identifying a set o f plans for achieving targets (HIVIAIDS, Agriculture). The Scope of the Sector PER Study 0 For some sectors, the coverage o f the PER study is not comprehensive as a result o ftime or data limitations (Health, HIVIAIDS, Water). l6 EmergingMarket Economics Ltd.Budget Sustainability and Full Financing of the Priorities. Processed. April 11, 2001 72 ExpenditureCoverageandTracking 0 The coverage o f sector expenditures and activities in some sector PERs i s not comprehensive, particularly with respect to activities carried out by LAs and projects that are externally funded. (Roads, Agriculture, HIV/AIDS). 0 The availability and accessibility o f qualitative and quantitative data and the compilation o f such data has been a challenge, particularly where there are many institutions/ministries involved or are institutionally responsible for the same sector (Agriculture, HIV/AIDS, Roads). OutputPerformance,Indicators,andTargets 0 Some reports have no output targets (HIV/AIDS, Agriculture). Hence, the link between actual expenditure and output to make a judgment about efficiency and effectiveness of sector expenditure cannot be clearly established. 0 Outputs/targets do not distinguish PRS output from sector output (Roads). 0 The expenditure impact cannot be measured owing exogenous factors that affect output or because output i s difficult to measure (Agriculture, HIV/AIDS). 0 Given resource and capacity constraints and past slippages, PRS targets may be too ambitious (Roads, HIV/AIDS, Water). ActinguponRecommendations 0 Some o f the sector PERs did not fully discuss or report on the status and actions taken upon the recommendations or findings in the previous PERs (Health, HIV/AJDS,Roads, Agriculture). ImplementationandAbsorptionCapacity Constraints 0 There are implementation and absorption capacity constraints at all levels (Education, HIV/AIDS,Roads, Agriculture). 0 Given capacity and implementation constraints, actual expenditures fall significantly behind allocated expenditures(Roads, HIV/AIDS). CostingandForecastingthe Full-financingfor PRSP 0 Not all sectors provide full and consistent costing/projections o f either PRS priority or Sector activities (Education, Roads, HIV/AIDS). Though the Health sector PER provided full financing cost in priority areas, there were no priority actions identified, or a breakdown o f what the aggregate figures may constitute. The following sub-section discusses these issues infurther detail. 0 Possibilities o f sector-based funding and cost-savings were not fully explored (HIV/AIDS,Agriculture). Ranking Sector PERs According to the Review Results 8.15 This section rates the sector PERs on the adequacy o f their analysis and information. Although in place only for a few years, the sector PERs have already proven themselves to be useful and have significantly contributed to progress made in policy and public expenditure planning, coordination, targeting, and execution, particularly of poverty reduction programs. The rating and the ranking will provide a simplified tabular representation of where, in aggregate, weaknesses remain and o f their 73 relative importance. In the culture of continuous improvement, the objective of this exercise i s to highlight the most pressing weaknesses and gaps, prioritize actions, and encourage improvements inthe sector PER work which is critical for the implementation o f the PRS. As stated before, this i s the first attempt at rating sector PERs and should thus be considered as a pilot for future improvement and development. In particular, while the present review was carried out by a consultant, in future it would be desirable to fully integrate this assessment of sector PERs in the work o f the external evaluation teams with full discussion by the sector and PER working groups. TheRatingmanking Method 8.16 The ranking o f sector PERs i s done according to the eight criteria discussed above. The scoring ineach category i s based on the detailed qualitative reviewpresented inAnnexes B and C. The score assigned ineach case tries to capture the adequacy ofthe analysis and information provided inthe sector PER. The scoring ranges from `0' to `5', where `zero' denotes the topic was not discussed, `1' is the lowest rating, and `5' i s the highestrating andrepresents fully satisfactory treatment. 8.17 Given that the adequacy o f sector PERs analysis and output can be affected by many factors, the assessment method i s such that it does not discriminate betweenthese factors. For example, if the depiction o f `expenditure trends' category o f a sector PER i s poor or not comprehensive because o f limited availability and accessibility o f information in the field, the score will be low. If the information is available and accessible inthe field but, somehow, was not adequately presented inthe PER, the score will also be equally low. 8.18 A limitation o f this ranking is that the review covers only eight categories that provide the most basic informatiodanalysis required from all sector PERs. It does not cover many other topics discussed at length by the various sector PERs that are also useful to MTEF and PRS implementation. TheRankingResults 8.19 Representing qualitative judgement innumeric form is not a perfect science. The exercise, however, i s not about providing precise numbers. It i s about highlighting and providing an approximation o f the relative distance o f the current analysis and information in sector PERs from providing fully satisfactory treatment. The gaps that emerge mean that improvements in policy articulation, institutional processes and procedures, or information gathering and dissemination, etc., are needed to make the PER a more useful document to MTEF and the PRS implementation. 74 Table 18. RankingFY03 sector PERs CriterialCategory Sector PERs Education Health HIV/AIDS Water Roads Agriculture Average a) Consistency of Sector and PRS targets 3.0 5.0 na 4.5 3.0 4.5 4. b) Assessment of policies to achieve targets 3.0 na 1.o 3.0 3.0 3.0 2.6 c) Sector Coverage, Expenditure Coverage 1.5 4.5 3.5 4.0 3.0 3.5 3.3 d) TreatmentofRecommendations ..I/ 5.0 2.5 2.5 5.0 3.5 2.0 3.4 e) Output Performanceagainst PRSTargets 4.0 4.5 1.0 4.0 4.0 2.0 3.3 f ) Identifying Absorption Capacity Constraints 2.5 3.5 4.0 3.5 4.5 3.5 3.6 g) Identifying priority activities and costing 2.0 0.0 0.0 4.0 3.5 3.5 2.2 h) ForecastingPRS and Sector Expenditures 1.5 1.5 0.0 4.0 2.0 3.5 2.1 i)ProspectsforSectorBasedFinancing 3.5 3.0 0.0 4.5 4.5 0.0 2.6 I / Discussing the status of previous PER recommendationsandproviding new ones. na: denotes not applicable. 8.20 The results are presented,inTable 18. The justification for each score inthe table i s provided in the detailed review in Annexes B and C. On aggregate, the results demonstrate that sector PER weaknesses/strengths vary considerably by sector and by category. On average, the Water and Roads PERs scored the highest while the HIV/AIDS PER scored the lowest. The Health, Education, and Agriculture PERs provided satisfaction treatment o f some issues, but displayed also significant deficiencies inother area. All ofthem, however, needto make improvements inkey categories. 8.21 Interms of categories, most sector PERs present a good discussion o f sector and PRS targets, which are typically presented as being well aligned. Sector PERs also did a good job in analyzing expenditure trends and in analyzing absorptive capacity constraints. With respect to criteria such as the treatment of recommendations, assessment o f performance against PRS targets, or the assessment o f the adequacy o f policies to achieve targets, there i s significant variation across PERs, with some o f them dealing with these issues in a fully satisfactory manner while others display severe weaknesses. With respect to `identifying priority areas and costing' and `forecasting PRS and sector expenditures' most PERs appear to be rather weak, although even in these areas there are examples o f very comprehensive and substantial coverage. FullFinancingCost Estimatesandthe FinancingGap Sector PER vs MTEF Cost Estimates 8.22 A required output of sector PERs is the identificationand the costing of bothPRS priority and sector-wide activities for full financing in the coming periods. If this requirement is properly carried out in the sector PERs, the projected expenditure in the MTEF will reflect those figures and will enable the MTEF to earmark adequate resources to PRS priority activities. The MTEF may choose to revise the PER figures for their weaknesses to the extent possible or adapt second-best estimates. This i s a key test for 75 the sector PERs. Table 19 below illustrates the two sets of estimates-the sector PERs and the MTEF. 8.23 To begin with, the estimates provided by the sector PERs exhibit varying degrees of focus and/or weaknesses. For example, inthe Agriculture PER, detailed projections o f only Agricultural Sector Development Program (ASDP) priority areas but no projections for other areas are provided. Inthe Health PER, only aggregate projected figures for the priority areas are provided, with no breakdown by activity or program, and no information i s made available on non-priority areas. For the Education PER, the projections encompassed only o f Secondary Education and Teacher requirements. The Water and Roads PERs provide only projections for the whole sector and no estimates are provided for priority areadactivities. Moreover, the Roads PER figures reflect only maintenance costs. Finally, the HIV/AIDSPER does not provide a cost estimation o f any kind. Table 19. Summary of PER and MTEFfull-financing estimates (Tsh billions) 2003104 -2005106 Sector 2003104 2004105 2005106 PER* MTEFRq Res.Av. PER* W E F R q Res.Av. PER* MTEFRq Res.Av. Agriculture n.a. 116 116 n.a. 134 112 n.a. 154 111 ASDP Priorities 56 56 56 60 60 60 72 73 73 Education ...11 364 518 492 287 540 513 215 579 544 Basic Education n.a. 364 364 n.a. 380 380 n.a. 407 407 Health ... 21 n.a. 362 236 n.a. 398 244 n.a. 437 256 Priority Activities 213 213 192 228 228 205 249 249 224 HIVIAIDS ... 31 n.a. 43 22 n.a. 43 26 n.a. 44 29 Water ... 41 98 98 86 97 97 82 96 101 92 Roads ... 4151 170 392 290 195 411 276 220 450 317 Total 1,485 1,116 1,580 1,136 1,451 1,250 (PER*) Requirement accordingto sector PER; (MTEFRq) Requirementaccordingto MTEF; (Res.Av.)ProjectedResource Availabilityaccordingto MTEF. ForPER columnonly: 1/ PER estimateIncomplete.Excludesprimaryeducation.Includesexpansion of Secondary SchoolRequirementsandadditional costs of Teacher Education. 2/ According to a list submittedby MoHto MOF. 3/ Not provided. 4/ Sector requirement 51 Includesmaintenance only. 8.24 Ifone compares the sector PER estimates against the MTEF, it seems that some of the sector PER projections were u s e h l to the MTEF. The MTEF directly used sector PER figures in the following categories: (i)ASDP priorities in Agriculture; (ii) Health priority activities; and (iii) Water sector projections. No sector PER provided estimates for both PRS and sector-wide activities. 8.25 The MTEF presents cost estimates for HIV/AIDS and, at least, considerably amended the PER projections for Roads, and Education. Moreover, behind the aggregate figures, there could be differences in composition. For example, the Education sector PER puts emphasis on Secondary education and attaches a Tsh 338.1 billion cost thereby 76 accounting for 92 percent o f the cost indicated in Table 19, while that o f the MTEF i s predominantly for basic (primary) education. 8.26 The above comparisons show that there is ample room for improvement in the cost projections o f the sector PERs. It would be desirable for each o f the SWGs to compare notes with the MTEF task team to see how they can improve their cost projections in the future. It would also be desirable for all sector PERs submit estimates for both their sector as well as for PRS requirements. TheFinancing Gap 8.27 The total financing gap (`MTEF requirements' less `MTEF resources available') for fiscal year 2003/4 i s Tsh 369 billion (Table 19). The resource gap i s about 25 percent o f the full financing estimate. The sector distribution o f the gap, however, i s disproportionate. HIV/AIDS activities face the most serious challenge in that available resources cover only half o f the requirements. In sectors where the priority areadactivities were identified (Agriculture, Education, and Health), the resource available for priority areas i s 97 percent o f the requirement. Thus, most o f the funding problems infiscal year 2003/04 appear to be inthe non-priority areas. TheInterpretations& Implicationsof the `FullFinancing' Estimates 8.28 The purpose of identifying the PRS priority sectors, areas, and activities, and the full-financing estimates i s to base budget decisions on the best information possible in reaching the PRS targets. Though it i s recognized that the poor costing in some sector PERs i s the consequence o f a lack o f detailed activity plans and necessary baseline information, estimations could have been made based on available and accessible information. Insome sectors (Education and HIV/AIDS), the failure to provide estimates i s worrisome, as `full financing' estimates were recently prepared. 8.29 Inbrief, some o fthe key implicationsofthe above are: The PER costing and projections as inputsto the MTEF are weak. It is unfortunate that some sector PERs did not identify and/or cost PRS priorities to be considered for MTEF programming and the full financing o f their requirements. The implication of this i s that PRS priority sectors are less likely to receive full funding and it is thus more likely that PRS targets and objectives will not be achieved. Since there are serious absorption capacity constraints in some sectors (Roads, HIV/AIDS), actual expenditures will likely remain far below the allocated budget. For these sectors, absorptive capacity (rather than full-financing requirement) should determine the level o f budget allocation. As discussed earlier, given that in some sectors (HIV/AIDS, Agriculture), there is no clear link established between a level of expenditure and output or there are no stated sector targets, it i s not clear what the `full financing estimates' o f Table 19 represent. Inthe strict ideal sense, `full financing' means or implies the costing o f projects/activities that will leadto a sought-after specified poverty reductiontarget. 77 StylizedLessons and `Next Steps' 8.30 Given these gaps, the question is what can be done about them and what lessons and next steps emerge for MTEF and PRS implementation. The individual sector PERs provide recommendations on further improvement o f the sector PER process to strengthen MTEF and PRS implementation in the coming year. The following provides some general indications. (a) Strengtheningthe Weaker Sector PERs 8.3 1 There is a need to strengthenthe sector PERs, notably for HIV/AIDS,but also for Agriculture, Education, Roads, and Health. To cover key gaps it may be necessary to provide support to the SWG and the concerned institutions to address identified weaknesses. (b) StrengtheningIdentification and Costingof Sector and PriorityActivities 8.32 There i s an urgent need to strengthen the costing o f sector and PRS priority activities so that it can be a more useful tool for MTEF preparation and priority targeting. After each annual cycle, the PRS, MTEF, and priority sector PERtask teams or working groups should meet and compare notes on how to improve their respective roles and inputsinto the following cycle. (c) Revisingand/or Developing PRS Targets 8.33 For sectors that have no sector-specific targets that are consistent with the overall PRS targets (HIV/AIDS and Agriculture), developing targets, intermediate or proxy, if necessary, should be priority. 8.34 For sectors that appear to be far from reachingtheir targets either due to financing constraints or absorptive and capacity constraints, it may be wise to review PRS targets for those sectors that have fallen behind (e.g., water). (d) Strengtheningthe Capacity, particularly of LAs 8.35 All sector PERs indicate that they face the most serious implementation and absorption capacity constraints at the LA level. These capacity constraints include institutional management, financial management, financial control and accountability, and procurement skills. There i s thus a need to strengthencapacity building at the LA levels. (e) Revisiting the Rate of Decentralization 8.36 Given the capacity and the implementation constraints o f LAs described above, there i s a need to review and orchestrate the pace o f decentralization across the priority sectors. The combined transfer o f fiscal responsibilities and resources to LAs will exert a tremendouspressure on LAs to manage resources and carry out programs. For example, inthe agricultural sector alone, 75 percent of total sector expenditure is plannedto be managed by LAs within 5 years. Consequently, the demand o f respective sectors for skilled and technical labor force (that was originally pooled at headquarters) will multiply. Since there is only so much human capital available inthe short-run, it may be wise to linkthe rate o f decentralizationwith the rate o f capacity building at the LA levels. 78 (f)TensionbetweenNationalandPRSPrioritiesandDecentralization 8.37 As decentralization progresses, an inevitable contradiction between the centrally orchestrated Sector Development StrategiedPlans and the demand-driven participatory planning-a key objective of decentralization, will have to be addressed. There i s thus an urgentneedto harmonize sector development programs and decentralization and the local government reform program. (g) StrengtheningInformationon Expendituresby DonorsandLocalAuthorities 8.38 Though there have been considerable improvements inthe financial management and accounting systems since the introduction o f the IFMS, almost all sector PERs indicate that there is unreported expenditure, mostly foreign financed expenditures at the LA level. The implication of these statements is that the real magnitude of sector expenditures are not yet fully accounted for. There is thus an urgent need to continue efforts to fully capture expenditures by donors and by local authorities. (h) Follow-upof Sector PER Recommendations 8.39 Future sector PERs should be requiredto follow up on the recommendations of past PERs. This does not mean that all recommendations proposed by any o f the sector PERs must be adopted. Rather, it means that recommendations need to be seriously considered and any actions taken uponthem should be reported. 79 PART111: SELECTEDISSUESINBUDGET MANAGEMENT 80 9. BUDGETMANAGEMENT ISSUES CashBudgetvs. CashFlow Management 9.1 Tanzania has maintained a cash budget system since FY97 whereby MoF limits aggregate expenditure in a month to average revenue collection in the previous three months plus program aid. MoF exercises this control by allowing vote holders to spend monies from their votes only to the amount it releases every month. Priority sectors get their releases every quarter. In the past, the cash budget has helped improve fiscal discipline by allowing continuous adjustment o f expenditures on other charges, to the resource stream. Monthly ex-ante control was desirable since information on budget execution was not available timely enough for adjustment and the cash budget system helped to prevent over-commitment and over-spending (above the appropriatedamounts) by the spending units. Also, it imparted realism to the budget-making process by taking away incentives to inflate revenue estimate^.'^ Coupled with a policy and desire on part o f the government to balance budgets, cash budgeting has thus been a useful instrument for controlling deficits. Most o f all, cash budget has been a credible indicator o f government's intentions to runa responsible fiscal policy. 9.2 Yet many things have changed since Tanzania instituted cash budgeting. (i) Budget management i s much stronger and budget monitoring and information systems have improved significantly. For example, budget execution can be monitored in almost real time with the IFMS. (ii) i s an effective PER process in place that has brought There improved transparency, and hence credibility, to expenditure management. (iii) The fiscal situation i s comfortable after HIPC debt relief and years o f prudent budget management. (iv) An MTEF is inplace. 9.3 Apart from the above-noted improvements that have weakened the argument for keeping a cash budget system, the cash budget has well known costs arising from low predictability o f resources for the vote holders. Importantly, the cash budget i s not an instrument to counter or offset the volatility of resource streams. It actually passes the volatility on to exchequer releases with some magnification. Figure29 shows that despite the cash budget exchequer releases are more volatile than actual expenditure and MDAs appear to be carrying out some `expenditure smoothing' on their own. l7Revenue estimates would be inflatedby the budget-makerswhen they want to theoretically provide for all the demands (often inflated) o f spending units. Against the fictitiously high revenue, units could then spendduring the year, resultinginhigher than designeddeficit when revenue did not materialize.The gap betweenestimatedrevenueandactual collectionhasbeendeclininginthe past four years. Infact inthe past two years collections were higher than estimates. Cash budget has also led to increased realism in expenditureestimates.PER FY02 found that the gap betweenactual andbudgetedexpenditures on O&M decreasedafter the institutiono fcash budget. 81 Figure 29. Monthly exchequer releasesand expenditure I Monthly Exchequer Releases and Expenditure j 200000 , 0 -Exchequer releases ----Recurrent+Local development expendnure 9.4 To translateimprovementsin financialmanagementinto betterpublic sector performance, government should consider relaxing cash budgeting and extend quarterly releases to all vote holders. Such relaxation would expectedly have associated fiscal risks arising mainly from delays and shortfalls in external program assistance (domestic revenue by itself i s fairly predictable and shows little variability). But these risks will be small and manageable. For example, in FY02 if government had fundedthe entire budgetedexpenditureas opposed to applying expenditure cuts underthe cash budget system, the fiscal deficit (after grants) would have been inthe range o f 1.1to 1.3 percent o f the GDP for the whole year. The risk in each quarter would be even smaller as adjustment after each quarter would still be an option. O n the contrary, the benefits o f more predictable resource flow to the MDAs would be many. Quarterly releases, if instituted, would need to be consistent with the pattern o f demand for resources. OfficialDevelopmentAssistance andFiscalManagement 9.5 The PER FY02 had listed various issues that needed watching in view o f the increasingly larger size o fprogram support. 0 Needto balance the recurrent and development expenditures ; 0 Additional risk of resource variability; 0 Possibility o f usingprogram aid as a buffer to domestic revenue variability; 0 Keepingpressure on for improving domestic resource mobilization; 0 Macroeconomic implications. 9.6 Many o f these issues are now coming to the fore and receiving government attention. The following paragraphs deal with some o f these issues. 82 Variability and shortfall of aid 9.7 As in other Sub-Saharan countries, aggregate aid displays higher year-to-year variability than domestic tax revenue in Tanzania (Table 20). Within aid, the program component has historically been more variable than theproject component. However, the variability o f the program component has been going down in recent years, specifically since 1998/99,when major bilateral donors switched to budget support, and while still more variable than domestic tax revenue, it now appears less variable thanproject aid. Table 20. Variability of revenue and aid Variance Variance Coeff o f Coeff o f (92/93 to (96/97 to variation Variation 01/02) 01/02) (92/93 to (96/97 to 01/02) 01/02) Domestic revenue 0.77 0.52 0.07 0.06 Tax Revenue 0.70 0.46 0.07 0.06 Non-tax Revenue 0.05 0.03 0.17 0.13 Grants 1.24 0.70 0.33 0.24 Budget support grants 0.57 0.49 0.51 0.56 project grants 0.51 0.44 0.40 0.34 Foreign Loans 2.55 0.26 0.66 0.23 Program loans 2.05 0.21 1.09 0.53 Project loan 0.36 0.34 0.54 0.44 Total Program aid 3.77 0.16 0.70 0.19 Total Project aid 1.47 1.22 0.42 0.34 Total (excl. HIPC relief) 6.39 0.81 0.45 0.17 9.8 Apart from year-to-year variability, within-year variability, particularly deviations from aid levels indicated by donors at the time o f budget preparation is also an issue. Figure 30 shows budget support resource flows in the last six quarters plotted against expected flows (based on annual commitment). While budget support fell much short o f the committed amounts inthe fourth quarter ofFY02, andthe first quarter of FY03, there was a sudden release o f funds by donors in the second quarter o f FY03. The volatility noticed here was not destabilizing because aid commitments were a relatively small part o f the budget. However, it i s an indication o f greater uncertainty that could accompany expectedly larger flows inthe future unless disbursementmechanisms are designed so as to create greater predictability o f disbursements. 83 Figure 30. Expectedand actually receivedexternal assistance 0 - Q1 Q2 Q3 Q4 Q1 Q2 W02 W03 9.9 In this context it is also important to acknowledge the effect of exchange rate fluctuations on the predictability o f disbursements. Part o f the difference between aid actually received and programmed in the budget reflects the changes between the bilateral exchange rates assumed at the time o f budgeting and actual at the time o f disbursement. Figure 31 plots average percentage changes in the exchange rate for four major currencies (US Dollar, Euro, Pound Sterling and Yen) during the last six quarters. Ineach o fthese six quarters the Shilling depreciated against all the four donor currencies leadingto a higher Shilling value o f aid. Figure 31. Average change in exchange rates, FY02-FY03 I Average Change in Exchange Rates 12.4 z Q2 FY02 Q3 Q4 1 Q1 FY03 9.10 Recent research covering a sample o f 37 countries shows that on average total aid fell short of donor commitments made at the time o f budgeting by about 20 percent, and program aid on average had higher shortfalls thanproject aid (about 32 percent compared to 20 percent)." These data precede the advent o f PRSC and PRBS in Tanzania, which are explicitly designed as annual budget support instruments. It i s expected that uncertainty surrounding actual disbursements should go down as these instruments get established. Yet it is unlikely that the annual gap between commitments and '*Aid and Fiscal Management by Ales Bulk and Timothy Lane, June 2002, IMF Working Paper wPi02I112 84 disbursements will completely go away because disbursements also depend upon the domestic budgetary processes and political and strategic priorities o f bilateral donors. Integration of Donor Assistance into the Budget 9.11 The PER FY02 had carried out a detailed assessment o f the objectives and modalities o f integrating external assistanceinto the budgetaryand financial management system. Donor assistance is getting more integrated into the budget in two ways: (i) through a general move from project support to program support; and, (ii)better recording o f in-kindand direct project support into government accounts. 9.12 Ascertaining precisely and documenting quantitatively progress in integrating foreign aid in the budget is difficult due to the lack of reliable and comprehensive information on all flows to Tanzania. We use four indicators to monitor progress. The first i s the share o f official development assistance (ODA) received by government (as shown in the central government operations tables prepared by the government and the IMF) intotal development assistance to Tanzania (as estimated by the OECD). Between 1999 and 2001,ODA to Tanzania i s estimated to have increased from US$ 990 millionto US$1,233 million. In 1999, central government operations tables showed 57 percent of total ODA accruing to government. By 2001, this share has increased to 58 percent. It i s worth mentioning that it i s unlikely for this share to increase to 100 percent since ODA accrues also to other parties than government, i.e, NGOs, the private sector, and the Central Bank, which is the recipient o f IMF support. 85 Figure 32. Capturing development assistance in the budget Government accounts show a marginally increasing A greater share of budgeted development share o ftotal ODA expenditure is shown at the end of the year in the appropriation accounts 59% , ~ 58% 8oy0 . ..-. ~ ._ ~ . . . ... . . ~ 70% 67% 1 58% 60% 58% 50% 40% 57% 30% 20% 57% 10% 56% 0% 1999 2000 2001 FY99 FYOO FYOI FY02 Programmehudget support accounts already for more A greater share of development expenditures is than 50 percent of development assistance to credit fmanced government c-- em,. f 58% ~ I"," 50% I50% 43% 40% 40% 32% 30% 30% 20% 20% 10% 10% 0% I 0% FY99 FYOO FYOl FY02 I FY99 FYOO FYOI FYO2 1 9.13 The second indicator we use i s the ratio o f development expenditure recorded in the appropriation accounts at the end of the fiscal year and budgeted development expenditure at the beginning o f the fiscal year and. This ratio has increased between FY99 and FY02 from 46 to 67 percent. This indicates improved recording o f development expenditures in the government accounts as well as improved project implementation and more realistic budget estimates. 9.14 The third indicator is progradbudget support as a share o f total development grants and credits received by government. Funds from this source are fully incorporated into the budgetary system and any relative increase in this source o f funding implies a better capture o f development assistance inthe government accounts. Between FY99 and FY02, the share o fprogrambudget support has increased from 32 percent to 58 percent. 9.15 Our last indicator i s the share of credit funding in total foreign project funding. The rationale for this indicator is that credit financed development expenditures are fully captured inthe budget and the appropriation accounts, as such development activities are 86 implemented by government using government systems. Between FY99 and FY02, the share o f credit financing inoverall project financing has increased from 28 to 42 percent. 9.16 Good progress has been made in improving recording and reporting o f development assistance. The Tanzania PER process has been very successful infostering improved recording and reporting o f development assistance, as acknowledged by the 2002 report by the Independent Monitoring Group. Starting in FY02 and improved further under PER FY03, a data collection exercise was done by the External Finance Department o f the Ministry o f Finance (EFD-MoF) in collaboration with development partners. In September 2002, EFD sent a request to Tanzania's development partners to provide data on disbursement and projections o f their support to the country. The information was subjected to donor verification as well as comparison with similar information on donor funded projects kept by ministries prior to being recorded in the Aid Flows Data base at MoF. This data was subsequently utilized as an input into the firming-up o f the resource envelope for the 2003/04 - 2004/06 budget guidelines prepared in December 2002. Donor compliance to the MoF request for aid information has been encouraging. More donor support i s increasingly being captured in the budget notably with the movement toward program support, PRBS and basket funding. 9.17 In spite o f the progress made thus far, there is scope for further improvement in disclosure and recording o f aid. One o f the remaining problems relates to reluctance/difficulty on the part of some donors to provide aid commitments and disbursement information, particularly where support i s given in the form o f direct project support, inkind or TA. Furthermore, even though MoF has prepared procedures on how to route development assistance through the Exchequer in the context o f the Public Procurement and Finance Acts (2001), some donors still find it difficult to apply the procedures. There are also capacity weaknesses related to effective involvement of sector ministries in the design o f the project work plan and related financing. The MoF can create better incentives for donors to provide information on the projections. If the donors see their projections being included in the budget guidelines, it will create incentives for them to give the exercise more attention. 9.18 Recent efforts to integrate development assistance into the budget process have almost exclusively focused on the central govemment budget. However, many donors also provide direct assistance to the districts. PO-RALG has launched an initiative to solicit and consolidate information on donor support to LAs. Local authorities typically maintain a list o f donors and NGOs operating in a particular district. Some local authorities also have quarterly consultations with the respective donors. However, there are significant differences in the level o f detail o f information on donor assistance maintained by the districts. 9.19 There i s also a need to incorporate the information compiled by PO-RALG into the Aid Flow Data Base inMoF. Inparticular, it would be useful if such a consolidated database could provide information on the regional distribution o f aid in Tanzania. What i s needed if for PO-RALG to improve the scope o f the information solicited from donors to correspond with the information compiled by MoF. Inturn, M o F needs to routinely get 87 this information from PO-RALG and incorporate/synchronize it with what already exists inthe Aid FlowDataBase keptbythe ExternalFinance Department. 9.20 Aside from the efforts o f MoF and PO-RALG to establish better information systems on donor assistance, there are also similar efforts going on at the sector level. For example the water sector i s trying to compile information on all water projects in Tanzania including those supported by donors and NGOs. 9.21 A continued problem is the adequate capturing o f donor funded technical assistance and little progress has beenmade inthis area. The problem arises from the fact that technical assistance i s frequently directly procured and provided by donor agencies inthe form of foreign experts. Aside from the problem of adequately capturingtechnical assistance inthe budget, Tanzania needs to address urgentlythe problem o f increasing the efficiency and effectiveness o f technical assistance and capacity building efforts, which claim a significant amount o f government and donor resources. Analytic work to assess the efficiency and effectiveness of technical assistance and capacity building efforts should be a priority to provide the basis for developing a coherent strategy for the use of technical assistance and capacity building efforts. Fiscal Management in thePresence of More Variableand Larger ProgramAid 9.22 Government is acutely aware o f the increased vulnerability o f its finances to a lumpier and a more centralized aid flow (as compared to decentralized project aid disbursedto sector ministries).Aid volatility has many costs: 0 Year to year variability o f aid makes it difficult to planinthe medium-term. 0 Delayed disbursements o f budget support within a year create cash flow problems. 0 Aid shortfalls within a year are the most difficult to manage because quick fiscal adjustment measures are difficult to implement, particularly where aid money finances recurring expenditures that cannot be easily adjusted inthe short term. 0 Sudden inflows o f aid can make liquidity management difficult for monetary authorities. 9.23 The MoFmust also contendwith an incentive problem linked to a greater share of budgetsupport inaid. When donor funds flow directly to the exchequer, sector ministries have less o f an incentive to aggressively work towards meeting the requirements o f release o f funds from donors; finding and getting resources then is a `problem' o f MoF and not o fthe sector ministries. 9.24 Government's response has been multi-fold: (a) Developing more accurate information on the volume and timing o f expected flows in the coming year as well as in medium-term during the budget preparationprocess. (b) Continue running a cash budget despite a relatively relaxed aggregate fiscal position and making only monthly releases to all sectors other than priority sectors. 88 (c) Impressing upon the donors (mainly for the budget support) to front load budget support disbursements. (d) Strengthening the financial management o f aid flows. 9.25 The Tanzania Assistance Strategy provides a framework within which government and donors are attempting to improve the predictability and pattern o f resource flows. The donors need to balance the costs o f disruption with the benefits o f ensuring good policies. The conditionality should be specific and appropriate. There should be a clear and common understanding o f conditionality by donors on one side and the authorities onthe other. 9.26 Government could also attempt making budget design more flexible in order to better manage aid variability. One option could be to develop a conservative baseline budgetby discounting donor commitments by, say, 10 percent and identifying items that would be financed only if disbursements exceeded the baseline. Expenditures that can be postponed without much disruption such as the purchase o f vehicles, training programs etc. could be budgeted against the highcase scenario. In such a budget, an inclination to spread the gap between base and high case scenario evenly over all line items should be avoided. Government could also resort to domestic short-term non-monetary financing andrunningdownreserves inline with ceilings andtargets set by the IMFprogram. Balancing recurrent and developmentexpenditures 9.27 Untiedprogram aid (budget support or balance o f payment support) can be used for increasing recurring expenditure, increasing development expenditure, reducingtaxes, or paying off debt. The exact combination o f these various possible uses depends on the nature o f aid - whether it i s temporary or permanent, and the current structure o f government spending. With donor resources shifting from project to budget support in Tanzania, government needs to pay more attention to providing adequate funding for development budget from its own resources and not leave it entirely to donor finances. While government has started funding some development expenditure (for example fast track road construction - T. Shs. 21.84 billion) from the total pool o f resources this year, non-donor-funded development expenditure i s expected to expand further in future. This will requiremore planningresources and capacity inthe concerned MDAs for developing and executing development budget. It will also require better integration between recurrent and development budgets to ensure that adequate funding i s provided for operation and maintenance o f development investments. Macroeconomic implications 9.28 The intendedpurpose o f aid is to accelerate growth andreduce poverty. Large and sustained volumes o f aid, however, have other macroeconomic implications both for the structure o f output and relative prices inan economy, inthe medium to long term. Inthe short-run aid transfers can increase the price o f nontradables relative to tradables, creating incentives for the movement o f productive resources from tradables to nontradables. This process could lead to shrinking o f the tradables sector. While this change i s not inherently bad, it may entail adjustment costs when aid finally tapers off. 89 Also, production o f traded goods leads to productivity improvements through `learning by doing', an opportunity which is lost whenless traded goods are produced. 9.29 On the external front, unilateral foreign transfers can cause real exchange rate appreciation, leading to loss o f export competitiveness. Where this happens, we would expect to see declining or negative growth inexports, everything else equal. But since aid transfers may be accompanied by a number o f other real shocks, for example terms o f trade improvements, a fall inexports may not be clearly visible. 9.30 Many experts argue that these `Dutch disease' effects may be offset in the medium-to long-term by the productivity gains, particularly those favoring production of domestic goods, that may accompany increased aid flows. 9.31 The empirical evidence on `Dutch disease' has been weak in most aid-receiving countries. In Tanzania itself, previous research actually showed that a depreciating real exchange rate accompanied increased aid flows inthe second half o f the eighties. Experts agree that policy response to observed `Dutch disease' i s not to reduce the volume o f aid. Possible policy responses to `Dutch disease' include removing import restrictions, investing aid to enhance productivity in tradable sector, and accumulating foreign reserves. 9.32 Analysis carried out inthe context o f the external evaluation provides preliminary evidence that the second half o f the nineties saw an appreciating real exchange rate - both in external as well as internal terms, despite a depreciating nominal exchange rate. The appreciation was accompanied by structural changes such as falling exports, particularly o f manufactured goods, and a declining share o f tradables in national output. Two possible reasons are (i) a higher domestic inflation rate compared with the main trading partners, and (ii) high level o f foreign aid flows. This analysis does not address the role o f various other `fundamentals' such as productivity changes, or changes in the terms of trade, inappreciation o f the real exchange rate. 9.33 The real exchange rate movements discussed above could well be equilibrating changes i.e., the real exchange rate might be moving towards a new equilibrium level. However, declining level o f exports, particularly o f manufactures, would be a source o f concern. Further liberalization o f imports i s a possible policy option that is likely to help exports and that would also take pressure off the exchange rate. Other possible policy responsesmust follow a more careful and detailed analysis. ImplementingNationalandPRSPrioritiesThroughLocal GovernmentBudgets 9.34 InTanzania, local authorities play an important role inthe implementation of the poverty reduction strategy. They are responsible for the delivery o f basic services inthe education, health, water, and agriculture sectors, which constitute the core priorities under the PRS. Ensuring the achievement of the national PRS targets requires thus effective coordination between the central government and the local authorities. At present, most o f the financing for these services i s provided through conditional grants from the central government to the local authorities, which gives the central government 90 significant leverage to ensure that activities o f local authorities contribute to the achievement o fthe PRS goals. 9.35 The government is currently pursuingdecentralization reforms with the objective o f strengthening public service delivery through deepening accountability and further empowering and capacitating local authorities. The local government reform program (LGRP) plays a key role in reforming the intergovernmental system and strengthening local authorities to enable them to take on greater responsibilities. A fiscal decentralization group with members from the government and the donor community i s currently considering options for fiscal decentralization and a reform o f the central government grant systembased on a comprehensive studylgcompleted inlate 2002. 9.36 Previous PERs have examined aspects o f local government public expenditure management including assessments o f the local government budget process, local government revenues, and financial management capacities at the local level. The focus o f this year's PER i s on the coordination o f central and local expenditure management through the local government budgetguidelines (LGBGs). 9.37 Local authorities prepare their own budgets and the fiscal year for local authorities runs from January to December, whereas the fiscal year for the central government runs from July to June. Budget guidelines prepared by the PO-RALG constitute the key link betweenthe central government budget and the budgets o f local authorities. The budget guidelines inform local authorities about the amount o f transfers provided to them in the central government budget as well as information on general government priorities. 9.38 However, representatives o f local authorities have often questioned the usefulness o f the budget guidelines. The first concern relates to the fact that the guidelines are often issued too late in the budget cycle, which makes it difficult for them to prepare their budgets in a timely manner. A second concern relates to the substance o f the budget guidelines, which is seen by some to impose undue restrictions on the use o f local authorities' own resources, thus undermining the broader goals o f decentralization. Another problem arises from the fact that due to the asynchronous budget cycles o f the central and local governments, local authorities usually only know the amount o f transfers they will receive during the first half o f their fiscal year, while funding for the second half remains uncertain. The budget guidelines contribute little to close this gap. Finally, there i s also the impression that the budget guidelines are prepared ina top-down manner with relatively little involvement of stakeholders including the local authorities themselves. 9.39 This section reviews the relevance and substance o f the LGBGs, their linkage with the central government budget guidelines, the MTEF, and the PRS, as well as the preparation process o f the LGBGs. The analysis i s based on the reviewof documents and l9URT. Developing a System of Intergovernmental Grants in Tanzania. Draft Report. November 2002. This study was conductedfor the Local Government Reform Programme by the Andrew Young School of Policy Studies Georgia State University 91 discussions with the President's Office Regional Administration and Local Governments (PORALG) in Dodoma, Dodoma Regional Secretariat, Dodoma Municipal Council, Dodoma District Council, Kibaha District Council and Regional Secretariat for the Coast Region. The following are the main findings: 9.40 Quality and Usefulness of the Budget Guidelines: The budget guidelines for LAs are supposed to be a vehicle through which PRS objectives and priorities expressed in the central government budget guidelines and budget are operationalized for implementation through the local authorities' budgets. However, at present the budget guidelines contain mainly instructions on the budgetprocess and budget formats and they are not elaborate enough on how to translate national priorities and PRS objectives at the local government level. The guidelinesmention the pro-poor sectors and assume that the LAs will separately understand objectives and policies o f these sectors and the standards that the sectors would like to see LAs maintain. However, interviews with a sample o f LAs indicated that this i s not always the case. The LAs admitted that they did not have some of the sector's documentatiordguidelinesdelineating sector objectives and policies and resource use. Some sectors expressed their dissatisfaction with LA budget guidelines, arguing that the guidelinesdo not clearly reflect their priorities. 9.41 It needs also to be recognizedthat there is a degree o f inevitable tension between decentralization imperatives - which imply widening local discretion over budgetary decisions - and national planning imperatives (as inthe PRS), which imply narrowing this discretion in favour o f determining local budgetary allocations through a top-down process. It is intrinsic to the basic rationale underlying decentralization that local governments are not simply implementerso f a national/sectoral priorities and plans. It i s only possible to capture the full economic benefits o f decentralization (allocational efficiencies) if decision-making about sectoral (and other) expenditures i s made by locally accountable bodies in line with local priorities rather than in narrow or strict accordance with national targets or plans(which may be allocationally inefficient). 9.42 Itwill be important as decentralization reforms progress inTanzania that the PRS and the budget processes, and the role and content o f the LGBGs are adequately adjusted. The above mentioned tension requires much more systematic examination, and a more clearly worked-through policy response, than it has receivedthus far. The type and nature o f the instruments that can regulate the tension between national planning and decentralization imperatives in the determination o f local budgets also will require -attention. standards or targets in education and health; procedural mechanisms such as The LABGs are one o f the primary mechanismshere, but there are also others national the role of the U S ' inbudget oversight and so on. A more fundamental review of both the substance and process o f the LABGs (anddirectly related instruments), probably on the basis of some technical inputprovidedby the LGRPwill be necessary as an important ingredient inthe discussion o f fiscal decentralization. 9.43 The guidelineswere also observed to be tedious, with too many tables to fill. Further, the requirement to detail progress in implementation o f the ruling party's election manifesto and promises contained therein was noted to be an overstretching exercise. Moreover, such a requirement was noted to impose undue restrictions, which 92 influence the allocation o f resources especially resources from own sources in order to meet these promises. Probably the requirement could be confined to issues related to implementation and efforts to meet the PRS targets. Otherwise as the practice i s now, resources are spread thinly in trying to meet all the election promises. The requirement o f setting aside funds for women and youth activities (10 percent o f budget) has been abolished by the Minister for Finance in his FY02 budget speech, on grounds that the resources were not used properly. The loans advanced to beneficiaries from these funds were not recovered, rendering the sustainability o f the funds impossible. MoF i s preparing a new procedure for managing such funds, that will involve using domestic banks to manage the funds on behalf o f government within the context o f micro finance regulations. However, this directive seems not to apply to LAs as they have continued to observe the restriction requiringthem to set aside funds for women and youth groups. 9.44 The guidelines could be improved further by: (i)putting more emphasis on national priorities (ii)clearly stating sector policies, priorities and strategies, albeit briefly, as mentioned earlier; (iii)requiring that LAs review past performance (covering revenue collection, expenditure, and inventory o f projects and resources).and status o f implementation o f national priorities; (iv) taking a longer term perspective (say 3 years) to budget guideline issuance, with only resource ceilings being provided on a yearly basis. Much o f the present content o f the budget guidelines concerning the budget process and forms to be usedcould be codified ina budgetmanual. 9.45 Budget guidelines preparation process: The basis for the preparation o f the LGBGs are the central government budget guidelines which are issued in DecembedJanuary and the central government budget, which i s issued in June. PORALG has primary responsibility for the preparation o f the budget guidelines for local authorities and typically issues the local government budget guidelines in October. Sector ministries are requested to provide inputs to the formulation o f the local government budget guidelines and both sectoral ministries and local and regional authorities have an opportunity to review the draft budget guidelines prior to their issuance. The sector ministries were o f the view that they would like to be more fully involved in the preparation o f the LA budget guidelines. L A s also shared a concern expressed by sector ministries that there was no clear forum between sectors and LAs, where sectors could elaborate their requirements to LAs and agree on how best to meet them. The Association of Local Authorities in Tanzania (ALAT), an important stakeholder in LA affairs, has, however, never been fully involved in the preparation o f LA budget guidelines. Restoring the role of the LGBGs in consolidating national and sectoral priorities for implementation by local authorities i s also important to ensure consistency between sectoral approaches to decentralization. The current system, where the dialogue between LAs and sector ministries takes place almost exclusively on a bilateral basis gives scope to inconsistencies and duplication and further underlines the need for a more open and participatory approach in the formulation o f LGBGs. The establishment o f a forum that brings together LAs, regions and ministries i s thus key to enhance the budget guidelines process for local authorities. The budget guidelines committee o f the central government provides a model for a similar setup for the preparation o fthe local government budget guidelines. 93 9.46 Inadditionto establishinga more consultative budget guidelines process for local authorities, it will also be important to ensure appropriate attention and participation by LA representatives inthe central government budgetprocess. This is of highimportance, since the bulk o f funding o f local authorities i s in the form o f transfers from the central government, decisions on which are made in the context o f the central government budget process. Enhanced participation in this area o f the central government budget process coupled with proposals to reform the local government grant system i s also important to enhance transparency in the allocation o f funds across districts. A formal process o f consultation with clearly identified LG representatives to take place annually should be considered. In the longer term, as i s the case in a number o f other countries, this process shouldbe legislated. 9.47 Timing of the Budget Guidelines: There have been concerns from LAs and other stakeholders that budget guidelines for LAs are issued too late in the budget cycle. This late issuance o f budget guidelines either made it difficult for LAsto prepare their budgets in a timely manner or compromised the quality of budgets, as these were to be prepared ina hurry. This practice was seen to go against the lawhegulation, which requires all LA budgets to be approved two months before the commencement o f the next fiscal year. Local authorities indicated that unlike inthe past few years, there was some improvement inthe timing of budget guidelines, with the LA budget guidelines for FY03 beingissued inSeptember 2002. The local authorities visited appreciated this turno f events andwere hopeful that this timetable will be maintained in future. Despite this improvement, more needs to be done to see to it that LA budget guidelines are issued on time as per laid down regulations and that LAs are given ample time to prepare their plans and budgets properly and ina consultative way. 9.48 Difference in Fiscal Yearsof Central Governmentand LAs: The fiscal year for the central government runs from July - June, while for local government the fiscal year coincides with calendar year, that i s January - December. In the budget guidelines for FY04, the government has indicatedthat the two fiscal years will be harmonized and that steps leading to the harmonization will be taken during the FY04 for operationalization during FY05. 9.49 The adoption o f asynchronous fiscal years had the objective to reduce the unpredictability o f resource availability for local authorities. Due to often unrealistic revenue estimates at the central level, which ledto significant alterations to the budget in the course o f budget implementation, local authorities suffered a large degree o f uncertainty. In addition, estimates o f resources to be transferred to LAs were in most cases not known until very close to the budget session. This left very little time for LAs to prepare their budgets before the start of a new fiscal year. Separation o f the fiscal years was undertaken as a measure against this problem. 9.50 However, with the asynchronous budget cycles o f the central and local governments, local authorities only know the amount o f transfers they will receive during the first half o f their fiscal year, while fundingfor the second half remains uncertain. The usual practice has been for LAs to assume that funding levels during the first half o f their fiscal year would remain the same during the second half o f the fiscal year. In addition, 94 the difference in fiscal years also complicates reporting by LAs. Inrecent years resource predictability for local authorities has been greatly enhanced following the adoption of the PRS which has brought about prioritization and protection o f allocations to priority sectors. This has created a situation where it i s appropriate to reintroduce identical fiscal years for central and local government. 9.51 To ensure that the realignment o f fiscal years yield the expected benefits, it will be important to 0 Carefully plan the transition for local authorities, who will have to prepare a budget covering the period January - June 2004 before they switch to the synchronized budgetyear with a budget for July - June 2005. 0 In order to provide sufficient time for the preparation of local government budgets, the central government guidelines will need to have a firm indications o f resources available to the local authorities. 0 The synchronization o f the budget year will also require a synchronization o f central and local government budget guidelines, creating an opportunity for improved involvement o f local government stakeholders inbothprocesses. 0 The MTEF will need to pay more attention to allocations to local authorities to fulfill its role of enhancing resource predictability and facilitating the eventual transition to a reforms grant allocation systemfor local authorities. 9.52 Innovations in the local government budget process: The budget preparation process described in the findings o f the PER FYOO has not changed much. A notable addition inthe process o f budgeting i s the discussion o f all LA budgets within the region by the Regional Consultative Committee (RCC) before they are finally approved the respective LA councils. The RCC i s chaired by the Regional Commissioner and Regional Administrative Secretary (RAS) is secretary to the committee. Other members include all district commissioners, all executive directors, all MPs inthe region, and all chairpersons o f various LA committees. Representatives o f NGOs operating in the region are invited as ex-official memberdobservers. Another notable positive development i s the effective involvement o f the regional secretariat in scrutinizing the draft plans and budgets o f LAs and in consolidating some o f LA reports to PORALG and subsequently MoF. However, the secretariats complained o f lack of capacity in some sectors and resource constraints for effective supervision and monitoring o f adherence to standards in service delivery by LAs. Updateon Assessment of GovernmentCapacityto Track PovertyReducing Expenditures 9.53 In2001, the Bank and the Fundcarried out ajoint assessment o f the capacity of Tanzania's public financial management system to allow reliable tracking o f poverty- reducing expenditures. The assessment used a list o f 15 benchmarks which cover budget formulation (comprehensiveness o f the budget, budget classification, and projections), budget execution (internal controls and reconciliation), and budget reporting. Eight out o f the 15 benchmarks were considered to be met at the time o f the initial assessment, which 95 put Tanzania among the bestperforming countries inAfrica. The assessment suggested that budget formulation is quite good; however, further improvements are required in budget classification and account codes, and in accounting for donors' disbursements; 0 budget execution procedures meet most standards except in the case of internal audits; and 0 budget reporting is relatively satisfactory; however, improvements are requiredin external audit and application o f GFS-compliant functional classification. 9.54 Based on this assessment, a concrete action plan was developed including steps to enhance the capacity for tracking poverty reducing expenditures. Table 21 revisits the assessment as well as progress in implementing the agreed actions. Overall, Tanzania has beenmaking steady, although sometimes slower then expected progress inaddressing weaknesses. As Tanzania has now more clearly definedpoverty reducing expendituresin the budget, that benchmark i s likely to be met, increasing Tanzania's rating to having met nine out o f the 15 benchmarks. Government is encouraged to accelerate the implementations in other areas, including completion of budget re-classijkation efforts, monitoring of expenditures by local authorities, and strengthening internal and external audit. However, it also needs to be acknowledged that many of the remaining weaknesses are related to the tracking o f donor funded and local government expenditures. Progress in these areas will thus require a concerted effort o f all parties involved. 96 e k- e, ? N m r m 3 hl 0 3 ANNEXES 104 ANNEXA: RECENTEXCHANGE AND TRADE RATE DEVELOPMENTS The Bagamoyo retreat had raised the issue of macroeconomic implications of continued high level o f foreign assistance. Several possible macroeconomic implications are described in Box 1. One key issue that often captures the attention of academics and policymakers alike i s the behavior o f real exchange rate, looked upon as a composite measure o f export competitiveness o f a country. Ifincreasing resource flows provided as program support are spent on non-tradables (say domestic services), they can create an upward pressure on the real exchange rate, affecting international competitiveness o f Tanzanian exports. A sustained increase in the real exchange rate can lead to shrinking o f the manufacturing sector in the economy. This annex presents the analysis o f actual behavior o f real exchange rates in Tanzania during the decade o f the nineties. However, it needs to be pointed out that although aid flows are likely to have an important impact on the exchange rate and competitiveness, other factors may have an equally important role. For example, the recent increase in exports o f gold could also be a source o f Dutch disease effects, with the volume o f recorded exports o f gold being similar to recorded levels o f aid inthe balance o f payments. RealEffective Exchange Rate Tanzania carried out major economic reforms in 1986 and 1989 which involved devaluation o f the currency, trade policy reform, and public sector reform. Foreign exchange was liberalized later in 1992 and 1993, and in 1994 a market determined foreign exchange rate was adopted. Figure A1 plots the evolution o f the real effective exchange rate (REER) during the last two decades.22REER sharply depreciated after 1985 till 1989, appreciated till 1991,remained relatively stable till 1994, and appreciated again till 2000. Inthe last two years it has slightly depreciated. Further analysis focuses on the post-1990period because of major changes inthe exchange rate regime after 1990, whichmakes previous years' outturns non-comparable. 22REER is calculated interms of domestic currency Le., a fall inREERindex indicates a real appreciation of Tanzanian Shilling vis-his the currencies o f the major trading partners. CPI is used as the price index. Actual trade shares o f 21 trading partners in 1993 are used as weights. The year 1993 was selected because by 1993 foreign exchange liberalization was completed, parallel market rates were very low and exports were stable. IMF reports REER based on Information Notice System (INS) trade weights that take third country competition into account. Calculations using I N S weights produce very similar results to the ones reported above. 105 Box 1. Macroeconomicimpact of aid The intended purpose of aid is to accelerate the growth and reduce poverty. Large and sustained volumes o f aid, however, have other macroeconomic implications both for the structure of output and relative prices in an economy, in the medium to long term. Inthe short-run aid transfers can increase the price of nontradables relative to tradables, creating incentives for the movement of productive resources from tradables to nontradables. This process could lead to shrinking of the tradables sector. While this change i s not inherently bad, it may entail adjustment costs when aid finally tapers off. Also, production of traded goods leads to productivity improvements through `leaming by doing', an opportunity which is lost when less traded goods are produced. On the external front, unilateral foreign transfers can cause real exchange rate appreciation, leading to loss of export competitiveness. Where this happens, we would expect to see falling exports (or growth in exports), everything else equal. But since aid transfers may be accompaniedby a number of other real shocks, for exampleterms of trade improvement, afall inexports may not be clearly visible. Many experts argue that these `Dutch disease' effects may be offset in medium- to long-term by the productivity gains, particularly those favoring production of domestic goods, that may accompany increasedaid flows. The empirical evidence on `Dutch disease' has been weak in most aid-receiving countries. InTanzania itself, previous researchactually showedthat a depreciating real exchange rate accompanied increased aid flows in the second half of the eighties. Experts agree that policy response to observed `Dutch disease' i s not to reduce the volume of aid. Possible policy responses to `Dutch disease' include removing import restrictions, investing aid to enhance productivity in tradeable sector, and accumulatingforeign reserves. FigureAl. REERinTanzania T m a F f E R g8 z z g z g s $ s g g 8 $ 8 8 8 " w m m 8 YearS The period 1993-2000 shows around 35 percent appreciation. Figure A2 plots the decomposition of the REER into two components: (i) nominal effective exchange rate the (NEER) which is a weighted average of bilateral nominal exchange rates with trading partners, and (ii)an index o f relative effective consumer prices, which is a weighted average of consumer prices in trading partner countries relative to Tanzania. This decomposition shows that REER appreciation duringthe 1993-2000 period occurred due to a greater increase in Tanzania's consumer prices compared to those in the major trading partner countries; the average annual inflation was about 15 points higher in Tanzania compared with the main trading partners during this period. NEER actually depreciated during this period; absent such nominal depreciation, REER would have appreciated by more than 100percent. 106 Figure A2. REERand itsDecomposition The analysis o f Tanzania's trade flows shows imbalances that need to be taken into account in the analysis o f exchange rate developments. For example, Tanzania's exports to Kenya and South Africa are negligible compared to imports. Therefore, REER for exports and imports were also examined. Separate REER for exports and imports show patterns similar to that for overall trade, except that the REER for imports appreciated more in the past 3-4 years. This appears to be mainly due to significant nominal depreciation o f the South African currency in the second half o f the nineties. Bilateral real exchange rates provide a measure o f export competitiveness vis-a-vis individual trade partners. Figure A3 depicts Tanzania's bilateral real exchange rates with selected major trade partners during the nineties. The figure shows a loss o f competitiveness with all the major trading partners during the second half o f the nineties. The maximum appreciation i s against South Africa, Japan and Germany, while that against Kenya has beenlimited. Figure A3. Tanzania's RERwith major trading partners Bilateral RER with Major Trading Partners -8 140 130 -t- India d 120 E - 110 100 Y x `0 a, 80 70 2a: 60 +South Africa 50 40 -0- United The above discussion o f competitiveness relies upon an external (PPP based) measure o f the REER. While it is useful for analyzing nominal shocks such as movements in nominal exchange rates and relative inflation rates, an internal measure o f the real exchange rate between tradables and nontradables i s better suited for analyzing the impact o f real shocks such as aid transfers (which can be seen as exogenous increases in income) or shocks to terms o f trade. Towards this end, we have calculated a three-good measure o f internal RER in terms o f composite exports, imports, and a nontraded good. 107 Figure A4 shows the evolution o f these rates over the nineties. It may be seen that these internal rates display trends similar to the external REER. Since 1993, the RER for both exports and imports has appreciated by more than 40 percent. FigureA4. RealExchangeRate(Internal) for Importsand Exports R e a l E x c h a n g e R a t e s (Internal) for I m p o r t s a n d E x p o r t s Y e a r +Exports RER +Imports RER As discussed in Box 1, real exchange rate appreciation has implications for the structure of output inthe economy. Table A1 has some o f the relevant macroeconomic variables for the nineties. These variables exhibit some o f the trends expected to accompany real appreciation. The share o f non tradable output in the economy has been growing and output o f tradables has been stagnant. However, we do not see any trend o f shrinking output o f tradables as indicated by a steady per capita level. Exports have been dwindling, despite buoyant tourism revenue, and the fall in manufactured exports is particularly dramatic. Figure A5 shows that the share o f manufactured goods has been going down inexports. TableA5. Tradables'" output and ExportsinTanzania Year Tradables Tradables Exports (% Export of output (% output per of GDP) Manufactures (% of GDP) capita of GDP) (constant 1992) 1990 51 53 12.6 1.7 1991 53 53 10.3 1.4 1992 53 53 12.4 1.4 1993 53 54 18.0 1.2 1994 50 54 20.6 1.7 1995 51 55 20.7 2.1 1996 52 54 17.6 1.9 1997 50 55 16.6 1.4 1998 49 55 13.6 0.4 1999 50 55 13.8 0.2 2000 50 55 14.6 0.4 2001 50 55 15.6 0.6 Source: WorldBank data. Tradablesinclude agriculture, mining * andmanufacturingoutput. 108 FigureA5. Share of Maufacturesin I PO I \ 0 ' 1 9 ' 1 9 ' 1 9 ' 1 9 ' 1 9 ' 1 9 ' 1 9 `19'19 `19 `20 `201 90 91 92 93 94 95 96 97 98 99 00 01 Conclusions on macroeconomicimplications The above analysis provides preliminary evidence that the second half o f the nineties saw an appreciating real exchange rate - both in external as well as internal terms, despite a depreciating nominal exchange rate. The appreciation was accompanied by structural changes such as falling exports, particularly o f manufacturedgoods, and a declining share o f tradables in national output. These developments could be consistent with "Dutch disease" effects o f increased levels and the changing composition o f aid. However, the analysis does not address the role o f various other `fundamentals' such as productivity changes, or changes inthe terms o ftrade, inappreciationo f the real exchange rate. The real exchange rate movements discussed above could well be equilibrating changes Le., the real exchange rate might be moving towards a new equilibriumlevel. However, declining level o f exports, particularly o f manufactures, would be a source o f concern. Furtherliberalization o f imports is apossible policy option that is likely to help exports and that would also take pressure off the exchange rate. Other possible policy responses mustfollow a more careful anddetailed analysis. 109 e sEeEa C e5E R C E R dPRb c 53 E 0 3 3 fC z c $R L c c i Eb -- II c c 3 3 3 d E b U sV U 9 cz m P rr 0 * Y cd c-) m $1 3 5a, U 0 23 c.' 0 S 0 C *" 4- 0 m 0 E : iEnE 5i c b c0 eCn 5C; 9 c 1rec E -E L c If! 3 00 3 .-0Y g 5 ru 0 .-mY C 0 .--0 w C 0 d E E E ztE2 d C cE f E d C C R C i C c C zEQE Q -- c 3 ANNEX D: SUMMARY PRIORITY SECTORPERREPORTSFISCALYEAR2002/03 OF 124 Listof Acronyms AIDS Acquired Immuno-deficiency Syndrome ASDS Agricultural Sector Development Strategy ASLMs Agricultural Sector Lead Ministries CAG Controller and Auditor General's COBET Complementary Basic Education DADPs Districts Agricultural Development Plans DDH Designated District Hospital DRE Discretionary Recurrent Expenditure DROMAS District Roads Management System EFA EducationFor All ESDP Education Sector Development Programme ESRF Economic and Social Research Foundation FY Fiscal Year FYOl 1July 2000 to 30 June 2001 FY02 1July 2001 to 30 June 2002 FY03 1July 2002 to 30 June 2003 GER Gross Enrolment Rate GOT Government o f Tanzania HIV HumanImmunodeficiencyVirus HTEMP Higherand Technical EducationMaster Plan HBS Household Budget Survey IMR Infant Mortality Rate IFMS IntegratedFinancial Management System IRP IntegratedRoad Project IWRM IntegratedWater Resources Management LA Local Government Authority MTEF Medium Term Expenditure Framework MTES Medium-Term Expenditure Strategy MoEC MinistryofEducationand Culture MoF Ministry o fFinance M O H Ministry o f Health M O W Ministry of Works MoWLD Ministry of Water and Livestock Development MAFS Agriculture and Food Security M C M Ministry o f Cooperatives and Marketing MWLD Ministryof Water and Livestock Development WSDP National Water Sector Development Plan NWP National Water Policy NER Net Enrolment Rate NFE Non-formal Education NMSF oc National Multi-sectoral Strategic Framework Other Charges PE Perrsonal Emolument PRS Poverty Reduction Strategy 125 PRSP Poverty Reduction StrategyPaper PEDP Primary Education Development Programme PO-RALG President's Office -Regional and Local Government R D S RuralDevelopment Strategy RFB Road FundBoard RMMS Roads Maintenance Management System TACAIDS Tanzania Commission for AIDS TANROADS TanzaniaRoadsAgency Tsh Tanzanian Shilling U5MR Under-5 Mortality Rate URT UnitedRepublic of Tanzania UWSAs Districts Urban Water Supply Authorities VET Vocational Education and Training EWIRA Water Utilities Regulatory Authority 126 ListofTables and Charts Table 1: Public EducationExpenditure, 1999/00-2002/03(T.Shs. Billion, Current Prices) Table 2: Comparative Analysis o f ResourceRequirementsandAllocations inPEDP (T.Shs. billion) Table 3: Share of recurrent expenditure, 1999/00to 2002103 (T.Shs. Million, Current Prices) Table 4: Sub-sectoralEducationDevelopmentExpenditure(T.Shs. Million) Table 5: PRS Targets andProgressinMeetingthe Targets Table 6: Total public health expenditureinTanzania(Billion shillings) Table 7: Public health spending, by funding type (Billion shillings) Table 8: Recurrentversus development compositionoftotal expenditure Table 9: Spendingby Type ofActivity, Tshbillion Table 10: Spendingby MOH departmenton drugs and supplies Table 11: Government subventions to LAs, FYOl-FY03 (Billion shillings) Table 12: Spendingon PRSpriority items, FYOOto FY03 (Tsh bn) Table 13: Finance-relatedhealth sector performanceindicators Table 14: Sources of funding o f HIV/AIDS programmesinthe GOT Table 15: Financingof HIV/AIDS Programmesinthe GOT accordingto Sector (Tsh) Table 16: FinancingHIV/AIDS: What Priority Ministries vs. IFMSReports (Tsh) Table 17: ExpenditureinWater Sector (Tsh) Table 18: AggregateResourceRequirementinMedium Term Plan(2003/04 -2005/06) Table 19: Roadnetwork length and conditionby category, 2002 Table 20: Government allocation o fresourcesto the roadsector (inmillionTsh) Table 21: Allocation to PO-RALG Table 22: Plannedvs Actual RoadMaintenance Table 23: PlannedImprovementsinRoadNetwork Conditions Table 24: FoodCrops Production 1998/1999-2000/01to 2002/2003 Table 25: CashCrops Production 1998/99-2000/01to 2002/03 * Table 26: Total Expenditure(Local andForeign-Funded)inMill Tshs Table 27: PlannedVersus PRSP ProposedExpenditure(Tshs Mill) Table 28: PlannedVersus Actual Expenditureinthe Development Budget inMillTshs Table 29a: Trends inFundingPriority Activities inthe MAFS Table 29b: Trends inFundingPriority Activities inMWLD Table 29c: Trends inFunding Priority Activities inthe MCM Table 30: Summary for Sub-programmePriority Costing2003/04-2006/07 (inT.Shs.) Figure 1: Distribution of Total UnspentHIV/AIDS Budget Allowance FY02 (Percent) Figure2: Donor Funds-MOW Roads 127 128 PRIORITY SECTOR PER REPORTS OF FISCAL YEAR 2002/03 This chapter summarizes this year's Public Expenditure Reviews (PERs) o f priority sectors: namely, Education, Health, HIV/AIDS, Water, Roads, and Agriculture. The main sector reports were prepared by the concerned ministries with support from consultants and inputs from the sector working groups during September 2002 to March 2003. Generally, the objective o f a sector report i s to help the budgetlMedium Term Expenditure Framework (MTEF) process for the upcoming fiscal year take informed decisions, making public expenditures more effective in attaining PRS and national development objectives. The sector PERs assess the sectoral budgetary performance in terms o f resource mobilization, allocation, and utilization, recommend improvements, and identify and provide costing o f priority interventions for the upcoming fiscal years. Inaddition, each Sector PER might contain particulartopic(s) considered by its Working Group to provide new and useful information. The following summaries preserve, to the extent possible, the content and follow instyle the main sector PERs. Education Introduction The education sector PER for fiscal year 2002/03 (FY03) builds upon the previous updates. It primarily reviews recent developments, tracks expenditure trends, discusses issues related to financial management and accountability, reviews sector plans and policies in relation to the achievement o f targets, especially those related to Primary Education Development Programme (PEDP), and HIV/AIDS and gender issues. This update also provides costing o f sector interventions and activities to feed into the budget guidelines and the preparation of the MTEF, and makes broad recommendations that can inform policy, enhance the process o f attaining a realistic education budget framework, and strengthenfinancial management and accountability. Recent Developments Since the last PER update o f FY02, progress i s beingmade on a number o f initiatives that will further improve coverage, policy planning and expenditure management o f the education sector. These initiatives include: 0 An ongoing broad review of post primary education and training covering secondary, vocational education and training (VET) and tertiary and higher education sub-sectors that will form the basis for developing strategies and plans for these sub-sectors; 129 The Higher and ~ e c ~ c~ad l~ ~ c aMastern Plan ( T t ~ o ~covering the period ~ ~ ) 2003-2018 has been completed; The ~ ~educat~Q~ ~ WFE) status report~has been~eo pletedr and strate~y ~ ~ a ~ for ~ ~ p ~ e ~ e nis~undero~reparat~onto address key issues related to literacy, t i n n ~ ~ r aandyvocational and life skills e d u ~ a t ~ ~ c foroyouths andadults; and The PEDP ~ n a n ~ imanage~~entand a c c o ~ t ~ nmanual as well as procure~ent a l g ~ a ~ u have been ~ e v g ~ o pand a l e ~ d~str~buted spending units ~ ~ i ~ ~ s t r i e s , to councils and schools). These manuals are expected to i ~ p r o v eprocurement systems, ~ n a n c ireporting, tracking, ~ a ~ i a g e mandtaccounta~~~ity a ~ e ~ . Table Ishows that s p e ~ d i ~ong educat~on(both r e c ~ ~ e n t k~reaseina ~ s and ~ in~real terms. ~ d u c a t i Q ~ ~ ~ ~ e spending as a percent of GDP has risen from 3.1 percent in 1~99/00to 4.4 percent in 2002/03. cent of total expei~d~ture p r ~ osectors i s a~locatedto educat~onsub- on r ~ ~ thin the educat~o~sector, priority is given to p ~ ~ ~e~da crayt~~n, ~ which continuesto get over 42 percent ofthe total resources. Source; URT, Poverty Reduction Strategy Progress Report2001102,November 2002, pg. 106 & 107. t ex re. ~ e c ~ ~spend~ngappears to be g a ~ n i n ~ e n t over ~ e v e ~ spendi~~,~ ~from abou~72.9 percent of total resources in 1~99/00to 81 o ~ ~ er%ng percent in 2002/03. ~~~e there are data p r o ~ ~ ethats may have caused an ~ u ~ i ~ ~ r e s t of~FU03~ oexternal s ~ ~ thereQ is~ an, a ~ ~ a r e nrise in recui~ent ~ a t ~ ~ t e x ~ ~ ~ i d ~ ~ ~ r e s . Though s p e ~ ~ has n ~d ~ amed PEDP resource r e ~ u i r e ~ e n as outline t s of July 2001, ~~~~ guidelines for 2 0 ~ 2 / ~ ~ - s ~ ~ ~ d ~ cintTable 2 a e d below, the d ~ v ~ a t ~bo n s and between PEDP and budget a~~ocat~onsover 40 percent and 50 percent, respectively. The deviations are between and actual budgetal~ocat~~nssmaller and s h r ~ ~ i ~ g . are ~~~~ The other priority sectors are: health, water, roads, a ~ i c ~ ~ judiciary,,and ~ ~ ~ / A ~ ~ S . t ~ ~ r e 130 Budget? 141.31 150.21 171.3 YODeviation4:MTEF vs. PEDP -43.91 -43.81 -49.5 %Deviations: Budget vs. PEDP -52.8 -53.5 -53 .0 96 Deviation:Budget vs. MTEF -15.9 -17.2 -7.8 ~ Q Computed by authors fiom Ministry o f Finance Budget documents (Various, Vote 46, 68 and ~ ~ ~ ~ : Volume 111Supply Votes for councils relatedto grants andsubventions). PRSProgressReport, Nov. 2002. it. ~ ~ u c a et x~poe~n ~ ~ tat~ ethe district level is concerned mainly with the ~ r o v ~ofspr~maryei ~ u ~ a ~~~r ~ v~ ~i ,sofi oe n u c a ~ati ~this ~ ~ ~ o d ~ level is d~ce~tral~zed i s largely the res~ons~b~litythe Local ~ o v e ~ e ~Thes . and of t ce~itral-local~ o v e r ~ split interms ofbudgeted resources ~ e x c ~ u dcapitat~o~grants e n ~ ~ n g and other d~norfunds) is about 15:85. For example, during FU02, a total of T.Shs. 1 5 ~ ~ io^ was. b~~ ~ ~ eoft which, 91.5 percent or T.Shs. ~ ~ 7 , 9~~~~~o~was ~ ~ 0 e ~ ~ 0 . ~ 131 under local g o v e r ~ ~ e responsib~lity26.During FYO1 and F Y Olocal g o v e r ~ e n t s n t ~ received more transfers for educat~on~ ~ ~ ~ than e~ ns~ t i a budgeted. ~ d ~ c a t i o n o s l ~ y transfers lo local g o ~ e ~ ~ ~are e nat s ~1.5 perce~ltto~2 p et~ c ~onftCDP. ~ ~ ~ ~ M Computedby the authors based on ~ ~ E ~ : i ~of Finance, Public Expenditure Supply Vales, Yofume ~ s t ~ 4,Jury 2002-.h?e2003. PRSProgressReport, November2002. ed to the gducat~onsector in relation to total devg~opment resources has been, on average, 12.4 percent for the past three years. Foreign financ~ng continue to d o ~ ~ n athe f i n ~ n co~f deve~opme~i~e~pen~iture,c o ~ ~ p95r percent~ of ~ t e ~ i ~ ~ s ~ the budget estimates in FV03. In teiins of sub-sectoral a~~ocation,primary education c ~ n ~ ~ ntoereceive the highest~ u s ~ ~(Table r4), ~being a~~ocatedover 74 percent o f the o t ~ deve~opinentexpendit~einFU03. r e ala a v a ~ ~ a bfrom the Treasury "AID,flows ~ ~ e ~ ~xterna~ ~ ~ ~ ~ ~ Finance ~ e p ~ ~ ~indicates tthat in FV02 dmiors were c o ~ m to~ provide~T.Shs. e i ~ , ~ e 106.4 b i ~ ~ i tonthe education sector. Actual ~ i s ~ u r s e ~ eweres T.Shs. 100.9 billion or o n t 94.8 percent o f the co itment which cIearIy demonstrates donors' resolute c o n ~ ~ ~ i t i n. e n t As ~ i s c ~ s inethe ~ o ~ l ~ w ~ n gmost of the Poverty ~ ~ d u Strategy~ (oF ~~ S ~ s ~ section, c ~ targets for FU03 have been met or are expected to be met before the end o f the fiscal year. Ach~ev~mento f these PRS targets is inpart due to the availabil~ty~ f f ~ dsuch s , as c a p ~ t a tand~deve1opme~tgrants, and c o ~ i ~ n ~on nthet part o f ~ e v e l o p ~ epartners i ~ e n t andthe ~ o v e ~ ~toeenx pt e d ~ t ~ o ~u s ~ ~ ~ ~the PE P~strategy andplans.~ ~ ~ e ~ The capitat~o~grant in support o f P~~~ i s set at US$ 10 per capita. Duringthe first year, schools with approved school plans would receive US$ 6 (with US$ 4 for textbooks r e t a ~ ~ eatdthe centre), while s c ~ ~ that s o ~ have no such plans would receive US$ 3, the '`URT,PubiicExpendlure Szppfy Voles (Regiod ,Appendices to VotumeIliT Estimates. 2001/02, Pg. ix 132 other US$ 3 being temporarily managed by the districts. As of January 2003, only 53 percent of schools had completed short-term or longterm plans for their schools. With regard to development grants, a total of T.Shs. 43,407 and 1,130 million was disbursedand received inthe districts for the construction of classrooms and pit latrines, respectively. (See achievements inBox 1). A total of T.Shs. 11,687.2 million was disbursed to the regions for the procurement of textbooks and other educational materials. Of this total, only T.Shs. 5,049.5 million or 43.2 percent of the funds were transferred to schools. The remaining funds were to be usedcentrally at the district levelto also procure textbooks. Review of Existing Sector Plans, Strategies,and Performance Current Statusof EducationSector PlansandStrategies Background. There are a number of sector plans, reforms, and strategies in place, including those in Education Sector Development Programme (ESDP), Primary Education Development Programme (PEDP), Secondary Education Development Plan (SEDP), Higher and Technical Education Master Plan (HTEMP), Teacher Education Master Plan, and Education For All (EFA) country plan. O f all the sub-sectors, primary education is the only sub-sector which has fully implementable plans and strategies. The secondary, teacher education and higher and technical education sub-sectors are in various stages of developing their plans and strategies. The PEPD is the first output of the ESDP and addresses access and quality issues inboth the formal and non-formal primary education system. PEDP's priority investmentareas are: (i)Enrolment expansion - focusing on classroom construction, teacher recruitment and teacher deployment; (ii)Quality improvement - encompassing in-service and pre- service teacher training, and teaching and learning materials provision; and (iii) Management improvements - through a range o f capacity building efforts. According to PO-RALG, year one of PEDP (FY02) implementation has shown a significant progress (Box 1). 133 ......................................................... .......................... ..... ............................................................................................................. 1: The Statusof the ImplementationofPEDP ..........,........ ......................................................................................................................................................................... Component ................................................................................................................................................................................................................... i Strategies Performance 1.Enrolment 1 I........................................................................................................................................................................................... .i Classroomconstruction 8,817 new classrooms against 13,868 ~ Expansion i j (more than 7,000 classrooms were at ' j /............................................................................................................................................................... !.......................................................................................................................................................... 1 ...................................................................................................................................................................................... j different stages of construction) i. Teacher recruitment ................................................................................. 7,277 new teachers against9,100 ................................................................................................................ ExpandCOBET programme / 0 NFEStrategybeing prepared I i! Centers 11,564 children enrolled in COBET i i................................................................................................................................................................. i ...................................................................................................................................................... Construction ofteachers houses 0 3,613 teachers houses completed against target of 8,000 houses per i j:................................................................................................................................................................ j annum or 45% ofthe target Increaseenrolment for both boys and 0 1,659,487 have been enrolled against girls I! 4 ......................................................................................................................................................................................... ....................................................................... the target of 1.5 million i 4 2.Quality j................................................................................................................................................................... II Upgradeteachers qualification I............................................................................................................................................................................................. 0 12,414 went through in-service Improvement j.................................................................................................................................................................... j.......................................................................................................................................................................................... 1 : training Increase textbook supply and other 1 0 TShs. 5,049,542,467.57 were made , teaching learning materials 3.................................................................... I1....................................................................................................................................................................................... available for purchase of textbooks, actualamount spent Tsh 3.208billion . .Capacity Conductpre-service training and 6919 Teachers trained in year one ~ I Building I................................................................................................................................................................... 1,1I + provide training for headteachers, against target of 9,000. , II school committees, ward education Preparationsare ongoing to commence coordinators, inspectors and LAs, capacity building at local, regional and regional andnational staff. I national levels. Source: URT, 2001 (PEDP), URT, 2002 (MoEC budget speech), and PO-RALG, 2002 (Annual Performance report, 2001 2002). - For year two of PEDP (2003), the targets were: (i)enrolment of 1,600,000 standard I pupils; (ii)recruitment of 11,65 1 teachers; (iii) training of 10,000 new teachers; (iv) construction of 13,396 classrooms; and (v) continue expansion of COBET (towards a target of 11,323 centres by the year 2006). So far, no progress reports have been made available on this initiative. PerformanceIndicators and Progresstoward Meeting PRS Targets The 2000/01 Household Budget Survey (HBS) suggests improvement towards meeting these targets and the PO-RALG progress report on the implementation of PEDP also shows that some of the targets (Gross Enrolment Rate (GER) and Net Enrolment Rate (NER)) have already been met and in some cases even surpassed (Table 5). For the primary education sub-sector commendable achievementshave beenattained. Only a few targets have shown little or poor progress, including raising the standard VI1 pass mark, andreducing the dropout rate. Quality improvements are yet to be attained. 134 Table 5: PRS Targets and Progressin Meeting the Targets Gross Enrolment Rate INet Enrolment Rate I 70 Girls:60.8 91.04 Boys: 56.7 Transition Rate (primary - secondary) 21 Not shown 19.5 Drop-out Rate 6.6 Not shown Insignificant change but exact percentagenot known Standard VI1 Examination Pass Rate 50 Not shown 27 Literacy Rate 74.8 70 Source: URT (2000), URT (2002) and PO-RALG(2002). CostProjections Cost projections focused on expanding secondary education (to better link primary and secondary education more fully) and teacher requirements. They indicate that resource requirements for secondary education will be nearly twice as much as currently planned for PEDP. However, the PER team recommends making cost projections only after the completion o f the on-going secondary education status study. Expenditure Management andAccountability The PER team observed several improvements and weaknesses in expenditure management and accountability, which can be summarized as follows: 0 The Controller andAuditor General's (CAG) report observed some improvements in resource use and accountability in both Vote 46 (Ministry o f Education) and Vote 68 (Ministry o f Science, Technology and Higher Education) and issued a clean certificate in2001/02. 0 Both the Annual PEDP Performance (2001-2002) Monitoring Report and Stock Taking Report (June 2002) observed improvements inresource use at the council and school levels. 0 Weaknesses infinancial management observed by the C A G relate to irregularities in voucher entries, and failure to audit schools and institutions which received grants. 0 Weaknesses at the Local Government levels relate to shortages o f qualified accountants and auditors, inability to follow financial regulations, weak control procedures, and use o f resources without proper documentation. Financial management manuals have been prepared and may help to ameliorate this problem inthe future. 135 TheImpact of HIV/AIDS The negative impacts o f HIV/AIDS poses a serious threat to the education sector. Anecdotal evidence suggests that the number o f HIV/AIDS cases among teachers i s i n c r e a ~ i n g .The school age population has experienced the immediate impact o f the ~ ~ increasing HIV/AIDS transmission rates through the death o f parents or guardians, and sick children who might face social stigma and discrimination. The consequences are erratic school attendance and therefore lower achievement for those that are affected. Inview of HIV/AIDS transmission and its impact on the education sector, MoEC put in place several preventive intervention programs, including HIV/AIDS awareness and counseling among children and teachers. The approved budget for the program in FY03 i s Tsh 587.7 million. Main Recommendations There are many recommendations inthe mainreport, including among them: 0 Allocationto Councils:The criteria and basis for allocation o f OC, development grants and capitation grants should be clearly debated, discussed, and disseminated so as to be transparent to all, including at the community level. The uses o f OC also need to be clarified and if needed made more consistent with policy inPEDP. 0 Capitation grant:(a) The full capitation grant o f $10 per enrolled child needs to be disbursed to the school level. (b) A rigorous recurrent cost estimation is required so as to provide solid grounding for determining the capitation grant amount inthe MTEFhudgetstarting FY 04/05. Management of Resources:Management o f resources needs to be strengthened at all levels, including clarifying roles and responsibilities, modes o f transfer o f funds, and improving transparency, accounting and accountability. 0 Quality improvements:A key way to improve quality is to ensure that quality improvement designs for curriculum review, assessment for examinations, pre- service and in-service training, monitoring and support and textbook guidelines are reflected inthe education budget. 0 Information:Many o fthe points noted above require effective andtimely flow of information, from top to bottom, bottom to top, and horizontally. Particular emphasis needs to be accorded to information access at the community/school level. A comprehensive communication program that can do this effectively and that can engage all key stakeholders i s yet to be developed andproperly costed, so that it can be included inthe budget. ''An ongoing study by ESRF to assess the socio-economic impact o f HIV/AIDS inTanzaniais expected to provide some quantitative information on this. 136 Health Introduction This year's health sector PER update (FY03) is the third consecutive PERundertaken and builds up on previous ones.** The objective is review and analyze recent overall and subsectoral trends in health sector income and expenditure, briefly explore spending at the Local Government Authority (LA) level, review sectoral performance in relation to stated PRS objectives mainly on primary health, and make recommendations and suggest next possible steps for strengthening the current budgetary review process and for future work. Recent Trends in Health Sector Expenditure The Overall TrendinExpenditure The health sector, as a priority sector within the Poverty Reduction Strategy, has benefited from increases in both the absolute level o f government funding, and in its share o f the budget. Table 6 below illustrates that the nominal spending on the health sector in Tanzania has risen from Tsh 132 billion in FYOO to Tsh 215 billion in FY02, with the allocated budget for FY03 showing Tsh 243 billion. This is largely due to an increase in recurrent spending of 32 percent, a substantial share o f which i s likely to be due to the increase inthe donor basket fund for the health sector which rose by over 80 percent between FY02 and FY03. Table 6: Total public healthexpenditureinTanzania (3illion shillings) r 1998/99 1999/2000 2000/2001 2001/2002 2002/2003 Budget Actual Budget Actual Budget Actual Budget Actual Budget Recurrent MOH 37.25 37.15 39.20 32.39 49.39 44.25 61.60 58.99 86.94 Region 9.25 8.68 9.36 9.01 6.21 5.61 7.06 6.58 7.60 Local Govt 15.72 16.34 18.69 17.95 36.35 35.67 46.26 46.28 57.66 Total rec. 62.21 62.18 67.25 59.34 91.95 85.53 114.92 111.86 152.20 Development MOH 21.21 17.27 17.75 10.19 20.47 14.84 32.07 21.12 33.78 Regions 5.00 0.67 2.57 0.79 4.62 1.39 2.35 1.28 4.75 Local Goa 0.62 1.18 1.06 1.73 1.52 1.70 2.04 I I I I I Total devt 26.83 17.94I 21.50 12.03I 26.81 17.74I 36.12 22.40I 40.57 Off budget expenditure Cost sharing 1.09 - 1.49 - 1.86 - 1.37 1.20 Other foreian funds 35.55 42.76 52.33 60.04 59.41 75.00 66.14 79.37 49.25 Total off buiget 35.55 43.85 52.33 61.53 59.41 76.86 66.14 80.74 50.45 Grand total 124.58 123.96II 141.08 132.91II 178.18 180.131I 217.18 215.01II 243.23 Note: basket f'unding included under recurrent or development as appropriate As a proportion o f the total government budget, the expenditure allocation to the health sector has increased slightly over the past few years since the publication of the PRS, from 7.5 percent in FYOO to an estimated 8.7 percent in FY03, only 0.2 percent of this 28The FY03 Public Expenditure Review health sector was undertakenby a Task Team fi-om the Ministry of Health (MoH), with support fi-om other colleagues inthe Ministry,officials from the President'sOffice - Planning and Privatisation Commission, President's Office -Regional and Local Government (PORALG), andthe Ministry of Finance. An external consultantprovided additional support to the Task Team. 137 occurring in the last two years. In addition to moving only slowly in the desired direction, these figures are still relatively low given the target o f 15 percent o f government budget for this sector, as was agreed inAbuja in2001. The Sub-sectoralTrends Spending by level of government: From FY99 to FYO1, there has been a trend o f increasing financial subventions to Local Authorities (LAs) for health spending. This trend has since been reversed, with a fall in budget estimates from 32.1 percent to 31 percent o f total on-budget estimates between FYOl and FY03, and from 36 percent to 34.5 percent in actual expenditures between FYOl and FY02 (Table 6). This i s contrary to the stated PRS objectives o f getting a greater share o f funds to the operational level. It should be noted however, that transfers inkindcontinue to grow. Funding: The share o f domestic funding in actual expenditure has declined from 47 percent o f the total expenditure in FYOO to 44 percent in FY02 (Table 7). For FY03, domestic funding o f the budget i s about 47 percent. Interms o f the allocated budget, the share o f domestic funding increased from 46 percent inFYOOto 48 percent inFY03. Table 7: Public healthspending, by funding type (Billionshillings) I I 1998199 I I99912000 I 2000/01 I 2001102 ~~ ~~ ~~~ ~~ I 2002103 1 Budget Actual Budget actual budget actual budget actual budget Recurrent Govt funds 62.21 62.18 60.73 57.98 75.47 74.90 91.63 90.29 109.39 Foreignfunds 6.52 1.36 16.48 10.63 23.29 21.57 42.81 Total 62.21 62.18 67.25 59.34 91.95 85.53 114.92 111.86 152.20 Development Govt funds 2.60 0.92 4.56 2.80 5.14 5.13 5.34 3.59 6.03 Foreignfunds 24.23 17.01 16.94 9.24 21.67 12.61 30.79 18.82 34.54 Total 26.83 17.94 21.50 12.03 26.81 17.74 36.12 22.40 40.57 Total budget 89.04 80.11 88.75 71.38 118.76 103.27 151.04 134.26 192.77 Off budaet - Govt funds 1.09 1.49 1.86 1.37 1.20 Foreignfunds 35.55 42.76 52.33 60.04 59.41 75.00 66.14 79.37 49.25 Total 35.55 43.85 52.33 61.53 59.41 76.86 66.14 80.74 50.45 Grand total 124.58 123.96 141.08 132.91 178.18 180.13 217.18 215.01 243.23 Budget Actual FY99 FYOO FYOI FYOZ FY03 FY99 FYOO FYOl FY02 Recurrent 70% 76% 77% 76% 79% 78% 83% 83% 83% Development I 30% 24% 23% 24% 21%1 22% 17% 17% 17%1 Recurrent versus development expenditures: The share o f the recurrent budget has continued to increase as a percentage o f the on-budget total, reaching 79 percent o f the FY03 budget, and 83 percent o f the FY02 expenditure (Table 8). This is largely due to the increasing share o fthe donor basket, and the fact that the majority o fbasket spending i s accounted for through the recurrent budget. Spending by type of activity: Regarding the trend in the allocation o f spending to Preventive, Hospital, or Administration, the largest percent increase was seen in Preventive spending in FY02, reaching 37.2 percent, and compared to growth o f 29 percent in spending on central level administration, and 1.9 percent on Hospital services (Table 9). This i s encouraging interms of the intra-sectoral allocation o f resources. 138 Table 9: Spendingby Type of Activity, Tsh billion FYOO FYOI FY02 PE oc Total PE oc Total PE oc Total MOH Admln, NlMR and TFNC 2.59 1.26 3.86 2.90 3.69 6.58 3.25 5.24 8.50 Hospitals MuhimbiliMedical Centre 4.93 1.36 6.29 3.79 1.20 4.99 4.78 1.72 6.51 MuhimbiliOrthopaedic Institute 0.44 0.26 0.69 0.54 0.30 0.85 0.56 0.38 0.94 Ocean Road Cancer Institute 0.26 0.15 0.42 0.28 0.23 0.51 0.29 0.43 0.72 Bugando Medical Centre 0.54 0.32 0.87 0.71 0.86 1.57 0.72 0.41 1.13 KilimanjaroChristian Medical Centre 1.02 0.61 1.63 1.20 0.83 2.03 1.33 0.71 2.04 Referralhospitals, MoH 2.82 2.49 5.32 1.74 4.06 5.81 2.03 0.32 2.35 Regionalhospitals 8.10 1.70 9.88 4.43 2.07 6.50 5.14 2.40 7.54 District hospitals 2.87 2.75 5.62 3.53 5.06 8.59 5.22 4.39 9.62 DesignatedDistrict Hospitals 1.91 0.89 2.79 3.07 1.27 4.33 2.94 2.23 5.16 VoluntarvAaencies - HosDital 1.03 0.08 1.11 1.91 0.23 2.14 1.94 0.09 2.03 Total hospitals 23.93 10.69 34.62 21.20 16.11 37.31 22.93 13.07 38.03 PreventivelPrimary health care I MoH preventiveservices 0.25 1.53 1.78 0.27 3.55 3.82 0.27 6.01 6.28 Regionalpreventive services 0.25 0.09 0.34 0.13 0.06 0.21 0.25 0.05 0.30 Council preventive 11.39 6.00 17.39 14.11 12.86 26.97 18.50 17.45 35.95 Total PreventivelPrimary 11.89 7.62 19.51 14.51 16.49 31.00 19.02 23.51 42.53 Total Health recurrent 57.98 74.90 89.06 On- versus off-budget expenditure:The share of on-budget spending has continued to increase, largely as a result o f the concerted effort by M O H and MoF to improve information on external funding to the sector and to capture a higher proportion of the total within budgetary estimates, and through improvements in the planning and budgeting process as the MTEF becomes a more familiar and applied tool. On-budget estimates rose dramatically from 69.5 percent to 79.3 percent o f the FY03 budget, while interms o f actual expenditure the figure rose from 57.3 percent to 62.4 percent between FYOl and FY02. Analysis of Spending Recurrentspending:Recurrent budget allocations, releases and expenditures continue to improve at the central level, both in terms o f GOT funding and the basket, with 98 percent release o f GOT funding and 83 percent o f basket. Interms o f actual expenditure, basket funds represents a substantial improvement on the 47 percent figure o f FYOl. over 99 percent of GOT releases were spent, while the figure o f 86.6 percent for the Recurrent spending on drugs and supplies: There has been a significant increase in recurrent spending on drugs and supplies, one o f the areas targeted for expenditure inthe context o f the PRS. The increase has been slightly higher interms o f spending on drugs and supplies for the Local Government priority level (36 percent). However, as Table 10 illustrates, there was a significant difference between the increase in government spending (7.6 percent) and spendingthrough the central basket (83 percent), which needs to be carefully considered duringthe budgetpreparation for FY04 given the uncertainties regardingthe future o f the central level basket. The majority o f hospital drug allocations in FY02 (57 percent) went to facilities at the district level. There are large per capita variations in hospital drug allocation by region which would benefit from further exploration. 139 FY02 FY03 GOT Basket GOT Basket 2001 Curative/Hospital services 11,270,002,202 2,905,668,252 11,000,000,000 8,208,070,000 2002 Chief Government Chemist 2,600,000 2,600,000 33,000,000 3001 Preventive services 1,614,111,333 4,038,815,780 2,800,654,600 4,492,385,000 4001 Tukuta 30.514.634 . . 580.000 10.415.000 3.200.000 . . 5001 Human Resource Devt I 6;300;000I Recurrent spending on HIV/AIDS: Spending on HIV/AIDS occurs both through the M O H recurrent and development budgets, and through off-budget expenditure both within and outside the sector, and more analysis is required to adequately compile this information. However, a preliminary analysis shows a substantial increase in spending from the central level basket, on both Preventive and Hospital Services. The allocation o f GOT funds to the National AIDS Control Programme was only 3.7 percent in FY03. At the local government level, all LAs had assigned funds for HIV/AIDS plans in their budgets, although at a very low level o f total LA health OC spending, but there has been a substantial rise betweenFY02 and FY03. Developmentspending:There is a significant mismatch betweenthe externally financed development projects in the database o f the MoF and those implicit in the budget. A limited analysis of the MoF External Finance database indicated that less than half (46 percent) o f external financing in FY03 was captured in the budget, and that over 75 percent o f projects in the sector are implemented through the MOH. Basket funding accounted for 40 percent o f the total external funding, while HIV/AIDS,TB and Malaria accounted for only 7.2 percent o f the FY03 allocation. The locally-financed M O H Development budget covers a number o f infrastructure rehabilitation projects, seen as a strategy for achieving PRS targets. Further analysis should be undertaken of the development budget, in order to reconcile spending figures both with the MoF database, andwith PRS targets. Local Government Budgets,Allocations and Spending L A Health Spending Overall leveland share: There has been a slight increase inthe level o fhealthbudgets at LA level, with health rising only from 17.6 percent to 17.7 percent o f total LA budgets between FYOl and FY03, with very little difference between urban and rural councils. Looking only at OCs, the figure for FY03 is 24.4 percent o f total LA OCs, with district councils allocating a slightly higher share than urban councils. These figures disguise wide variations between councils-which would be interestingto explore further. 140 FYOl Amroved estimates FY02 Approved estimates FY03 estimates PE oc Total PE oc Total PE oc Total Urban 4.47 1.57 6.04 4.90 2.16 7.06 6.34 2.52 8.86 District 16.58 6.47 23.05 18.83 9.59 28.42 23.60 11.22 34.81 Total 21-05 8.04 29.09 23.73 11.75 35.48 29.94 13.74 43-68 Governmentsubventionsto LAs: Government subventions to councils for spendingon healthhave fallen slightly as a share o f total on-budget spendingon health. However, the absolute level has risen as shown inTable 11. Releases vs the budget: In general, releases matched budgets for FY02, with only 8 o f the 115 districts receiving a different amount thanbudgeted. No apparent reasons could be found for these variations which would benefit from further exploration. Per capita allocation: The mean per capita allocation (PE and OC) for Tanzania mainland for health spending at LA level was calculated at Tsh 1,288 in FY03, up 19 percent from Tsh 1,079 in FY02. The range varies for FY03 from Tsh786 in Kagera to Tsh2,708 in Coast. There appears to be no correlation between regional poverty levels and LA allocations, total or OC, and the picture appears to have worsened between FY02 and FY03. Fiscal decentralization and allocation formulae Steps were taken in the sector to define a more objective, transparent resource allocation formula for allocating funds across the region and councils, and parallel activities have beenundertaken under the Local Government Reform Programme. It remains to be seen what decisions will be taken in the hture, but standard baseline figures should be developed inorder to monitor progress over time towards a stated equity target. Sector Performance Health performancein relation to PRS. The national health goal o f the PRS is firstly to arrest the decline in life expectancy and thento raise it to 52 years bythe year 2010. By 2003, policies are intendedto: 0 Lower the Infant Mortality Rate (IMR) from 99 to 85 per 1000live births; 0 Reduce Under-5 Mortality Rate (USMR)from 158 to 127per 1000live births; 0 Lower maternal mortality from 529 to 450 per 100,000 births; 0 Reduce malaria-related fatality for under-fives from 12.8 percent to 10 percent; and 0 Raise the proportion o f the rural population with access to safe and clean water from 48.5 percent to 55 percent. Strategies to achieve these objectives include strengthening immunization services, improving the availability o f drugs and supplies, providing quality health services through delivery o f the essential health package, and strengthening and reorienting the 141 delivery o f secondary and tertiary health services to ensure more effective support of primaryhealthcare. It is not possible to report here on the success or otherwise o f the sector in terms of meeting these objectives, and delivering on these strategies. However, it is possible to broadly show the extent to which stated priorities, as articulated in successive budgets, have been met with actual expenditures. The data presented earlier in Table 6 indicate that the sector as a whole has received an increase in both in the nominal level and the share o f the budget since publication o f the PRS. Whether resources went into all priority itemdareas i s less clear. Health sector performance in relation to stated PRS priorities was analyzed for government spending only, and further analysis should be undertaken o f basket and other foreign funding within the total in order to obtain a clearer picture o f the whole sector. For government spending there was a mixed picture. Table 12 shows that the absolute level o f spending on primary health increased by 184 percent between FYOO (actual spending) and FY03 (budget estimate). However, as a share o f government spendingin the sector, the FY03 estimates represent a slight fall from 30.1 percent to 28.2 percent. FYW FYOl FY02 FYo3 Priorityltems-PrimryHealth Vote Subvote Item Actual Actual Actual Estimates OCSubventiontoL4 7&89 + 95 1001 all 2.63 8.W 11.75 13.74 1Adrugsallwtim buclgetedun& MOH 52 2001 260402 5.75 4.62 7.63 10.40 M UpreventiveSubvoteOC 52 3001 all 3.22 5.40 8.82 8.99 FsgionspreventiveSubvoteOCI 7&89 + 95 3002 all 0.09 0.08 0.05 0.12 Total: Primary Health 11.69 18.14 28.26 33.24 Total healthsoendin0 Govt - 60.78 80.03 93.88 115.42 The share o f primary health within discretionary recurrent expenditure (DRE) has increased from 1.7 percent in FYOO to 2.5 percent in FY02.29 The budget for FY03 indicates a slight increase to 2.6 percent. This demonstrates that the priority items within the health sector have absorbed an increasing share o f overall (unadjusted) DRE, albeit at a very low level. 29Defmed as the recurrent budget available for allocation between different sectors or priority areas after the deduction o fstatutory expenditure requirements such as debt service. 142 Table 13: Finance-relatedhealthsector performanceindicators llndicator I Level I Baseline I FY02 7 health activities against the total overall funds available for Financing performance In terms of finance-related performance indicators for the sector, there have been improvements in nominal per capita GOTallocations at the central, regional and district level, in total per capita health spending, and in per capita GOTrecurrent spending by level, with the largest increase in per capita preventive spending at 37.6 percent (Table 13). A slight fall inthe share o f health funds allocated to LA health activities i s noted. Future Costs and Revenues Information on future costs and revenues remains one o f the weaker parts o f the PER. No detailed costing has yet been undertaken. Estimates from four priority programmes indicate that if their requirements were met, they would absorb almost 25 percent o f the total sectoral resource envelope (on and off-budget) for FY03. The picture regarding revenues is gradually improving, particularly with respect to foreign funding, as the database developed by the Ministry o f Finance improves in accuracy and completeness, but there i s still some way to go. Concerns regarding the future level of basket funding for the central level, and the extent to which shortfalls will be offset by general budget support remain for the period beyond FY04. Recommendations and Next Steps A numbero frecommendations andnext steps are outlined inthe main Sector Report: 0 Key among them is the need to better institutionalize the process o f the PER within the M o H in terms of actual commitment and involvement, and reduced dependence on external support. A clear articulation o f the information needs, shared among parties within the sector (MoH, PORALG, MSD etc) and timetable for undertaking smaller pieces o f analysis which would feed into the process could be developed quite simply. 0 Various improvements are required in terms o f information, particularly relating to local government spending, cost-sharing income and expenditure, and arrears within the sector, particularly o f larger institutions. More analysis is requiredof the development budget, and beyondthat o f the information contained within the MoF External Finance database in order to determine the type o f expenditure 143 supported by external finance, and the extent to which spending through the Development Budget meetings PRS objectives. Inorder to better assess the realismoffuture estimates, andto be able to prioritize available resources between activities and levels o f the system, some type o f standard costing should be undertaken, based on agreement as to key interventions, and feasible levels o f coverage. This would also enable better geographical targeting to under-resourced areas, or those with greater healthneed. Various aspects o f the government budget structure and coding need to be reviewed, as has beenpointedout inprevious PERreviews. Agreement shouldbe reached on the most appropriate split for analysis (e.g. preventive curative, or district, referral, central). Planned harmonization o f central and local government expenditure coding and financial years should assist in this process. However, review o f some sub-item codes is required(e.g. HIV/AIDSepidemics, the various categories o f allowances), as i s a better articulation of cost centres across the levels o f the system, rather than heavy reliance on central level transfers to different institutions and geographical areas. HIV/AIDS Introduction The update o f the HIV/AIDS Multi-sectoral Public Expenditure Review reports on the state o f HIV/AIDS financing and e~penditure.~' The principle objective o f the update is to report recent expenditure and budget estimates for HIV/AIDS programmes from a multi-sectoral perspective, assess the feasibility o f improving the tracking o f HIV/AIDS- related financial flows across all sectors, recommend ways to improve the draft narrative section in the National Budget Guidelines for FY2003/04, and assess and recommend ways to build the capacity o f key sectors to formulate effective plans and budgets for their distinctive contributionto the multi-sectoralresponse. Background:The economic impact of HIV/AIDS in Tanzania is severe - through lost productivity, increasing dependency ratio, and escalating cost (to government and households) for the care for infected individuals. Moreover, the impact o f HIV/AIDS strains the GOT'S ability to deliver social services. Therefore a rapid acceleration o f preventive programs to mitigate its impact and care for those affected is a top priority. The importance o fthis priority is clarified inthe mission statement o fthe National Multi- sectoral Strategic Frameworkon HIV/AIDS WMSF) 2003-2007. Translating the policy commitment into action requires the provision o f adequate budgetary resources across a range o f sectors and at different levels o f government. The NMSF identifies the mainstreaming of HIV/AIDS programmes as a key strategic 30The review in this update is mainly limitedto preventiveprogrammesthat are dedicatedto HIV/AIDS, such as the "National AIDS ControlProgramme" inthe Ministry of Health(MoH) and the "Programme to Train Teachers in HIV/AIDS Counselling" in the Ministry of Education and Culture (MoEC). For example, expenditures on treatment, administration, capital formation, medical and support staff are generallyexcluded. 144 objective. Therefore the inclusion o f HIV/AIDS interventions into the work plans and budgetsof all sectors should be a normal part o fthe planningprocess. Expenditure and Budget Estimates Compared to the actual expenditure in FY02, the budget allocation in FY03 o f the financing of HIV/AIDS programmes represents an increase o f almost two-folds (Table 14). Both the GOT and development partners are channeling more resources to HIV/AIDS and are re-doubling their efforts to fight HIV/AIDS in the public sector. Moreover, GOT contribution will boost its share o f total funding for HIV/AIDSprograms from 20 percent inFY02to 33 percent inFY03. Overall financing of HIV/AIDS programmes in the GOTrose sharply (Table 14). The rise i s mainly due to increases in few sectors, most notably TACAIDS. Most sectors did not increase financing in FY 2002/03 considerably beyond expenditure levels in FY 2001/02 (Table 15). This situation indicates that although some progress has been achieved in mainstreaming of HIV/AIDS planning across sectors, there i s considerably more work to be done. Table 14. Sourcesof fundingof HIV/AIDS programmesinthe Governmentof Tanzania 2001/02 (Actual Expenditure) 2002/03 (Budget Allocation) Amount Share Amount Share Source of Funding (Tsh) (%) (Tsh) ("/.I GOT 2,295,562,466 20 7,251,694,571 33 Development Partners 9,109,720,764 80 14,693,112,036 67 Bi-lateral 4,985,146,220 9,834,876,996 Multi-lateral 4,124,574,544 4,858,235,040 Total 11,405,283,230 100 21,944,806,607 100 Sources: MoF IFMS,2002 UNAIDSDonor Mapping Stu&, MoF AID Flow Database TACAIDS has replaced the M o H as the most important govement institution regarding HIV/AIDS financing. The share of total HIV/AIDS financing by TACAIDS increased from 10percent inFY02 to 48 percent inFY03. Meanwhile the share o f total HIV/AIDS financing by the M o Hdecreased from 69 percent inFY02to 35 percent inFY03. 145 Figure 1.Distribution of Total UnspentHIV/AIDS BudgetAllowance FY02 (Percent) TACAIDS MOEC Other Sectors MLYDS MOJ PO-RALG Planning There is concern, however, that TACAIDS does not yet have the capacity to absorb its sizeable budget allowance. Eighty-threepercent of the TACAIDS budgetallowancewas unspentinFY02, and eighty-onepercent remainsunspent, so far, inFY03. TACAIDS i s responsible for about 65 percent of what was not spent of the HIV/AIDS budget allowance inbothFY02(Figure 1) andFY03. 146 Table 16. FinancingHIV/AIDS: What PriorityMinistriesvs. IFMS Reports(Tsh) (Based on limited available information) FY2002/03 - BudgetAllowance Sector IFMS Difference PriorityMinistries Request* Reporting** Amount YO Ministryof EducationandCulture 997,000,000 587,661,32 1 409,338,679 41% Ministryof Works 368,3 13,800 0 368,313,800 100% MinistryofLands 237,000,000 5,000,000 232,000,000 98% MinistryofHealth 3,962,545,000 1,567,676,000 2,394,869,000 60% PO-RALG 23,600,000 23,600,000 0 0% Total 5,588,458,800 2,206,937,321 3,381,521,479 61YO There i s also concern that the Integrated Financial Management System (IFMS) under- reports actual budget and expenditure data on HIV/AIDS. Four o f five priority ministries with available data report an average o f 61 percent higher budget for HIV/AIDS than what i s recorded by the IFMS (Table 16). The discrepancy indicates that a sizeable amount o f expenditure goes un-recorded by the IFMS in many sectors. This gap in reporting may undermine efforts to accurately measure the multi-sectoral response to HIV/AIDS. Tracking Financia1Flows The existing system for tracking HIV/AIDS-related financial flows needs to provide more comprehensive and detailed information in order to be effective in monitoring HIV/AIDSbudgetsand expenditure across sectors. As it i s presently configured, the IFMS provides insufficient information for a complete and detailedanalysis o f HIV/AIDSbudget and expenditure patterns. The lack of detail i s because HIVIAIDS programmes are not coded in the IFMS as an upper-level classification. Instead HIV/AIDS programmes are typically classified at lower levels as itemsor sub-items. Recommendations: 0 To provide more comprehensive and detailed information on HIV/AIDS financial flows thus requires that: (i) the work plansand budgets for HIV/AIDS activities in the sector MTEFs must be linked with the strategic objectives o f the NMSF; and (ii)the coding of HIV/AIDS and key sub-components (e.g. fundingsources, strategic objectives and line items) must be integrated into the classification structure o f the IFMS. It i s recommended that HIV/AIDS be coded at the activity level o f the IFMS classification system. 0 The proposed re-classification of HIV/AIDS in the IFMS could be used to: (i) rapidly obtain detailed information on recent HIV/AIDS expenditure; (ii)link HIV/AIDS activities in the MTEF to the NMSF objectives, thus providing information in the monitoring o f the execution o f the NMSF; (iii)monitor progress in the mainstreaming of HIV/AIDS among sectors (sectors not making progress in mainstreaming HIV/AIDS programs into their work plans could be identified easily); (iv) produce regular and consistent multi-sectoral HIV PER 148 updates, which are mandated annually by the NMSF; and (v) measure progress on Goal 3 of the NMSF: Ensuringthat Political and government leaders consistently give highvisibility to HIV/AIDS intheir proceedings and public appearance. Budget Guidelines The guidelines for the FY 2003/04 budget provide insufficient guidance to sectors in planning and budgetingfor HIV/AIDS programmes for the following reasons: 0 The narrative section makes only a general reference to the importance of HIV/AIDS and remarks that each sector should plan for HIV/AIDS activities in its budget. Moreover, there i s no guidance on what specific activities sectors shouldbe engaged in. 0 The recommended budget ceilings for HIV/AIDSproposed in the guidelines are organized according to priority area, which i s based on an out-dated classification since the introduction o f the NMSF strategic objectives. They are not organized according to sector, which would be useful in indicating an appropriate sector budget. 0 The budget ceilings do not indicate appropriate requirements for HIV/AIDS. The FY 2003/04 Budget Guidelines mentions 42.6 billion T.Shs. and while this represents a large increase on previous years, it i s clear that there remains scope for greater expenditure on this priority area. Recommendation: The National Budget Guidelines should: 0 Provide more detailed guidance on planning and budgeting for HIV/AIDS programmes. First, the guidelines should identify types o f HIV/AIDS activities that are appropriate for each sector. The suggested activities also should be linked to the objectives o f the NMSF. Furthermore the activities should identify targets o f outputs to be produced by each activity as well outcomes to be expected. Second, the guidelines should emphasize the importance of HIV/AIDS programming and that it is to be a normal part o f sector activity. Third, the guidelines should indicate realistic funding requirements and inform sectors ofthe amount o f funds available for HIV/AIDS. 0 Provide the following: (i) rationale for including HIV/AIDS work plans into sector budgets; (ii) comprehensive list o f NMSF objectives and activities to be conducted; (iii) suggested activities for each sector mapped to NMSF objectives including funding requirementsand estimated unit costs; (iv) realistic output and outcome targets for each activity; and (v) a list o f sector HIV/AIDS coordinators and contact information. CapacityAssessment TACAIDS i s concerned that it does not have the capacity to effectively manage its financial responsibilities in FY 2003/04. According to the Tanzania Commission for AIDS Act, 2001, TACAIDS is mandated to: (i) mobilize, disburse and monitor resources and ensure equitable distribution; (ii)develop a strategic framework and national 149 guidelines to support planning, co-ordination and implementation o f the national multi- sectoral response to HIVIAIDS at all levels; and (iii)develop and facilitate implementation o f the national multi-sectoral strategy for mobilization and utilization of resources for HIVIAIDS. The responsibility for executing these duties rests squarely with key departments inTACAIDS. TACAIDS is also responsible for implementation guidance and financial management systems for HIV/AIDS activities, including: mobilizing line ministries and local government authorities to mainstream HIV/AIDS in work plans and budgets; adjusting the IFMS classification to improve the capacity to track HIV/AIDS finance flows; developing a supplement to the National Budget Guidelines that provides detailed guidance to sectors in HIV/AIDS planning and costing; providing one-on-one technical guidance to sectors in HIVIAIDS planning and budgeting; and monitoring the flow o f all funds for HIV/AIDS in Tanzania, including contributions from the GOT, development partners, andNGOs. Presently, to undertake these tasks, the key responsible departments at TACAIDS do not have adequate staffing. It i s clear that additional financial staff will be necessary to meet these responsibilities. Recommendations: 0 Indeveloping its strategic work planfor FY 2003/04, TACAIDS should focus on the resource needs of its Department o f Finance, Administration and Resource Mobilization, which i s responsible for implementing a number o f financial management systems. 0 TACAIDS should foster stronger ties to key institutions involved in the HIV/AIDS planning and financing process. The most crucial ties are with the offices o f the Budget Commissioner inthe MoF and with PO-RALG. Water Introduction This year's update o f the water sector PER isjointly prepared by the Economic Research Bureau o f the University o f Dar-es-Salaam and the Ministry o f Water and Livestock Development (MoWLD). It builds upon the past water sector PERs. The main objective o f the report i s to review sector strategies and policies, assess the sectoral budgetary performance, discuss output performance, identify the medium-term targets and priorities, estimate resource requirements necessary to achieve PRSP targets for the period 2003/04-2OOYO6, and provide recommendations and concluding statements. Sector Strategies and Policies Strategies and Policies InJuly 2002, the revisedNational WaterPolicy was launched. Itfocuses on participatory planning and cost sharing in the construction, operation and maintenance o f community- based domestic water supply systems and cost recovery for urban water supplies. It 150 addresses the relevant issues on Integrated Water Resources Management (IWRM) and adopts a comprehensive policy framework that treats water as an economic good, addresses environmental protection, and seeks institutional reform and greater participation o f stakeholders. The National Water Sector Development Plan (WSDP), which is now at the early stages o f preparation, i s plannedto be completed inDecember 2004. It will become the strategic document for the sector and i s expected to speed up and enhance the implementation o f the National Water Policy (NWP). The National Strategic Action Plan for the Water Supply and Sanitation Sector has been established mainly based on the Tanzania Assistance Strategy, Medium-Term Expenditure Strategy (MTES), Poverty Reduction Strategy Paper (PRSP) and the National Water Policy. The Ministry o f Water and Livestock Development is also preparing a number o f sub- sector strategies through various donor-funded projects, including Urban water Supply and Sewerage Strategy, the Rural Water and Sanitation Strategy, and a similar action plan for Water Resources Development and Coordination. There i s a concern that all these parallel efforts for sub-sector strategies may jeopardize the ongoing preparation o f the WSDP. Reforms. Institutional reforms are underway in the water sector that would increase private sector and community participation. A key component of the reform i s that the role o f government/LA has changed from being direct implementer of water sector programs to that o f a facilitator, regulator and monitor. Accordingly, the Ministry o f Water has re-organized its structures to bring it in line with its new role. However, structures at regional and district levels have not been worked out completely in line with the reforms. Also in line with the sector policies, there i s public divestiture. The privatization o f DAWASA is almost complete with the government preparing for the take over o f its operations by the tender winners Biwater Co (England) and Gauff (Germany). The private operator i s expected to take charge by July 2003. It i s expected that the private operator will soon revamp the water systemand supply service inDar es Salaam. The government intends to strengthen the Energy and Water Utilities Regulatory Authority (EWURA), by institutingpowers for technical regulation. Budget Performance Overalltrend. Table 17below illustratesthat the overall trend inexpenditure is such that the budget for FY03 is more than double o fthe actual inFYOl, There has been a decline in FY02 primarily reflecting the decline in the actual foreign financing of the development budget. 151 Table 17: Expenditure in Water Sector (Tsh) 2000/01 (Actual) 2001/02 (Actual) 2002103 (Budget) 1. RecurrentBudget 4,939,597,508 6,503,403,900 10,151,212,600 PE 1,968,299,218 2,137,411,400 2,835,805,900 oc 2,971,298,290 4,365,992,500 7,3 15,406,700 2. DevelopmentExpenditure 18,854,007,000 14,414,964,000 41,400,687,600 LocallyFinanced 2,334,007,000 2,966,483,000 3,164,256,000 ForeignFinanced 16,520,000,000 11,448,481,000 38,236,43 1,600 3. Total Expenditure 23,793,604,508 20,918,367,900 51,551,900,200 RecurrentExpenditures Recurrent expenditure has doubled over the past three years and the proportion o f OC in the recurrent budget has been increasing from 60.3 percent in FYOl to 72.1 percent in FY03. The steady increase in PE expenditure, in absolute terms, from 1.968 billion Tshs in2000/01 to 2.137 billionTshs in2001/02 and2.835 billionTshs in2002/03 was mainly due to increased sector activities includingmore humanresource requirements. With respect to OC, its composition in the water sector Technical Departments show greater percentages for the leading core sub-sectors (Urban Water Supply and Sewerage, Rural Water Supply and Water Resources Assessment and Exploration). These core sub- sectors absorb about 75 percent-80 percent o f total OC expenditure in the Technical Departments because o f their nature o f activities, which cover the entire country. The main cost components are electricity and chemicals. The urban water supply does not seem to receive a larger share o f the recurrent budget as it usedto be inthe past. The rural water supply instead appears to receive the lion's share o f the recurrent budget, despite the fact that over time the proportions are declining. In 2002/03 the budget allocation for the rural water supply was 22.5 percent o f the total recurrent expenditure lower than 29 percent o f the total recurrent, which the sector was allocated in2000/0 1. DevelopmentExpenditures The decline o f foreign financing o f development expenditure is the outcome o f donor's and government's fulfillment o f their commitments. For this reason, the trend in the development budgetis unpredictable. Unlike the trend revealed under the recurrent expenditure, the development budget figures for urban and rural water sub-sectors show that resources allocated to urbanareas are consistently and significantly higher than the ones allocated for the rural areas. The share o f rural sector inthe development budget was 18 percent, 10 percent and 27 percent 152 during the 2000/01-2002/03 period. This trend does not conform to the PRSP targets and objectives that sought to increase the proportion allocated to rural sector overtime. The share o f the rural sector in the total external donor funding o f development expenditure o f the sector was 13.8 percent, 1.2 percent and 26.3 percent in FYs 2000/01, 2001/02 and 2002/03 respectively, while the foreign budget proportions for urban water supply and sewerage were 65 percent, 98 percent and 72 percent during the three years. Thus, the distribution of development resources between rural and urbansectors, ex post, has beenmore o f a function o f donor preferences. Sanitation. Sanitation services are not covered inthe PERo fthe water sector. Disbursements and utilization. An analysis of the first half of only FY03 shows that there i s some delay in the disbursement o f funds. The funds released as o f January 2003 amount to only 5.5 percent o f the total development expenditure. Even if the funds are released before the end o f the fiscal year, the implementation o f those projects and programs will be disrupted. The problem i s more a reflection o f procedures than a lack o f adequate capacity. Sector revenue. Contributions by water users and Local Authorities to the Sector activities are still inadequate. The revenue collection by UWSAs improved from Tshs 8.5 billion in 2000/01 to Tshs 11.5 billion in 2001/02, an increase o f 35 percent. If arrears are excluded from the 2001/02 revenue, the increase i s only marginal at 10 percent and it i s attributed to increased tariffs and clientele. The clientele increased by 8,993 to 127,426 and accumulated arrears decreasedby 17.6 percent from Tshs 6.8 billion to Tshs 5.6 billion. Unlike inthe past when the arrears problem was more with the public sector consumers o f water (PER for Water Sector, 2002), currently about 60 percent o f the uncollected revenues originate from private consumers. Thirteen out o f the total eighteen UWSAs depict this trend. OutputPerformance Based on the Annual Report on the UWSAs produced by M o W L D in 2001/02 18 UWSAs produced a total volume of water amounting to 99,207,000m3 equivalent to an average o f 271,800m3/day. This volume is enough to supply a population o f 2.7m assuming losses o f 20 percent and gross per capita demand o f 80 lpcd. The current population o f all 18 urban centers (regional headquarters), i s estimated at 3.4 million and hence gross coverage level o f 73 percent only. Thus, urban water supply coverage has increased from 68 percent inDecember 2000 to 73 percent inJune 2002. The coverage o f rural water supply has also increased but only marginally from 48.5 percent in December 2000 to 50 percent in June 2002. In addition, some improvements were made inthe quality o fwater. 153 The sewerage coverage inurban areas is still very low. Only the cities of Dar es Salaam and Mwanza and eight municipalities of Arusha, Moshi, Tanga, Tabora, Morogoro, Dodoma, Iringa and Mbeya have sewerage systems with an average coverage of about 17 percent3'. Thus, the performance of the water sector during the past three years i s lagging slightly behindthe targets specified under the PRSP and the MTEF. The PRS target was that 55 percent of the rural households should be supplied with clean and safe water by 2003. The corresponding target for the MTEF was 53 percent by 2003. Medium-term Targetsand ResourceRequirements The targets set for the sector inmediumterm plan(2003/2004 -2005/2006), are: Raising the proportionof rural population that has access to safe and clean water from 50 percent in 2002 to 53 percent in 2003 and 60 percent by the year 2006 and the corresponding ratios for urban population being 73 percent, 75 percent and 78 percent, respectively, over the same period; 0 Increasingthe coverage of urbansewerage and drainage facilities from 17 percent of urban population in2002 to 30 percent by year 2006; 0 Reinforcing legal and institutional involvement of local communities and the private sector in developing and managing water services, and reduce water pollution levels from 20 percent in2002 to 10percent by year 2006; and a Training gender sensitive and competent professionals and qualified technicians/engineers for the sector development and managementby 2006. The government interventions towards achieving the mediumterm targets include: Rehabilitating water supply systems, pumping facilities, treatment plants, and maindistributionnetworks; Expanding the existing systems thereby increasing service levels inall areas; Constructing new water supply systems so as to bring service to areas not covered by existing water supply systems inbothrural and urbanareas; Developing, extending and upgrading urbansewerage facilities; Rehabilitation and acquisition of new earth moving and drilling equipment for borehole drilling and dam construction; Rehabilitation and expansion of hydrological, hydro geological and hydro meteorological networks; Exploration of underground water to find alternative or supplement of surface water sources especially indry areas to ease drilling of boreholes; Protection and conservation of water sources from pollution and close monitoring o f water quality; and Enhancing human resource development at all levels in the fields of water resource development and management. 31This is an average coverageo f the 10municipalities. 154 Resource Requirement. Annual financial requirements based on full costing o f the priority interventions i s summarized in Table 18. The aggregate resource requirement for the fiscal year 2003/04 i s Tshs 97.6 billion. lType of Budget and Costing Center Approved Projection Budgets Estimates of (In Million Tshs) Budgets 200212003 200312004 b00412005 b00512006 Recurrent ersonal oWLD 2,836 2,92 1 2,979 3,039 molument (PE) Councils 2,252 2,409 2,578 2,758 Total PE 5.088 5.330 5.557 5.797 11,860 22,030 23,911 4,962 5,458 6,004 Total OC 11,828 16,822 27,488.30 29,914.99 Total Recurrent Budget 16,916 22,152 33,045 35,712 bevelopment Local MoWLD 3,271 29,442.50 12,993 12,963.90 Councils 1,107 1,661 2,49 1 3,736 Foreign MoWLD 38,236 46,000 51,000 47,000 I I lcouncils IO 10 Io lo I Total Development Budgets 41,507 75,442.50 63,993 59,963.90 Aggregate Resource Requirement 58,423 97,595 97,038 95,676 Source:MoWLD, Resource Estimatesand Projections Inputsto the Budget Guidelines(2003104-2005/06) and Budget Estimates (Budget Memorandum FY 2002/2003) Availability of Resources The potential source of funds for the water sector inthe Medium Term Plan will remain the Central Government, Local Government (Town and District Councils), water users, UrbanWater Supply and Sewerage Authorities (Regions and District town headquarters) and already established River Basin Offices (Water User Fees). The key sources o f funds remain to be the Central Government and donor community. The funds collected from other sources are usually utilized in their respective local areas o f water sector service delivery. The sector will likely continue to receive resources inadequate to meet PRSP targets. According to the Ministry o f Finance, the expected allocations, from the Central Government and donor community, for FYs 2003/04, 2004/05 and 2005/06 are 82 percent, 77 percent and 73 percent of sector resource requirements, respectively. Thus, the resource envelope needs to be increasedto makereachingPRSP targets attainable. Conclusions and Recommendations Tanzania has abundant water resources and ifmanaged properly and sustainably they can support the country's population as well as the increasing socio-economic activities. 155 Inadequate funding, however, i s the main constraint to realizing PRSP targets and improving the management and protection o f water resources and the supply adequate and quality water, sewerage and sanitation services. Some Constraintsinthe Sector There is lack o f capacity at Local Government Authorities, which are the central organs inimplementation water sector activities. 0 There is confusion regarding the policy guidance on small towns and district towns. While the Revised National Water Policy provides guidance that these small towns and district towns shouldbe managed through the urbanwater supply sub sector, the current set up at the ministryis contrary to this. They are therefore managed through the rural water sub sector and currently there are no efforts to rectify this. Water sector resource requirements from Regions and Local Government Authorities are yet to be integrated into the sector ministry's input o f the water sector to the Budget Guidelines. 0 Monitoring networks, databases and facilities to process and analyze water resources related data and information are not well developed. The data that exist are often scattered across several governmental departments and are difficult to access. Inaddition, the sector lacks gender segregateddata. 0 Water sector continues to depend substantially on donor financing (development) while future donor fundingis unpredictable. Recommendations The following few key recommendations are made to improve information about the sector and performance (more recommendations are inthe main Sector Report): More financial resources are neededfrom all stakeholders. Speed up the use o f the sector wide approach, which i s an integrated approach usingthe water basins as a unit o fmanagement andcoordination. Continue sensitizing communities and private sector to increase their participation, ownership and contribution to the water sector. Collect and prepare a gender disaggregated data base inthe water sector Increase the financing o f capacity building activities at all levels. Create a budget line in the Ministry for sanitation probably in the DPP so as to carry out research for revealing andpublicizing sanitationissues. Increase more funding to the sector in general and inparticular to the rural water supply and sanitationandfor capacity building, as well as to finance activities that will helpto reduce the spread o f HIV/AIDS. 156 Roads Introduction This year's Public Expenditure Review was prepared for roads under the Ministry of Works jurisdiction (trunk and regional roads) and .those under Local Authorities jurisdiction (district, urban, feeder and improved unclassified roads). The main objective of this PER i s to analyze recent trends and financial performance o f the Sector, review existing plans and strategies, discuss the institutional performance, and provide some recommendations. TheRoad Network -A Background The actual road network in Tanzania i s not accurately known and past efforts at improving the road condition have not beentracked for all classes o f roads. Although the official road length i s 85,0OOkm, the roads authorities show a total length o f 78,5 1O h . The latter is a useful figure for budgetingand assessingperformance. The road condition in2002 is showninTable 19below. Category Good km (?A) Fair km (%) Poor km (99) Total Trunk roads 3,300 (33) 3,900 (39) 2,845 (28) 10,045 Regional 3,725 (20) 7,475 (40) 7,265 (40) 18,465 . PO-MLG 3,995 (8) 9,755 (20) 36,250 (72) 50,000 TANROADS and PO-RALG are the two key implementing agencies using funds from RFBand other sources such as Government ofTanzania (GOT) and donors. The agencies implement the program as funds become available. Apart from the reporting that the Ministry of Works (MOW) receive from the implementing agencies there is no other follow up mechanismthat is inplace to supervise the implementation o f the program. Financia1Performance There has beena general increase inthe budget and actual allocations to the sector from FYOl to FY03. The allocated budget to the M O W for FY03 is more than double that of FYO1. As a share o f total domestic revenues, the allocation for the M O W has increased from 4.6 percent to 7.2 percent inthe same period. Approximately 93 percent, 87 percent and 95 percent of the budget allocation to M O W in FYO1, FY02 and FY03, respectively, went to roads sector departments. The increased fundingto M O W has facilitated the flow o f funds to trunk andregional roads. 157 IDevelopment 4,5541 4,3541 95.61 16,4541 16,0341 971 34,5191 Recurrent 37,604 55,982 148.9 48,192 48,877 101 54,115 / D e " e l ~ ~ r r c lI 4,554~ 4,354~ 16,4541 16,034 97 34,519 Recurrent 34,695 52,839 39,769 42,566 I07 49,989 ~~~~~~: ofWorksF i ~ a ~ cAccounts2001/02;EstimatesofPublic Expenditure Ministry i ~ l ConsolidatedFund Services2000101, 2001/02,2002/03 Most of the increased a l l o ~ a t ~has~b e ~ e ~~ ~e e v~ ~e ~l ~~ e ~nwhich increased~ o i ~~ ~ et ~ ~e seven fold between FUO1 to FU03 (Table 20). Recurrent e x ~ e ~al~oca~ionincreased ~ i ~ ~ r ~ by 44 percent inthe same ~ e r ~ of this increase~exclusively b ~ ~ - ~ s ~ ~ ~OC. ~ e ~ ~ d 158 Actual 10,470 IO, 850 15,240 NA Variance (?A) NA (23) 0 NA Resources that are budgeted for urban, district and feeder roads (LA roads) under PO- RALGhave also increased over the past three years (Table 21). GOT maintenance expenditure over the past three years however has not changed significantly with the road fund revenues being the main source o f funding for roads maintenance. Additional domestic resources in the roads sector, mainly from additional GOT tax revenues have been spent on the special trunk roads projects. The GOT allocates Tsh 22 billion annually for special trunk road projects. This i s not sufficient to cover the projects' financing requirements for 2003/04, which are Tsh 61 billion. About Tsh 39 billion more will have to be raised by the government duringthe coming financial year. The total finance gap for the special trunk roadprojects duration is Tsh 101billion. Resources from donor community. In line with GOT priorities, donors have also been providing significant resources Fig. 2: Donor Funds-MOW Roads (see Figure 2) to the roads sector, most I o f which targeted trunk and regional roads development. There are however, disparities between donor commitments and actual funds released with the latter being lower than commitments. There are significant donor funds going to LA roads but it i s difficult to determine the actual level, particularly due to the lack 1999/00 2000/01 2001/02 2002/03 o f a central co-ordination mechanism for Source: Estimates of Public Expenditure MoF such funds. It i s thus difficult to assess the contribution o f such funds to overall LA roads maintenance and development. Utilisationcapacity. The level o f utilisation o f GOT finds by the roads authorities has not expanded in line with actual allocation and disbursements leading to rollovers. This situation holds for LAs and TANROADS. For example in 2000/01, out o f the Tsh 25.6 billion released to TANROADS, Tsh 8.6 billion (33 percent) o f the released funds remained unspent at the year-end. Mainreasons for rollovers are: 0 The duration taken before the Performance Agreements (PAS)between RFB and the agencies are signed i s very long; they effectively cover only 9 months o f the year. Funds are not disbursed before signing PAS. The RFB i s currently considering the use o f interim PASfrom next fiscal year when waiting for final PASto be signed. 159 The time required for procurement of goods and services under the Procurement Act is very long. More often than not, funds released in April and onwards are prone to rollover. The TANROADS has forwarded its proposals to the government seeking to rectify the procedures and remedial actions are being worked out. Inadequate procurement expertise. Some efforts are being made at TANROADS to improve their delivery capacity and procurement expertise. Mismatch between timing of funds release and the best time for carrying out the roads works. The low capacity o f local contractors in terms o f equipment, finance and sometimes expertise. There was some training for engineers and contractors in view o f enhancing their capacity to deliver. More needs to be done particularly on equipment. The financial performance for TANROADS roads maintenance works has been around 70 percent since 2000. Ifthe utilisation capacities are improved, the performance is likely to grow further. The table below illustrates the financial performance o f TANROADS for the past three years. In 2000/01, for example, LAs were provided with Tsh 14,159 million for roads maintenance, but managed to spend only Tsh 5,916 million, which was 42 percent o f the budget. This low level o f performance was mainly due to low contracting capacity and inadequate capacity at district level to manage road contracts. Most o f the actual maintenance expenditure in LAs went to periodic and spot/emergency works, compared to routine maintenance. Giventhe utilisation capacity problems inTANROADS and LAs, the Table 22 illustrates the low level ofroadnetwork maintained compared to plan. Roadscategory 2000lO1 2001/02 Planned Actual Actual against Planned Actual Actual against km km plan(inpercent) km plan(Inpercent) TANROADS 18,538 13,248 72 18,869 14,471 77 LAs roads 8,758 6,161 70 10,120 6,599 65 Total 27,296 19,409 28,989 21,070 General (OCAG) report for 2000/01,the financial performance within GOT organisations Financial audit. In general, according to the Office o f the Controller and Auditor has improved recently. The Ministry o f Works received a qualified audit report as against an adverse report in the previous year. The financial performance of TANROADS as was judged by the OCAG report 2000/01 as reasonably fair and received a qualified report. RFB performance was even better as it received a clean report. For local authorities, there has been also a general improvement trend in financial performance. The reports for 2001/02 were not yet out duringthe PER review. 160 Existing Plans and Strategiesfor the Road Sector Planned improvement. Based on existing plans, programmes and projects in the road sector, it i s anticipatedthat the road network condition inthe next three years will change as shown inTable 23. Itis noteworthy that by the target year, all tarmac roads will be ingood condition and will therefore require full maintenance. Over 60 percent o f the road network will be in good or fair condition by 2005/06 current commitments are implemented. 2002 2006 I 1 Good 1 Fair 1 Poor 1 Good 1 Fair 1 Poor I Unpaved 3,700 7,300 7,220 10538 6012 1520 Unpaved 3,955 9,657 35,888 12,4771 7,427 29600 The Poverty Reduction Strategy (PRS) Road Sector Progress report dated August 2001, has 15,524 km o f road network that will be subject to various types o f interventions ranging from rehabilitation to routine maintenance. The network was identified in the twelve regions considered to be among the poorest. Out o f the road network identified, 12,780km are district roads; 418 km urban roads and 2,326 km are regional roads. The targets for the PRS were to: (i) rehabilitate 4,500 km inthe regions by 2003; (ii)improve 6,105 km o f the road network in the regions that i s in faidpoor condition to good condition by 2003; (iii) undertakeroutine and periodic maintenance promptly on all rural road networks. Cost consequences and financing requirement. Based on the current commitments in the road sector, it is anticipated that road maintenance requirements will be approximately Tsh 171 billion excluding administrative costs (equivalent to US$ 169 million). This estimate i s based on the current unit costs o f maintenance as provided by the Ministry o f Works and the estimated maintainable road length from the various programmes and projects. Inlight o f the current revenue for the road fund account, (Tsh 52.9 billion in FY02, whereby only 90 percent goes out for road maintenance) there i s a huge funding gap for road maintenance budgetthat is equivalent to Tsh 118.1 billion (the current RFB funding therefore, accounts for about 31 percent o f the roads sector maintenance requirements). Even though funding needs for maintenance o f the network are huge, the absorption capacity on the ground i s still limited. 161 On the basis o f the current status o f financing road maintenance it i s recommended that: (i) RFBshouldbesupportedtoestablishasustainablefundingmechanismforroad The maintenance based on the proposed projections; (ii)The level of funding of road maintenance should take into account the absorption level o f the road authorities-- TANROADS and LAs; and (iii) The M O W should spearhead efforts at increasing the capacity and capability o f the road contractors and road engineers, particularly at the council level, to raise capacity to absorb more funds for road maintenance. Institutiona1Performance The institutional framework for the road sector has now been completed and is about to be firmedup inthe Roads Act 2003. The reforms have beenundertaken since 1985 with the objective of separating the policy and regulatory functions with the executive functions. The changes in the structure and functions o f road maintenance and development in the Ministry o f Works led to: (i)Enactment of the Roads Tolls (Amendment)Act Number 2 of 1985; (ii) Tolls Act o f 1998; (iii) Roads Formation of the Road Fund Board (RFB) in December 1998; and (iv) Formation o f the Tanzania Roads Agency (TANROADS) inJuly 2000. The institutional framework for the maintenance and development o f LA roads, which include urban, district, feeder, and improved unclassified roads i s still under the Local Government Authorities. There are measures towards a sector-wide approach, although at a nascent stage. These include the drafting o f a Roads Act that specifies the roles and responsibilities o f each institution in the sector, and the establishment o f a PER process that now covers the whole sector and recommends the required resources for the budget. Additional measures include strengthening the regulatory role o f the M O W in setting standards for all contractors and engineers inthe country. The Draft Roads Act provides a good basis for firming up the current institutional framework but does not provide clear role and responsibilities o f LA. A weakness is that there are no specific criteria for identifying roads for maintenance at the Council level. As a result, the road funds that are allocatedto Councils do not seek to achieve specific targets and outcomes. The District Road Management System (DROMAS) that i s currently being developed will provide effective and transparent criteria to the Council Management Team in identifying roads that should receive maintenance. 162 Recommendations Issuesfrom theprevious PER that need to be taken up in the MTEF The MTEF should focus on completing the issues that have been identified inthe PER 01 and PER 02 with respect to: Completing the establishment and implementation of the R M M S and DROMAS; Ensuringdetailed monitoring o f performance of the road authorities with respect to unit costs o f maintenance; Up-dating o f the 10 year RSDP and undertaking a RSDP for the LA roads Implementingthe roadimprovement programme underthe PRS; Implementing measures proposed by the RFB to establish sustainable mechanisms for financing roadmaintenance; Buildingcapacity inthe LAsto enable themmanage roadworks effectively; Building capacity in the local construction industry in order to ensure improved quality o f services to the road authorities; and Training local communities in the maintenance and management o f local community roads. Among the many new issues discussed inthe main report and that need to be taken up in the MTEFare: Both the RFB and the authorities need to strictly monitor performance measures that included inthe performance contracts. The funding o froad maintenance shouldtake into account the absorption capacity o fthe road authorities. The draft Roads Act needs to be amended. The role and responsibilities o fthe LA inthe new institutional set uphasto be well spelt out. The relationship between Councils, PO-RALG and MOW need to be defined; and essential role of Regional Road Boards will be clarified, particularly inrelation to LAs. Establishing clear mechanisms for implementing cross cutting issues such as the environment, women participation, HIV/AIDS, labour based technology and capacity buildingto enable the road works programme to contribute positively to welfare ingeneral. 163 Agriculture Introduction This year's PERupdate for the agriculture sector builds uponthe previous ones. It covers all three Ministries that deal with the sector, namely, the Ministry o f Agriculture and Food Security (MAFS), Ministry o f Cooperatives and Marketing (MCM) and the livestock sub-sector o f the Ministry o f Water and Livestock Development (MWLD). The overall objective o f this PER update i s to review sector performance, public service delivery, recent trends in public expenditure, and undertake costing o f priority expenditureprograms and activities to feed into the budget guidelines for the upcoming fiscal year. Sector Strategiesand Programs: A Background Agriculture i s the dominant sector in the economy and contributes about 50 percent to GDP, provides 85 percent o f rural employment, and generates 73 percent o f all export earnings. The sector also harbors most o f the poor in Tanzania. Hence, accelerating agricultural growth i s crucial for meeting poverty reduction objectives and as well as for the health of the overall economy. It is from this background that the government formulated the Rural Development Strategy (RDS) and subsequently the Agricultural Sector Development Strategy (ASDS) in line with the National Vision 2025 as the principal strategic documents to guide the policy initiatives for achieving the PRS objectives and targets.j2 To operationalise the ASDS, the Agricultural Sector Development Program (ASDP) i s also being developed and the draft document i s already out. Within the framework o f RDS and therefore ASDS, the government is committed to creating the proper environment for agricultural transformation through: (a) decentralizing administrative and implementation-responsibility to district councils, (b) liberalizing agricultural markets and dismantling state monopolies in foreign trade; (c) redefining food security objectives; (d) increasing the reliance on the private sector (comprising smallholders, commercial farmers and pastoralists) for all agricultural production and marketing activities; and (e) strengthening government responsibility for assisting and regulatingthe industry. A key outcome from the sector strategies and programs will be that most of the responsibilities and activities in the sector will devolve to local government authorities (LAs). Through the Districts Agricultural Development Plans (DADPs), LAswill execute 75 percent o f the activities in the agricultural sector, while the Agricultural Sector Lead Ministries (ASLMs) will manage the remaining 25 percent. The ASDP seeks to realize this vision withinfive years. This makes the success o fASDWASDP crucially dependent on the capacity o f LAs to develop and execute agricultural programs in a participatory manner. 32The vision is to transform the economy such that, by the year 2025, poverty would have been eradicated inthe country (Tanzania Development Vision 2025, page22). 164 Sector Performance and Public Sector Delivery At present, interms o f sector performance, it is not possible to separate the influence of policy interventions from those o f natural factors to explain the performance o f the agricultural sector. With regard to assessing improvements in service delivery, there are no records o f actual performance to compare against the current state. Establishing such data inthe future will helpto monitor and evaluate actual progress made. Sector Performance As in the past, agricultural growth during the period 1993 to 2001 varied from year to year. Between 1993 and 1998 the annual growth rate o f real agricultural GDP averaged 3.4 percent It subsequently increased to 4.1 percent in 1999 before declining to 3.4 percent in 2000. In 2001 and 2002, GDP growth increased again to 5.5 percent and 5.0 percent, respectively. Duringthe last season, the performance of food crop production improved mainly due to good weather conditions, and some policy measures, such as increased efforts to control crop diseases and pests. Though the productionvolume o f cash crops increased, its value has suffered from declining prices inthe world market. Cassava 1674 1,725 3.1 Potatoes 655 950 45.1 Total Non-Cereals 3718 4,111 10.6 Food crops. Table 24 presents an overview o f food crop production over the past two years and the forecast production for 2001/03, together with the annual average production inthe previous 3 years period 1998/99 to 2000/01. While there are variations between years, mainly due to climatic and environmental factors, the continuous growth inmaize, sorghum/millet, rice, pulses, potatoes andtotal non-cereals over recent years is significant. 165 The assessment o f food security situation in the country indicates that the national food self-sufficiency ratio, defined as a ratio o f production to the amount o f food required, was 94 percent for FY02 and i s expected to improve to 102percent for FY03. Cash crops. As indicated inTable 25, with the exception o f cashew nuts, there has been a significant and sustained increase in the production o f traditional cash crops in recent years. These favourable trends inproduction, however, have not always been reflected in export values, mainly because o f variations in world prices and, in some instances, because o f falling quality. Livestock Meat production. Meat production over the past years has been growing, but slowly. The annual growth rate was 0.8 percent in FY02 and expected to be only 1.1percent in FY03. Milk production. Milk production has also displayed a consistent and marked growth over the recent years. However the annual growth rate i s expected to decreased from 10.6 percent in FYO1-FY02)to 4.7 percent in (FY02-FY03). Generally, the traditional sector continues to play the major role in milk production within the country, and i s expected to contribute about 62.4 percent o f total milk production inFY03. 'YODeviation from %Deviation from Base Period Base period Source: Ministry ofAgricultureandFoodSecurity. Notethat a baleof lint cotton= 181kgs. (*)indicatesforecasted figures. Hides and skin production. Cattle hide collection has been increasing at decreasing rates for the past three years. In FY02 it increased by 7 percent while in FY03 it i s expected to increase by 6 percent. Meanwhile goat and sheep skins collection has fallen by 25 percent and 20 percent respectively. These figures seem to be relatively low considering the total ruminant livestock population inthe country. Most o f the hides are being exported inraw form with very few hides exported as wet blues. 166 Public ServiceDelivery Given the fact that at present there i s no adequate data on public service delivery, this review is limited to describing the current situation and should not be regarded as comprehensive inassessingpublic service delivery by ASLMs. So far, there are no established standard performance indicators for delivery o f public services by the ASLMs. Recently, the ASLMs have each issued a client service charter, explaining among other things, the core principles o f public service delivery, nature and standards of services, clients' rights and responsibilities, and a feedback mechanism for evaluation o f actual performance. The Charters also commit the ASLMs to report performance against standarddtargets. The targets have been set up in terms o f time needed to deliver a service. However, the criterion that has been usedinthe setting up o f targets i s not clear and needto be revisited. Overall, inadequate funding i s the main constraint that adversely affects public service delivery in the agricultural sector. It disrupts smooth implementation o f the planned activities. For example, due to inadequate funding, ASLMs have not been able to provide transport facilities to facilitate provision o f extension services. Maintenance o f 4,637 bicycles, and 915 motorcycles used by frontline extension staff could not take place due to inadequate funds. As a result o f transport bottlenecks, only 65 percent o f targeted farmers could be reached. Shortage o f extension staff due partly to freezing employment inthe civil service constrains the delivery of extension services to farmers. Inthe MAFS, the ratio o f extension agents to farmer is low. It stands at 1:1000-1500 and is far away from the recommended ratio o f 1:600-700. Another key constraint i s the conventional top-down extension methodologies. Adoption of demand driven extension service i s neededto redress the situation. Budget Performance Overall trends. The combined approved and actual expenditure o f ASLMs has been generally on the rise over the past three years (Table 26). This i s also true for both recurrent and development budgets. The sharp increase o f the actual recurrent expenditure in2000/01 was largely an outcome o f the creation o f the new ministries. Table26: Total Expenditure(Local andForeign-Funded)inMill Tshs Expenditure FYOO FYOl FY02 FY03* Approved Actual Approved Actual Approved Actual Approved Recurrent 12,725.6 7,418.8 24,644.1 33,796.9 33,718.5 31,964.3 46,426.6 Development 35,635.0 15,413.4 28,708.4 21,361.5 25,074.0 20,842.7 38,113.6 Total 48,360.6 22,832.2 53,352.5 55,158.4 58,792.5 52,792.5 84,540.2 Note: * means figures for 2002/03 do not include Export Credit Guarantee SchemeFundworth Tshs 6.5 billion. Source: documents for 2002/03 -2004/05. Appropriation Books of Accounts for the Sector Ministries-1999100, 2000101 and 2001/02 and MTEF Generally, actual disbursement appears to be considerably lower than the approved expenditure, except for the FYOl recurrent expenditure. The main causes for deviations are: (a) budgetary management at the national level leading to a situation where there 167 were inadequate funds to meet approved expenditures; and (b) the Cash Budget Management System and the policy-influenced allocation o f resources within the mini~tries.~~gap betweenthe approved and the actual, however, hasbeennarrowing. The Donor funding. The local-foreign resource split shows that, for the past three years, external funding has continued to dominate the budget. The local contribution has been very small. Planned/actual vs PRS expenditure. Despite the general increase in expenditure toward the sector, the allocated budget i s considerably small relative to PRS proposed expenditure(Table 27). Development expenditure. There are distinct differences in the deviations between planned and actual expenditure in both local and foreign fund contributions (Table 28). With the introduction o f the MTEF approach to budgeting in 2000/01, local funds released for development project under counterpart funding arrangements in the agricultural sector were 100percent. PlannedExpenditure* 28,000.13 36,395.10 53,009.03 Actual Expenditure** 55,158.48 52,792.50 NA PRSP Proposed Budget*** 66,149.00 79,111 .OO 95,95 1.oo Percentageofthe Planned (local contribution only) to PRSP 42.33 46.01 55.25 Proposed Expenditure Percentageo fthe Actual (local and external contributions) to PRSP 83.4 66.7 NA ProposedExpenditure ~~ 33 Since cash budgeting was introduced as one o f the measures to improve government fiscal policy management, the sector ministries have faced considerable financial constraints. This is because funds were only disbursed on a monthly basis in relation to revenue collections and these were usually insufficient to meet sectoral requirements. 168 Table28: PlannedVersusActualExpenditurein the DevelopmentBudget inMillTshs 1999/00 2000/01 2001/02 %Actual/ YOActual/ %Actual/ Sourceof Fund Planned Actual Planned Planned Actual Planned Planned Actual Planned Local 3,294.50 147.0 4.5 3,355.99 3,665.56 109.2 2,676.60 2,672.60 99.9 Foreign 32,340.56 13,902.42 43.0 25,352.44 17,696.01 69.8 22,397.40 18,170.18 81.1 Total 35,635.06 14,049.42 39.4 28,708.43 21,361.57 74.4 25,074.00 20,842.78 83.1 I% localltotal 9.2 1.0 11.7 17.2 10.7 12.8 YOforeignitotal 90.8 99.0 -- 88.3 82.8 -- 89.3 87.2 Source: MAFS Other charges. Analysing 10 major OC expenditure items inMAFS only for fiscal year 2001/02 and their corresponding approved estimates for FY 2002/03 indicates that those items that benefited most from the increased level o f expenditure were the priority ones as stipulated in both the ASDS and ASDP. These items include acquisition o f breeding stocks and seeds for research and seed multiplication services, air operations for control o f crop pests and diseases, utilities for maintaining efficient delivery o f services, and acquisition o f technical equipment andtools. Trendsin the Funding of Priority Areas The following tables give the trends, by ministry, inthe resource allocationto the priority areas that are identified for increased finding inthe forthcoming budget. It i s clear from the Tables 29a, 29b, and 29c, that required funding i s significantly higher than what has beenallocated before historically. This is partly due to the new activities that have been earmarked for funding consistent with the ASDWASDP. These new activities are listed inmainSector report. These activities canbetraced inthe log frame ofthe ASDS. 169 Table 29a: Trends in FundingPriority Activitiesin the MAFS S/No Activitvlhout " . 200012001 2001102 2002103 2003104 Actual Actual Budget Full Cost Requirement 1 Policy, Regulatory and Inst. 659,120,758 859,3 18,505 890,498,412 2,2 15,569,933 Framework 2 Agricultural Sector Information 49,832,488 31,943,903 53,768,562 374,100,388 3 Str. Training, Advisory and Tech. 4,751,895,713 2,572,111,616 4,885,801,836 9,608,925,056 Services 4 Agriculture ResearchServices 303,331,578 220,084,345 370,450,000 1,787,950,000 5 Private Sector Development 4,6293 15 7,792,990 86,967,600 6 Cross Cutting Issues 213,419,000 TOTAL MAFS 5,764,180,538 3,688,088,184 6,208,311,800 14,286,931,977 Table 29b: Trends in FundingPriorityActivities inMWLD S/No A. Activitylinput 200112002 200212003 2003104 Actual Budget FullCost requirement 1 Policy and InstitutionalFramework 308,133,600 750,000,000 2 Livestockinformation 18,505,021 257,936,200 3,000,000,000 3 Researchand Training 742,652,559 785,118,100 3,372,960,000 4 Animal production, extension and 879,305,291 1,584,552,100 6,116,550,000 rangelanddevelopment) 5 Delivery of Veterinary Services 659,865,048 2,017,991,700 5,126,064,000 6 Cross cuttingissues(A1DSand gender) 0 0 600,000,000 Total MWLD 2,608,461,519 4,645,598,100 18,965,574,000 Table 29c: Trends in FundingPriority Activities in the MCM SIN0 Activity 200112002 Actual Approved Approved Full cost 200112002 budget budget requirement per 2002103 2003104 activity 1 PolicyandRegulatory Framework 80,890,823 287,040,576 324,983,478 410,000,000 2 Strengtheningtraining, cross cutting issues 979,560,300 1,254,361,464 1,157,753,639 1,300,000,000 3 Tech. services, information, andresearch 750,20 1,200 439,480,800 572,783,379 2,640,000,000 4 Priv. sector development and advisory serv. 1,860,500,136 734,314,178 909,953,737 5,753,600,000 5 Agricultural finance 46,475,200 62,809,282 105,619,630 120,000,000 Total MCM 3,717,627,659 2,778,006,300 3,071,093,863 10,423,600,000 Costingof Priority Interventions The costs o f resource requirementsfor the next financial year and beyond i s based on the detailed costing o f individual items and activities, which are consistent with the ASDS ASDP, log frame.34These individual activities are only those that have been already 34Ingeneralthe priority areas of interventionsthat are proposedhere are well informedby the PRS, ASDS, andthe draft ASDP. 170 identified. The cost projections give some allowance to inflation and the required gradual shift ofresources toward LAsactivities. Table 30:Summary for Sub-programme Priority Costing Estimates 2003104 - 2006107 (in TShs.) Category 2003104 2004105 2005106 2006107 A 19,922,000,000 29,479,000,000 42,123,000,000 58,646,000,000 B 35,336,268,411 32,3 13,432,670 32,344,270,502 33,656,035,232 C 531,419,000 565,740,140 617,320,950 671,929,407 TOTAL 55,789,687,411 62,358,172,810 75,084,591,452 92,973,964,639 The projected costs are illustrated in Table 30. Category A are activities undertaken by the districts. Category B are activities undertaken by ASLMs which are essentially national in nature and deal with issues such as the regulatory framework, commodity boards, and research and development. Category C activities are cross-sectoral issues also undertakenby ASLMs. By the end o f the first five years of ASDP, Categories A, B, and C, are envisaged to executing 75 percent, 20 percent, and 5 percent, o f the total budget resources, respectively. Ideally all activities under sub-programme A are to be identified by the District Agricultural Development Plans (DADPs) through a participatory approach and the district council. In this sense, local needs and priorities will be properly taken into account. Recommendations There are many recommendationproposed inthe sector report, The mainones are: While the introductiono f the client service charter marks a beginning o f a new era in which the performance o f the ASLMs in delivering public services could be monitored and evaluated, the setting o f targetdstandards needs to be revisited in order to come up with more realistic targets. The ASLMs should also establish data base to keep records o f actual performance in delivery o f public services for monitoring and evaluation purpose. Efforts should be made to inform the general public/clients o f the clients' service charter, so that they can understand their rights and responsibilities. A feed back mechanism must also be instituted to facilitate monitoring and evaluation. Since agriculture i s a priority sector, it i s important for the government to increase overall funding, and particularly the share o f local funding. As more responsibilities devolve to the LAs, new challenges emerge. Inview of the fact that already many LAs have failed to produce DAPDs intime for budget preparation, it i s proposed that: (a) Inter-ministerial coordination i s strengthened; (b) a block grant be set aside in the budget for the agricultural activities that would be carried out under the LAs to be accessed only by the LAs that produce the DAPDs; and (c) a monitoring and evaluation mechanism be put in place to assess the performance o f the LAs. In particular, expenditure-tracking studies should be commissioned regularly by the ASLMs to assess the use o f agricultural funds by the LAs. 171 ANNEXE: Data 172 CONTENTS Table A1: TanzaniaMacroeconomicIndicators.................................................... 167 Table A2: Balanceof Payments ....................................................................... 168 Table A3: 170 Table A4b: BudgetFrame- Analytica (as YOof GDP) .............................................. Table A4a: BudgetFrame- Analytical(in Bill.Tsh.) ............................................... Summary ofCentralGovernmentOperations .......................................... 171 173 Table A5a: 174 Table A5b: BudgetFrame- Accounting( as % of GDP) .......................................... BudgetFrame- Accounting(in Bill. Tsh.) ............................................. 175 Table A6a: 176 Table A6b: Sectoral- DevelopmentExpenditure(inBill.Tsh.) .................................. Sectoral- RecurrentExpenditure(in Bill. Tsh.) ...................................... 179 Table A6c: Sectoral - TotalExpenditure (inBill Tsh.) ............................................ . 181 Table A7a: Actual RecurrentExpenditure............................................................ Actual RecurrentExpenditureby Vote as the Shareof Total 183 Table A7b: Actual DevelopmentExpenditureby Vote as the Share of Total Actual DevelopmentExpenditure................................................................. 185 Table A7c: Expenditure................................................................................. TotalActualExpenditureBy Vote As The Share OfTotal Actual 187 Table A8a: Actual Recurrent Expenditure as a Percentageof BudgetedRecurrent Expenditureby Vote ....................................................................... 189 Table A8b: ActualDevelopment Expenditureas a PercentageofBudgetedDevelopment Expenditureby Vote ....................................................................... 191 Table A8c: Actual TotalExpenditureas a PercentageofBudgetedTotal ExpenditurebyVote ....................................................................... 193 Table A9b: DevelopmentExpenditure- Regions(in Bill.Tsh.) ................................... Table A9a: RecurrentExpenditure- Regions(in Bill.Tsh.) ....................................... 195 196 Table A9c: TotalExpenditure- Regions(inBill. Tsh.) ............................................ 197 Table AlOa: ActualRecurrentExpenditureas a PercentageofBudgetedRecurrent Expenditure-Regions .................................................................... 198 Table AlOb: ActualDevelopmentExpenditureas a PercentageofBudgetedDevelopment Expenditure-Regions .................................................................... 199 Table AlOc: ActualTotalExpenditureas aPercentageofBudgetedTotalExpenditure- Table A11a: Sectoral - Recurrent ExpenditureBefore andAfter Reallocation..................-201 Regions ...................................................................................... 200 Table A1lb: Sectoral - Recurrent ExpenditureBefore andAfter Reallocation,FYOl ............202 Table A1IC: - Table A12b: Regions- RecurrentExpenditureBefore andAfter Reallocation,FYO1 ..........205 Table A12a: Regions- Recurrent ExpenditureBefore andAfter Reallocation,FYOO ..........204 Sectoral Recurrent ExpenditureBefore andAfter Reallocation,FY02 ...........203 Table A12c: Regions- RecurrentExpenditureBeforeandAfter Reallocation,FY02 ..........206 PrioritySector Expenditures (T.Sh. million) ........................................... . Table A13: 207 173 f 1 0 9 w m g N m N b w - w w b - 3 1 I m m - m - w m o 2 2 4 - N - k k hh o o TableA2: BalanceofPayments in Mill.of USdollars). 1997 1998 1999 2000 2001 20021 L. Current Account -403.4 -905.4 -829.5 -498.6 -480.0 -251.1 ioods: exports E0.b. 752.6 588.5 543.3 663.3 776.4 902.5 Traditional 435.3 356.3 301.2 292.8 231.1 206.1 Nontraditional 317.2 232.2 242.1 370.5 545.3 696.5 Goods: importsf.0.b. -1148.0 -1382.1 -1415.4 -1367.6 -1560.3 -1511.3 'alance on Goods -395.4 -793.6 -872.1 -704.3 -783.9 -608.7 ervices: credit 482.4 521.3 600.3 627.3 679.3 665.8 Transportation 59.4 41.8 53.7 56.8 68.4 61.2 Travel 339.0 399.0 463.7 376.7 412.8 437.1 Other 83.9 80.4 82.8 193.8 198.1 167.5 Services:debit -800.2 -955.3 -795.0 -682.4 -689.5 -712.5 Transportation -205.7 -224.7 -189.9 -205.7 -194.2 -176.7 Travel -407.2 -493.3 -369.7 -337.3 -327.3 -337.4 Other -187.4 -237.3 -235.4 -139.4 -168.0 -198.3 alance on Services -317.9 -434.1 -194.8 -55.1 -10.3 -46.7 'alance on Goods and Services -713.3 -1227.7 -1066.9 -759.3 -794.2 -655.5 icome:credit 43.0 44.4 49.0 50.4 55.3 74.3 icome:debit -164.8 -149.4 -148.3 -180.4 -140.5 -90.6 Direct investment income -0.2 -8.2 -16.8 -13.2 -1.1 -2.0 Interestpayments (scheduled) -163.9 -124.8 -116.0 -146.9 -117.3 -67.6 Compensationof employees -0.6 -16.5 -15.4 -20.3 -22.1 -21.1 'alance on Goods,Services and Income -835.1 -1332.7 -1166.1 -889.4 -879.4 -671.8 'urrent transfers 431.7 427.3 336.6 390.8 399.4 420.7 'urrenttransfers: credit 499.3 454.2 445.6 463.7 469.5 472.9 Government 433.6 421.0 411.4 427.8 418.4 427.7 o\w Multilateral HIPC relief 0.0 0.0 0.0 40.6 71.3 68.8 Other sectors 65.7 33.2 34.2 35.9 51.1 45.2 'urrenttransfer: debit -67.7 -26.9 -109.0 -72.9 -70.0 -52.3 1.Capital Account 270.9 252.4 270.6 330.4 365.2 354.1 'apitaltransfers: credit 270.9 252.4 270.6 330.4 365.2 354.1 GeneralGovernment 259.9 235.9 256.4 314.7 342.9 324.0 Project 259.9 235.9 256.4 3 14.7 342.9 324.0 Program 0.0 0.0 0.0 0.0 0.0 0.0 Debt forgiveness 0.0 0.0 0.0 0.0 0.0 0.0 Other sectors 11.0 16.5 14.2 15.7 22.3 30.1 'apitaltransfers: debit 0.0 0.0 0.0 0.0 0.0 0.0 175 Table A2: BalanceofPayments (cont'd) (inMill.ofUSdollars). 1997 1998 1999 2000 2001 2002 Prov. Prov. C. FinancialAccount, excl. reservesand 236.3 -166.1 613.3 572.5 98.2 306.8 relateditems 0.0 0.0 0.0 0.0 0.0 0.0 Direct investment abroad Direct investment in Tanzania 157.8 172.2 516.7 463.4 327.2 240.4 Portfolio investment 0.0 0.0 0.0 0.0 0.0 0.0 Other investment 78.5 -338.3 96.6 109.1 -229.0 66.4 Assets -69.1 -18.5 14.8 -134.0 -76.7 2.9 Currencyand deposits -69.1 -18.5 14.8 -134.0 -76.7 2.9 Banks -69.1 -18.5 14.8 -134.0 -76.7 2.9 Other sectors 0.0 0.0 0.0 0.0 0.0 0.0 Liabilities 147.6 -319.8 81.7 243.2 -152.4 63.6 Trade credits 11.6 46.9 15.5 12.8 13.0 13.2 Loans 113.6 -305.6 58.0 176.8 -123.0 17.0 Generalgovernment 7.8 -3 16.4 100.0 225.4 -238.8 63.4 Drawings 397.5 254.6 335.8 486.7 157.1 177.0 Repayments -389.7 -571.0 -235.8 -261.3 -395.9 -113.6 Scheduledpayments -162.4 -204.3 -151.2 -251.3 -264.5 -113.6 Rescheduleddebt -227.3 -366.7 -84.6 -10.0 -131.4 0.0 Banks 0.0 0.7 0.2 -0.9 2.0 -0.8 Other sectors 105.8 10.1 -42.2 -47.7 113.9 -45.6 Drawings 117.8 43.0 8.3 2.5 171.6 17.2 Repayments -12.0 -32.9 -50.5 -50.2 -57.7 -62.8 Scheduledpayments -12.0 -32.9 -50.5 -50.2 -57.7 -62.8 Currencyand deposits 22.5 -61.1 8.3 53.6 -42.4 33.4 Other liabilities 0.0 0.0 0.0 0.0 0.0 0.0 D.Net Errors and Omissions -303.6 313.3 -163.3 -439.5 34.1 -81.0 Overall balance (Total, Groups A through 0) -199.8 -505.8 -109.0 -35.1 17.5 328.8 E. Reservesand RelatedItems 199.8 505.8 109.0 35.1 -17.5 -328.8 Reserveassets -182.0 21.8 -175.5 -197.3 -186.6 -372.4 Use of Fundcreditand loans 77.4 11.0 51.3 49.4 15.6 26.0 304.4 473.0 233.2 183.0 153.5 17.6 227.3 366.7 84.6 10.0 131.4 0.0 0.0 0.0 0.0 0.0 0.0 0.0 14.6 73.1 92.7 81.2 14.4 18.4 62.5 33.2 55.9 91.8 7.7 -0.8 0.0 0.0 0.0 0.0 0.0 0.0 4,709 5,572 6,433 7,226 8,260 9,181 7,692 8,382 8,636 9,028 9,425 9,498 -5.2 -10.8 -9.6 -5.5 -5.1 -2.6 -10.9 -15.8 -14.4 -10.3 -9.5 -7.1 623.1 599.0 775.6 974.4 1156.6 1529.0 3.8 3.1 4.2 5.7 6.2 8.3 272.6 283.7 405.1 542.7 761.2 1058.4 -84.4 -11.1 -121.4 -137.6 -218.5 -297.2 624.6 681.0 797.3 803.3 916.3 976.3 IExchangerate (annual average) 612.1 664.7 744.9 800.4 876.4 966.6 Source:Bankof Tanzania, InternationalEconomicsDepartment 176 Table A3: Summaryof Central Government Operations ;in Bill. Tsh.) FY95 FY96 FY97 FY98 FY99 FYOO FYOl FY02 FY03* Total Revenue 331 448 572 619 689 778 930 1043 1195 Tax revenue 300 384 505 566 616 685 828 938 1089 Nontaxrevenue 31 65 67 53 73 93 102 104 106 Total Expenditure 413 446 644 730 863 1180 1275 1463 1949 Recurrent expenditure 346 416 519 544 680 820 989 1118 1426 Developmentexpenditureand 67 30 125 187 183 360 286 345 443 net lending 3verall balance (checks issued) -82 2 -72 -111 -174 -402 -345 -420 -754 3rants 67 47 115 119 170 292 293 383 506 ExpenditureFloat 0 0 0 0 0 0 -23 -34 0 3verall balance checks issued (after grants) -110 -75 -71 -248 4djustment to cash andother -20 -85 0 7 -45 -2 -4 35 0 items(net) 3verall balance (checks-cleared) -35 -36 43 15 -49 -112 -79 -36 -248 Financing: -35 -36 43 15 -49 -112 -79 -36 -248 ForeignFinancing, net /1 -24 -35 9 64 27 105 87 118 273 DomesticBorrowing 58 71 -53 -50 21 8 -2 -24 -24 PrivatisationFund 0 18 25 0 7 0 27 0 0 Bank & Parastatal 0 0 0 0 0 89 0 0 0 Recapitalization Changes inarreas 0 0 -34 0 0 0 -32 -59 0 Contigency 0 0 0 0 0 0 0 0 0 FinancingGap 0 0 0 0 0 0 0 35 0 4s % of GDP at market prices Total Revenue 12.5 13.2 L3.5 12.6 12.5 11.3 12.1 12.2 12.5 Tax revenue 11.3 11.3 L 1.9 11.5 11.1 10.0 10.7 11.0 11.4 Nontaxrevenue 1.2 1.9 1.6 1.1 1.3 1.3 1.3 1.2 1.1 Total Expenditure 15.5 13.1 15.2 14.8 15.6 17.1 16.6 17.1 20.4 Recurrentexpenditure 13.0 12.3 12.3 11.0 12.3 11.9 12.8 13.1 14.9 Developmentexpenditure and 2.5 0.9 3.0 3.8 3.3 5.2 3.7 4.0 4.6 part of net lending 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 3verall balance (checks issued) -3.1 0.1 -1.7 -2.3 -3.1 -5.8 -4.5 -4.9 -7.9 Srants 2.5 1.4 2.7 2.4 3.1 4.2 3.8 4.5 5.3 ExpenditureFloat 0.0 0.0 0.0 0.0 0.0 0.0 -0.3 -0.4 0.0 3verall balance checks issued -0.5 1.5 1.0 0.2 -0.1 -1.6 -1.0 -0.8 -2.6 [after grants) 4djustment to cash and other -0.8 -2.5 0.0 0.1 -0.8 0.0 0.0 0.4 0.0 items(net) 3verall balance (checks-cleared) -1.3 -1.0 1.o 0.3 -0.9 -1.6 -1.0 -0.4 -2.6 Financing: 1.3 1.0 -1.0 -0.3 0.9 1.6 1.o 0.4 2.6 ForeignFinancing, net/ l -0.9 -1.0 0.2 1.3 0.5 1.5 1.1 1.4 2.9 Domestic Borrowing 2.2 2.1 -1.3 -1.0 0.4 0.1 0.0 -0.3 -0.4 PrivatisationFund 0.0 0.5 0.6 0.0 0.1 0.0 0.3 0.0 0.0 Bank & ParastatalRecapitalization 1.3 0.0 0.0 0.0 Changes in arreas 0.0 0.0 -0.8 0.0 0.0 0.0 -0.4 -0.7 0.0 Contigency 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 FinancingGap 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.4 0.0 Note: Likely Outtum * I Source: The Tanzanian authorities. 177 Table A4a: Budget Frame Analytical(in Bill. Tsh.) - I 1994/95 1996197 1998199 2000101 2002/03 2004/05 2004/061 Actual Actual Actual Actual Actual Actual Actual Actual Actual Likely Proj. Proj. Domestic revenue 331 448 572 619 689 778 930 1043 1184 1393 1572 1774 OIW Fuel Levy and Transit Fees 0 0 34 37 38 38 40 50 64 62 66 71 Total Expenditure 413 446 644 730 863 1180 1275 1463 1729 2418 2518 2768 Recurrent expenditure 346 416 519 544 680 820 989 1118 1406 1611 1745 1982 Intereston external debt 22 39 38 48 60 47 52 57 66 63 96 107 Interest on domestic debt 25 27 74 53 30 81 78 65 49 72 70 73 Wages/salaries 110 156 199 219 232 285 308 342 411 464 533 602 Goodslservicesitransfers 189 194 209 224 358 406 551 655 801 1012 1046 1200 o/w Road Sector 16 16 20 17 38 38 55 50 61 62 66 71 Specialexp. 6 35 29 44 42 0 37 0 75 18 81 113 CFS (Others) 19 35 24 23 23 56 62 80 79 85 84 90 T W C M S A 0 0 16 17 20 22 0 0 34 Parastatal Wages 0 0 24 28 34 22 0 55 68 76 88 100 Retention Scheme 0 0 11 14 29 23 26 27 51 63 66 74 Election Costs 0 0 0 0 0 0 26 0 0 0 0 0 Census 0 0 0 0 0 0 0 0 10 0 0 0 Other Charges 168 153 85 82 171 216 409 443 646 708 660 753 Development expenditure 67 30 125 187 183 360 286 345 323 807 773 785 Projects 67 30 84 187 183 360 286 345 323 807 773 785 Local 20 5 20 24 19 19 35 50 96 140 106 118 Foreign 47 25 64 163 164 340 251 294 227 667 667 667 Other Programme Assistance 0 0 40 0 0 0 0 0 0 0 0 0 EnergyFundSongosongo 0 0 0 0 0 0 0 0 0 0 0 0 Overall deficit (checks issued)- -81 2 -72 -111 -173 -402 -345 -420 -545 -1025 -945 -994 beforegrants Grants 67 47 115 119 170 292 293 383 397 728 719 767 Budget supportgrants 35 30 67 2 69 73 114 183 190 260 265 302 oiw MDFPRBS 0 0 0 0 70 71 84 143 163 234 265 302 Basket Funds .. 82 124 project grants l / 33 17 49 118 101 208 124 140 74 241 339 339 HIPC interim relief-Multilateral 0 0 0 0 0 11 56 60 51 103 114 126 21 Expenditurefloat 0 0 0 0 0 0 -23 -34 -79 0 0 0 Overall deficit (checks issued)- -14 49 43 8 -4 -111 -75 -70 -148 -298 -226 -226 after grants Adjustment -20 -85 0 7 -45 -2 -4 35 0 0 0 0 Overalldeficit(checksc1eared) -34 -36 43 15 -49 -113 -79 -35 -228 -298 -226 -226 178 Table A4a: BudgetFrame Analytical (in Bill. Tsh.) (cont'd) - Financing 34 36 -43 15 49 113 79 35 228 298 226 226 Foreign (net) -24 -35 9 64 27 105 87 118 228 287 292 287 Programmmeloans 0 8 47 80 37 55 40 33 160 145 120 127 Project loan 14 8 15 57 63 133 128 154 153 303 328 328 Amortization -38 -50 -53 -72 -73 -82 -80 -69 -86 -161 -156 -167 Domestic (net) 58 71 -53 -50 21 8 -2 -24 0 10 -10 -10 Bank (net) 55 63 -48 -24 8 8 -19 -59 0 21 0 0 Non-bank 3 8 -5 -26 13 0 16 37 9 0 0 0 borrowing 6 - 6 4 2 7 4 5 0 0 0 2 0 0 0 0 amortization -3 -4 0 -53 -38 -234 0 -2 -10 -28 -10 -10 Recovery from NBC Bond 0 0 0 0 PrivatisationFunds 0 1 8 2 5 0 7 0 2 7 0 0 1 7 0 0 Change inArrears 0 0 -34 0 0 0 -32 -59 0 0 0 0 Bank Recapitalisation 0 0 0 0 0 89 0 0 0 0 0 0 Contigency 0 0 0 0 0 0 0 0 0 - 1 - 5 6 -51 Financing Gap 0 0 0 0 0 0 0 35 0 0 0 0 memo: CDPmp 2,660 3,394 4,236 4,929 5,532 6885 7703 8564 9493 10490 11623 12925 OC for distribution 0 0 0 0 0 299 489 637 764 927 962 1110 Primary Dejcit(check issued) -34 68 40 -10 -83 -274 -216 -390 -802 -891 779 -813 Government Saving(check -15 32 53 75 9 -42 -59 -157 -213 -218 -172 -208 issued % ofGDP -0.6 1.0 1.2 1.5 0.2 -0.6 -0.8 -1.9 -2.3 -2.1 -1.5 -1.6 Note: I/It has been assumed that 80% of projectgrants pass via the exchequer 21HIPC relieffrom IMF, World Bank and AfDB Source:TanzaniaAuthorities 179 Table A4b: BudgetFrame Analytica(as % of GDP) - 1994195 1995196 1996197 1997198 1998199 1999100 2000101 2001102 2002103 2003104 2004105 2004106 Actual Actual Actual Actual Actual Actual Actual Actual Likely Budget Proj. Proj. Outtum (In percent) lomestic revenue 12.5 13.2 13.5 12.6 12.5 11.3 12.1 12.2 12.5 13.3 13.5 13.7 OiW Road Fund 0.0 0.0 0.8 0.7 0.7 0.5 0.5 0.6 0.7 0.6 0.6 0.5 otal Expenditure 15.5 13.1 15.2 14.8 15.6 17.1 16.6 17.1 18.2 23.1 21.7 21.4 Recurrent expenditure 13.0 12.3 12.3 11.0 12.3 11.9 12.8 13.1 14.8 15.4 15.0 15.3 Intereston extemal debt 0.8 1.1 0.9 1.o 1.1 0.7 0.7 0.7 0.7 0.6 0.8 0.8 Intereston domestic debt 1.0 0.8 1.7 1.1 0.5 1.2 1.o 0.8 0.5 0.7 0.6 0.6 Wages1salaries 4.1 4.6 4.7 4.4 4.2 4.1 4.0 4.0 4.3 4.4 4.6 4.7 GoodsiservicesAransfers 7.1 5.7 4.9 4.5 6.5 5.9 7.2 7.6 8.4 9.6 9.0 9.3 oiw RoadFund 0.6 0.5 0.5 0.3 0.7 0.5 0.7 0.6 0.6 0.6 0.6 0.5 Specialexp. 0.2 1.o 1.o 0.7 0.9 0.8 0.0 0.5 0.0 0.8 0.2 0.7 0.9 CFS (Others) 0.7 0.6 0.5 0.4 0.8 0.8 0.9 0.8 0.8 0.7 0.7 TRA \ , 0.0 0.0 0.4 0.3 0.4 0.3 0.0 0.0 0.4 ParastatalWages 0.0 0.0 0.6 0.6 0.6 0.3 0.0 0.6 0.7 0.7 0.8 0.8 RetentionScheme 0.0 0.0 0.3 0.3 0.5 0.3 0.3 0.3 0.5 0.6 0.6 0.6 Other Charges 6.3 4.5 2.0 1.7 3.1 3.1 5.3 5.2 6.8 6.7 5.7 5.8 Development expenditure 2.5 0.9 2.9 3.8 3.3 5.2 3.7 4.0 3.4 7.7 6.7 6.1 Projects 2.5 0.9 2.0 3.8 3.3 5.2 3.7 4.0 3.4 1.7 6.1 6.1 Local 0.8 0.2 0.5 0.5 0.3 0.3 0.5 0.6 1.o 1.3 0.9 0.9 Foreign 1.8 0.7 1.5 3.3 3.0 4.9 3.3 3.4 2.4 6.4 5.7 5.2 Other ProgrammeAssistance 0.0 0.0 1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 lveralldeficit (checks issued) -before -3.1 0.1 -1.7 -2.3 -3.1 -5.8 -4.5 -4.9 -5.7 -9.8 -8.1 -1.7 rants rants 2.5 1.4 2.7 2.4 3.1 4.2 3.8 4.5 4.2 6.9 6.2 5.9 Programmegrants 1.3 0.9 1.6 0.0 1.3 1.1 1.5 2.1 2.0 2.5 2.3 2.3 01w PRBSiMDF 0.0 0.0 0.0 0.0 1.3 1.o 1.1 1.7 1.7 2.2 2.3 2.3 project grants 1.2 0.5 1.2 2.4 1.8 3.0 1.6 1.6 0.8 2.3 2.9 2.6 HIPC interimrelief 0.0 0.0 0.0 0.0 0.0 0.2 0.7 0.7 0.5 1.o 1.0 1.o lveralldeficit (checks issued)- after -0.5 1.4 1.o 0.2 -0.1 -1.6 -1.0 -0.8 -1.6 -2.8 -1.9 -1.7 rants idjustment -0.8 -2.5 0.0 0.1 -0.8 0.0 0.0 0.4 0.0 0.0 0.0 0.0 lveralldeficit (checks cleared) -1.3 -1.1 1.o 0.3 -0.9 -1.6 -1.0 -0.4 -2.4 -2.8 -1.9 -1.7 inancing 1.3 1.1 -1.0 0.3 0.9 1.6 1.0 0.4 2.4 2.8 1.9 1.7 Foreign -0.9 -1.0 0.2 1.3 0.5 1.5 1.1 1.4 2.4 2.7 2.5 2.2 importsupport loans 0.0 0.2 1.1 1.6 0.7 0.8 0.5 0.4 1.7 1.4 1.0 1.0 project loan 0.5 0.2 0.4 1.2 1.1 1.9 1.7 1.8 1.6 2.9 2.8 2.5 amortization -1.4 -1.5 -1.2 -1.5 -1.3 -1.2 -1.0 -0.8 -0.9 -1.5 -1.3 -1.3 Local (net) 2.2 2.1 -1.2 -1.0 0.4 0.1 0.0 -0.3 0.0 0.1 -0.1 -0.1 Bank (net) 2.1 1.8 -1.1 -0.5 0.1 0.1 -0.2 -0.7 0.0 0.2 0.0 0.0 Non-bank 0.1 0.2 -0.1 -0.5 0.2 0.0 0.2 0.4 0.1 0.0 0.0 0.0 borrowing 0.2 -0.2 0.1 0.6 0.8 0.0 0.0 0.0 0.2 0.0 0.0 0.0 amortization -0.1 -0.1 0.0 -1.1 -0.7 -3.4 0.0 0.0 -0.1 -0.3 -0.1 -0.1 PrivatisationFunds 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Change in Arrears 0.0 0.5 0.6 0.0 0.1 0.0 0.4 0.0 0.0 0.2 0.0 0.0 Bank Recapitalisation 0.0 0.0 -0.8 0.0 0.0 0.0 -0.4 -0.7 0.0 0.0 0.0 0.0 ontingency 0.0 0.0 0.0 0.0 0.0 1.3 0.0 0.0 0.0 0.0 0.0 0.0 vmo: 'DPmp(Bill. Tsh.) 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 rimaty Deficitlchecks issued) -1.3 2.0 0.9 -0.2 -1.5 -4.0 -2.8 -4.6 -8.4 -8.5 6.7 -6.3 'overnmentSaving(checks issued -0.6 1.o 1.2 1.5 0.2 -0.6 -0.8 -1.8 -2.2 -2.1 -1.5 -1.6 ource:Table 4a I 180 Table A5a: BudgetFrame Accounting(inBill. Tsh.) - - _ . 1994195 1995/96 1996197 1997198 1998199 1999100 2000101 2001102 2002103 2003104 2004105 2004106 Actual Actual Actual Actual Actual Actual Actual Actual Likely Outtum Budget Proj. Proj. ITOTAL RESOURCES 453 500 731 856 974 1,262 1,411 1,626 2,219 2,607 2,740 2,996 Domestic revenue 331 448 572 619 689 778 930 1,043 1,184 1,393 1,573 1,774 Programme loan and grants 35 38 114 82 106 128 154 216 350 405 385 429 Project loans and grants 47 25 64 174 164 340 251 294 309 473 667 667 Basket Support Loans 71 Basket Support Grants .. 124 Recovery from NCB Bond 0 0 0 0 HIPC Interim Relief Multilateral 0 0 0 0 0 11 56 60 51 103 114 126 HIPC interim relief-Paris Club 0 0 0 0 0 0 0 0 0 0 0 0 Non Bank Borrowing 6 -6 4 27 45 0 16 37 9 0 0 0 Bank Borrowing 55 63 -48 -24 8 8 -19 -59 0 21 0 0 Drawdown of reserves 0 0 0 0 0 0 0 0 0 0 0 0 Adjustment to cash -20 -85 0 -23 -45 -2 -4 35 0 0 0 0 Privatisation Funds 0 18 25 0 7 0 27 0 0 17 0 0 [. TOTAL EXPENDITURE 453 500 731 856 974 1,262 1,411 1,626 2,219 2,607 2,740 2,996 RECURRENT 387 470 606 670 791 902 1,125 1,282 1,406 1,800 1,911 2,159 .XPENDITURE 1,485 CFS 107 155 188 250 225 266 272 271 289 409 416 447 Debt service 88 120 164 227 202 211 210 192 210 324 332 357 Interest 48 66 112 101 91 128 130 121 115 135 166 180 amortization 41 54 53 126 111 82 80 70 95 189 166 177 Others 19 35 24 23 23 56 62 80 79 85 84 90 RecurrentExp.(excl. CFS) 280 315 384 420 566 636 821 917 1,117 1,391 1,495 1,712 01w Salaries & wages 110 156 199 219 232 285 308 342 411 464 533 602 Other Charges 168 153 85 82 171 269 409 506 801 909 880 997 DesignatedItems 11 0 0 100 119 163 135 145 133 75 18 81 113 Payment ofArrears 0 0 34 0 0 0 32 59 0 0 0 0 ExpenditureFloat 0 0 0 0 0 0 23 34 79 0 0 0 DEVELOPMENT 67 30 125 187 183 360 286 345 323 807 773 785 XPENDITURE Projects 67 30 84 187 183 360 286 345 323 807 773 785 Local 20 5 20 24 19 19 35 50 96 140 106 118 Foreign 47 25 64 163 164 340 251 294 227 667 667 667 Basket Funding 3/ 0 0 40 0 0 0 0 10 82 0 0 0 Note: 11Includescontingentexpenditures that will be voted at alater stage 21Fiscal risk provision 31From2002103 included inprogrammegrants Source:Ministry of Finance 181 N2:" r." 0 0 0 0 0 0 0 N b 0 V ) W b b N t - t - m O O b - - m N O - 0 0 0 0 0 0 ri N 0 ~ * t - o o o o w w o V ) o 111? 9 9 9 9 9 9 9 2" ? - 0 0 0 0 0 0 0 1 9 ? 9 9 9 N 9 9 N "b - 0 0 0 0 0 0 Do 0 Do 0 ? 9 ? ? 9 ? 9 o o o q o o o z ? 9 N N 9 9 n o o o q o o o 2 N 9 9 - 9 9 9 Do 0 0 0 0 0 0 0 0 1 \ 4 \ 4 0 U m - N x oy D o - 1 1 1 9 0 " O m 0 s 9 ? 0 1 1 1 1 - ? = ~ " 0 - C 36 p ! ? - . - ? p ! w p c o o m o : w - - q - , - 0 0 0 0 9 9 ? ? 9 - ? - - 9 ? 9 p ! ? ? ? m o o m o ~ o ~ o o m m o mN o o - o 'CI 393 c ? \ 4 P - ? N ~ WN Q 1'40'5 z u r - - CI 3 " ? 4 y c ? r l N O O " 3 r 9 -. 9 9 9 9 t : I - - 0 0 0 0 0 0 a 9 9 1 9 : P m e o 0 0 c .. a 09 9 9 9 9 f 09 9 y o 1 : u ~ a 0 0 0 0 0 0 0 - 0 0 P W O ? : : P m o o r : e oooooomoo 1 9 9 9 9 9 9 9 ' 9 0 1 : 0 0 0 ~ 0" : - 0 0 0 0 0 0 0 - c e.0 e, : w o o 000,eqo : m r - 0 0 0 0 - q 9 9 9 9 9 - 9 9 0.00 : U 0 0 0 0 0 0 0 0 0 0 0 r s9999-9r? a 0 0 0 0 0 0 0 0 * N.9999-99 o o o o o w o o '9990.9 r m o o o o c -T090- 0 0 p 0 0 9 9 T 9 1 r c - v - 9 9 9 9 r 0 0 0 0 0 : 1 9 1 9 9 " 9 " ! o o o o o m o o a I- u .-C 'ZYI s ? e 9 9 9 9 9 9 0 0 0 0 0 0 0 0 c?999CI 9 9 9 0 0 0 0 0 0 0 0 0.9 9 0 0 0 9 9 9 9 9 9 9 9 0 0 0 0 0 0 0 0 9 9 9 0 0 0 9 9 9 9 9 00'??9V! 9 9 9 0 0 0 0 0 w o -No-rw- m 0 0 0 00 3 9 9 9 9 9 w - - y 9 9 9 0 0 0 0 0 & O O e h 0 0 0 9 9 9 9 9 9 9 9 0 0 0 0 0 0 0 0 C r 9 9 9 Y 9 9 9 0 0 0 0 0 0 0 0 ? N 9 9 \ q 9 9 9 0 0 0 0 0 0 0 0 c?-99'? -9000: 0 0 0 0 0 0 0 0 0 E rL U Y V n Y 4 rableA6c: Sectoral TotalExpenditure(in Bill. Tsh.) - VOTE VOTE HOLDER 1996197 1996197 1997198 1997198 1998199 1998/99 1999100 1999100 2000/01 2000101 2001102 2001102 2002/031 Budget Actual Budget Actual Budget Actual Budget Actual Budget Actual Budget Actual Budget ADMINISTRATION 23.0 AccountantGeneral 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 0 23.6 23.1 100.1 90.5 26 0 27.0 Registrarof Political Parties 0.0 0.0 3.4 3.4 2.6 2.6 2.7 1 3 2.8 2.7 3.9 3.8 8 0 30.0 PresidentsOffice 9.8 9.7 10.9 10.6 15.7 15.5 23.2 22 9 35.1 29.9 40.2 38.3 48 9 !6 & 31 Vice President 0.5 0.5 2.7 1.o 3.3 1.7 7.0 2 6 9.7 1.9 6.9 2.6 20 9 32.0 Civil Service Dept. 2.1 2.1 3.8 2.6 4.0 3.7 7.3 117 11.2 11.1 14.4 13.9 160 33.0 Ethics Secretariat 0.1 0.1 0 2 0.2 0.2 0.2 0.2 0 1 0.2 0.2 0.3 0.3 0 3 34.0 ForeignAffairs 10.6 12.0 13.3 15.2 18.3 19.4 17.1 16 6 20.5 24 5 29.3 27.8 28 3 35.0 Perm Comm. Enq. 0.1 0.1 0.2 0.2 0.3 0.3 0.2 0 2 0.3 0.0 0.0 0 0 36.0 Civil Service Comm. 0.1 0.1 0.1 0.1 0.2 0.2 0.2 0 2 0.3 0.5 0.6 0.5 0 6 37.0 Prime MinistersOfice 7.3 7.1 26.5 11.8 14.2 6.1 8.1 7 7 9.2 5.4 13.3 8.6 18 4 40.0 Judiciary 3.6 3.6 4.6 4.5 5.4 5.3 7.9 6 0 9.6 9.6 124 12.0 12 8 41.O Justice 1.o 0.9 I .2 0.9 2.4 1.9 1.1 0 9 1.5 1.5 3.6 2.9 6 6 42.0 Office Of The Speaker 3.3 3.2 4.0 3.7 5.3 5.4 5.9 5 9 7.1 7.0 8.7 8.4 104 45.0 Exchequerand Audit 0.5 0.5 0.7 0.7 0.8 0.8 1.1 1 1 2.3 1.8 2.9 2.0 3 7 50.0 Finance 76.7 34.1 116.8 35.7 68.5 51.9 56.1 50 2 50.8 42.4 123.3 59.6 142 6 51.O HomeAffairs 1.6 1.6 2.3 1.7 4.6 4.6 5.8 4 8 5.6 5.6 6.9 5.9 9 0 54.0 RadioTanzania 0.3 0.3 1.1 1.1 1.2 1.1 1.6 1 4 1.9 1.5 2.0 1.7 3 4 55.0 Comm of HumanRights & .. 0.9 0.7 2 1 GoodGov. 56.0 Regional Adminisstrationand 6.2 2.7 22.7 167 31.0 17.0 90.8 78.7 99 6 Local Gov. 57.0 Defence & National. 3.5 1.7 3.4 5.2 5.5 3.1 3.3 2 3 5.1 5.1 4.4 4.4 117 59.0 Law Reform. Commi. 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0 1 0.2 0.2 0.7 0.3 0 5 60 0 IndustrialCourtOf Tz. 0.1 0.1 0.1 0.1 0.1 0.1 0.2 0 1 0.3 0.3 0.3 0.3 0 4 61.O ElectoralCommision 0.9 0.9 1.4 1.5 3.8 17.0 5 9 30.9 306 0.7 0.7 1 7 63.0 Local Govt. Servi. Comm. 0.1 0.1 0.1 0.1 0.2 0.2 0.2 0 2 0.3 0.3 0.4 0.4 0 6 64.0 CommercialCourt .. 0.7 0.5 1 1 66.0 PlanningCommission 2.0 1.1 8.3 2.4 4.1 3.2 3.2 2 9 4.8 4.5 16.6 13.9 33 9 90.0 Land Court 0 4 91.0 Drugs Control Commission 0 6 Sub Total 124.3 79.8 205.3 103.0 166.9 133.0 192.3 162.0 240.7 203.8 484.1 378.6 508.6 DEFENCEAND SECURITY 28 0 PoliceForce 24.6 24.0 29.1 25.2 29.7 29.7 34.9 32 2 42.0 41.9 43.4 42.2 54 5 29 0 Prison Services 12.7 12.3 17.4 15.1 16.4 16.4 19.4 162 20.9 20.8 23.4 23.1 31 4 38 0 Defence 63.8 63.5 72.5 66.7 73.2 81.7 78 5 89.2 89.2 100.0 99.7 1185 39 0 NationalService 7.5 7.5 9.5 8.9 11.4 11.8 12.1 I O 9 12.5 12.5 16.4 16.2 20 4 Sub Total 108.7 107.3 128.5 115.9 130.8 131.9 148.1 137.8 164.6 164.4 183.1 181.2 224.8 SOCIAL SERVICES 46.0 Education 16.2 15.3 24.8 24.4 30.4 29.8 37.7 32 1 50.8 48.4 56.4 43.9 I21 8 49.0 Water, 2.8 1.7 10.4 3.0 37.1 25.5 34.2 12 0 35.5 25.2 25.4 15.0 62 1 52.0 Health 22.8 16.3 51.7 31.1 58.7 48.7 57.0 42 2 74.3 65.4 91.6 78.3 117 1 53.0 Comm.Dev.Wome.A& 3.3 0.9 4.7 1.4 4.1 3.4 4.9 2 8 3.6 3.5 4.4 4.4 6 1 65.0 Labour Youth Develop. 3.6 3.7 5.7 2.7 3.7 2.3 5.1 2 2 6.2 3.3 6.0 6.0 7 8 67.0 Teachers Service Comm 0.1 0.1 0.1 0.1 0.1 0.1 0.2 0 2 1.5 5.7 1.8 1.7 1 8 68.0 Science, Tech.& HEd 24.2 21.0 28.1 21.8 37.5 34.8 37.4 32 2 49.1 44.7 60.1 58.6 70 2 '0 & 89 Regions 110.7 100.8 140.0 122.7 150.1 138.8 197.8 158 2 207.5 206.8 306.6 275.7 308 2 Sub Total 183.7 159.9 265.6 207.2 321.8 283.5 374.3 281.9 428.6 394.2 552.3 483.6 695.1 ECONOMICSERVICES 47.0 Works. 10.7 26.5 70.0 17.7 56.1 25.4 73.8 49 6 87.7 72.6 104.1 99.1 I87 6 48.0 Lands, Hous.,& Urb. Dev. 1.3 0.5 2.8 0.6 3.4 2.8 4.1 3 6 5.2 5.1 5.4 5.4 14 9 58.0 Energyand Minerals. 6.1 1.7 25.1 7.2 25.8 18.0 24.2 127 34.1 8.9 30.8 27.2 137 0 62.0 Comm.& Transport 6.3 6.2 32.7 30.5 22.6 22.2 20.7 21 9 32.8 32.8 19.6 18.8 48 2 Sub Total 24.3 34.8 130.6 56.0 107.9 68.4 122.8 87.9 159.9 119.5 159.9 150.4 387.7 PRODUCTIVE 24 0 Cooperatives & Marketing 5.5 5.5 7 1 43.0 Agriculture & Food Security 15 0 119 192 17.5 42.6 27.8 40.4 192 28.8 25.1 25.6 23.4 33 3 44 0 Industriesand Trade 2 0 1 4 3 3 3 2 5 0 4 7 4 8 3 6 4 5 4 4 7 0 6 4 8 1 188 Table A6c: Sectoral Total Expenditure (in Bill.Tsh.) - Icont'd) VO VOTE HOLDER 1996197 1996197 1997198 1997198 1998199 1998199 1999100 1999100 2000101 ZOO0101 2001102 2001102 20021031 TE Budget Actual Budget Actual Budget Actual Budget Actual Budget Actual Budget Actual Budget 69.0 Tour.,Nat.Res.& Env. 4.2 3.1 8.9 7.7 15.6 8.1 12.5 7.4 19.0 16.7 19.2 18.0 30.9 Sub Total 21.3 16.3 31.4 28.4 63.1 40.6 57.7 30.3 52.3 46.2 57.2 53.3 79.4 CONSOLIDATED FUhBSERVICE 20 0 Statehouse 0 9 0 8 0 8 0 8 0 9 0 9 I O 1 0 1 1 1 1 1 4 1 4 1 9 22 0 Public Debt 2177 2109 207 1 2532 2380 2322 2944 2861 2864 261 9 3375 275 1 321 6 Sub Total 218.6 211.7 207.9 254.1 238.9 233.1 295.3 281.7 287.5 263.0 338.9 276.5 323.5 GMhBTOTAL 680.8 609.9 969.3 Note * Datafor FY03 are from the ExpenditureFlashReports(July 2002 December 2002) -764.6 1029.5 890.4 1190.5 987.5 1333.7 1191.1 1775.6 1523.5 2219.2 Source MoF, AppropriationsAccounts and ExpenditureFlash Report 189 Table A7a: ActualRecurrentExpenditureby Vote as the Share of Total ActualRecurrent Expenditure VOTE VOTE HOLDER 1995196 1996197 1997198 1998199 1999100 2000101 20011021 ADMINISTRATION 23.0 Accountant General 0.00 0.00 0.00 0.00 0.00 2.17 7.01 27.0 Registrar o f PoliticalParties 0.00 0.00 0.50 0.33 0.15 0.26 0.30 30.0 PresidentsOffice 1.57 1.65 1.54 1.99 2.19 2.39 2.41 3 1.O 2nd Vice President 0.04 0.09 0.11 0.11 0.12 0.14 0.19 32.0 Civil Service Dept. 0.34 0.35 0.36 0.47 0.38 0.42 0.43 33.0 Ethics Secretariat 0.00 0.02 0.03 0.02 0.01 0.02 0.02 34.0 ForeignAffairs 3.12 2.09 2.23 2.48 1.88 2.37 2.16 35.0 Perm. Comm. Enq. 0.03 0.03 0.03 0.03 0.02 0.03 0.00 36.0 Civil Service Comm. 0.02 0.02 0.02 0.03 0.02 0.05 0.04 37.0 Prime Ministers Office 1.15 0.68 0.84 0.77 0.53 0.49 0.64 40.0 Judiciary 0.63 0.62 0.66 0.68 0.68 0.85 0.92 41.O Justice 0.11 0.13 0.13 0.25 0.10 0.15 0.18 42.0 OfficeOfThe Speaker 0.56 0.55 0.52 0.69 0.67 0.64 0.65 45.0 Exchequer and Audit 0.09 0.09 0.10 0.10 0.12 0.17 0.15 50.0 Finance 3.34 3.45 4.14 6.65 5.51 3.96 4.53 5 1.O HomeAffairs 0.29 0.27 0.25 0.60 0.55 0.54 0.46 54.0 Radio Tanzania 0.05 0.05 0.04 0.14 0.13 0.11 0.11 55.0 Comm of HumanRights & Good Gov. 0.05 56.0 Regional Adminisstration andLocal 0.35 1.89 0.37 1.58 Gov. 57.0 Defence & National. 0.26 0.27 0.25 0.39 0.26 0.40 0.26 59.0 Law Reform. Commi. 0.01 0.01 0.02 0.01 0.01 0.02 0.02 60.0 IndustrialCourt Of Tz. 0.01 0.02 0.02 0.02 0.02 0.03 0.02 61.O Electoral Commision 4.95 0.17 0.21 0.38 0.67 2.96 0.05 63.0 Local Govt. Servi. Comm. 0.03 0.02 0.02 0.02 0.02 0.02 0.03 64.0 Commercial Court 0.04 66.0 PlanningCommission 0.11 0.11 0.13 0.33 0.15 0.22 1.05 90.0 LandCourt 91.O Anti-DrugCommission Sub Total 16.71 10.69 12.16 16.85 16.11 16.61 23.29 DEFENCEAND SECURITY 28.0 PoliceForce 4.27 4.15 3.67 3.81 3.65 4.05 3.25 29.0 PrisonServices 2.49 2.14 2.21 2.11 1.84 2.02 1.79 38.0 Defence 11.14 11.11 9.79 9.48 8.91 8.63 7.74 39.0 National Service 1.27 1.31 1.31 1.51 1.24 1.21 1.26 Sub Total 19.17 18.71 16.98 16.91 15.64 15.91 14.04 SOCIALSERVICES 46.0 Education 2.47 2.10 2.14 2.50 2.34 2.77 2.79 49.0 Water, 0.38 0.26 0.37 0.82 0.38 0.52 0.69 52.0 Health 2.75 2.78 3.79 4.76 3.64 3.86 3.74 53.0 Comm.Dev.Wome. Aff 0.16 0.16 0.21 0.23 0.16 0.19 0.23 65.0 Labour Youth Develop. 0.30 0.22 0.22 0.26 0.17 0.30 0.33 67.0 Teachers Service Comm. 0.01 0.02 0.02 0.01 0.03 0.15 0.13 68.0 Science, Tech.& HEd 3.94 3.40 3.09 4.20 3.47 4.33 4.29 70 89 Regions - 18.43 17.30 17.21 16.54 17.95 20.02 19.61 Sub Total 28.45 26.23 27.66 29.33 28.15 32.14 31.82 ECONOMICSERVICES 47.0 Works. 1.23 4.08 2.60 0.81 3.30 5.42 3.80 48.0 Lands, Hous.,& Urb. Dev. 0.29 0.08 0.08 0.35 0.41 0.40 0.36 58.0 Energy and Minerals. 0.01 0.09 0.12 0.25 0.33 0.35 1.75 62.0 Comm.& Transport 0.10 0.40 0.53 1.66 1.35 1.01 0.87 Sub Total 1.63 4.65 3.34 3.08 5.38 7.19 6.781 190 Table A7a: Actual Recurrent Expenditure by Vote as the Share of Total ActualRecurrent Expenditure(cont'd) VOTE VOTE HOLDER 1995196 1996197 1997198 1998199 1999100 2000101 2001102 PRODUCTIVE ~~ ~ 24.0 Cooperatives & Marketing 0.42 43.0 Agriculture & Food Security 2.37 2.06 1.83 2.35 0.84 1.17 0.68 44.0 Industriesand Trade 0.47 0.23 0.42 0.59 0.38 0.40 0.40 69.0 Tour., Nat. Res. & Env. 1.28 0.39 0.34 1.02 0.84 1.12 1.10 Sub Total 4.11 2.69 2.59 3.96 2.07 2.69 2.60 CONSOLIDATEDFUNDSERVICE 20.0 State house 0.10 0.14 0.12 0.11 0.11 0.11 0.11 22.0 Public Debt 29.83 36.89 37.15 29.77 32.53 25.35 21.36 Sub Total 29.93 37.03 37.27 29.88 32.65 25.46 21.47 GRAND TOTAL 100.00 100.00 100.00 100.00 100.00 100.00 100.00 Source: Table 6a, 6b, and 6c 191 Table A7b: Actual DevelopmentExpenditureby Vote as the Share of Total Actual Development Cxpenditure VOTE VOTE HOLDER 1995196 1996197 1997198 1998199 1999100 2000101 2001102 ADMINISTRATION 23.0 Accountant General 0.00 0.00 0.00 0.00 0.00 0.42 0.08 27.0 Registrar of PoliticalParties 0.00 0.00 0.00 0.00 0.00 0.00 0.00 30.0 Presidents Office 0.79 0.54 0.12 0.00 3.39 3.28 3.06 3 1.O 2nd Vice President 0.00 0.00 0.33 0.74 1.51 0.30 0.08 32.0 Civil ServiceDept. 0.22 0.21 0.20 0.00 7.85 4.28 3.57 33.0 Ethics Secretariat 0.00 0.00 0.00 0.00 0.00 0.00 0.00 34.0 ForeignAffairs 0.00 0.00 0.00 0.00 0.00 0.00 0.00 35.0 Perm.Comm.Enq. 0.00 0.00 0.00 0.00 0.00 0.00 0.00 36.0 Civil Service Comm. 0.00 0.00 0.00 0.00 0.00 0.00 0.00 37.0 PrimeMinistersOffice 4.35 8.49 7.32 0.09 2.83 0.25 0.17 40.0 Judiciary 0.00 0.00 0.00 0.00 0.00 0.50 0.08 41.0 Justice 0.01 0.46 0.03 0.00 0.00 0.00 0.25 42.0 Office OfThe Speaker 0.15 0.20 0.21 0.00 0.00 0.30 0.00 45.0 Exchequer andAudit 0.00 0.00 0.00 0.00 0.00 0.03 0.04 50.0 Finance 3.54 37.70 9.00 0.00 1.53 0.94 0.54 51.O HomeAffairs 0.57 0.00 0.00 0.00 0.00 0.00 0.00 54.0 RadioTanzania 2.75 0.00 0.95 0.00 0.25 0.19 0.13 55.0 Comm o f Human Rights & GoodGov. 0.00 56.0 Regional Adminisstration and Local 0.00 0.04 8.25 24.75 Gov. 57.0 Defence&National. 6.46 0.21 4.30 0.00 0.00 0.63 0.42 59.0 Law Reform.Commi. 0.00 0.00 0.00 0.00 0.00 0.00 0.04 60.0 Industrial Court OfTz. 0.00 0.00 0.00 0.00 0.00 0.00 0.00 61.0 ElectoralCommision 0.00 0.00 0.00 0.00 0.00 0.00 0.00 63.0 Local Govt. Servi. Comm. 0.00 0.00 0.00 0.00 0.00 0.00 0.00 64.0 Commercial Court 0.00 0.00 66.0 PlanningCommission 2.07 1.17 1.75 0.55 1.42 1.37 0.17 90.0 Land Court 91.O Anti-Drug Commission Sub Total 20.90 48.99 24.21 1.39 18.82 20.74 33.40 DEFENCEAND SECURITY 28.0 PoliceForce 1.02 0.74 0.24 0.00 0.00 0.00 0.13 29.0 PrisonServices 0.28 0.15 0.00 0.00 0.00 0.00 0.00 38.0 Defence 0.00 0.00 0.00 0.00 0.00 0.00 0.00 39.0 National Service 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Sub Total 1.30 0.89 0.24 0.00 0.00 0.00 0.13 SOCIAL SERVICES 46.0 Education 4.67 8.73 6.87 9.29 10.79 12.49 3.35 49.0 Water, 0.65 0.53 0.60 17.32 8.07 6.91 2.59 52.0 Health 0.65 0.92 6.29 10.50 9.59 16.05 12.78 53.0 Comm. Dev.Wome. Aff. 0.52 0.00 0.00 1.39 1.26 0.92 0.64 65.0 LabourYouth Develop. 15.67 6.52 1.36 0.32 0.61 0.16 0.72 67.0 Teachers Service Comm 0.00 0.00 0.00 0.00 0.00 0.00 0.00 68.0 Science, Tech.& HEd 2.49 4.16 0.89 1.85 1.47 2.65 1.40 70 & 89 Regions 13.08 5.10 6.49 8.91 0.00 0.00 9.81 Sub Total 37.74 25.96 22.50 49.59 31.78 39.19 31.29 ECONOMIC SERVICES 47.0 Works. 23.09 8.33 0.00 17.31 19.38 10.50 21.27 48.0 Lands, Hous.,& Urb.Dev. 0.17 0.00 0.00 0.00 0.00 0.56 0.30 58.0 Energy and Minerals. 0.02 3.17 7.70 14.56 9.27 3.34 1.95 62.0 Comm.& Transport 0.01 10.18 32.38 8.33 9.41 14.10 3.23 Sub Total 23.29 21.67 40.08 40.21 38.06 28.50 26.75 192 Table A7b: Actual DevelopmentExpenditureby Vote as the Share of Total Actual Development Expenditure(cont'd) VOTE VOTE HOLDER 1995196 1996197 1997198 1998199 1999100 2000101 2001102 PRODUCTIVE 24.0 Cooperatives& Marketing 0.00 0.02 43.0 Agriculture & Food Security 5.93 0.28 6.12 8.60 11.09 8.19 6.24 44.0 Industriesand Trade 2.89 0.09 0.35 0.11 0.26 0.15 0.51 69.0 Tour., Nat. Res. & Env. 7.96 2.12 6.51 0.10 0.00 3.22 1.66 Sub Total 16.77 2.49 12.97 8.81 11.35 11.57 8.43 CONSOLIDATED FUNDSERVICE 20.0 Statehouse 0.00 0.00 0.00 0.00 0.00 0.00 0.00 22.0 Public Debt 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Sub Total 0.00 0.00 0.00 0.00 0.00 0.00 0.00 GRAND TOTAL 100.00 100.00 100.00 100.00 100.00 100.00 100.00 Source: Table 6a, 6b, and 6c 193 9 1 ? - . 9 9 9 9 9 ? ? b m - ~ - m m o m w m ~ m m - m - m t - m m m ~ m m - 008 V P ' m m c p ' r 4 b U m o m o o o - o o o o 0 o o m o o o m o o o o o o o " 1 9 1 - 9 - 1 9 9 9 9 9 9 :p' 1 0 : m 2 - b - N O N " 1 9 9 9 9 : w 1 0 0 0 0 0 0 0 : ~ - N o \ N: - - 0 0 0, c? 0 0 0 0 0 0 : z 0 : 2 0 w m w c 9 0 9 c " m r 3 Table A8a: Actual RecurrentExpenditureas a Percentageof BudgetedRecurrentExpenditureby r`ote VOTE VOTE HOLDER 1995196 1996197 1997198 1998199 1999100 2000101 2001102 ADMINISTRATION 23.0 Accountant General 100 91 27.0 Registrar o fPolitical Parties 100 100 50 99 98 30.0 Presidents Office 100 100 100 99 99 100 100 31.O 2ndVice President 91 100 91 100 100 100 100 32.0 Civil Service Dept. 88 101 101 100 100 100 95 33.0 Ethics Secretariat 100 99 98 83 100 113 34.0 ForeignAffairs 112 114 115 106 97 120 95 35.0 Perm.Comm. Enq. 93 100 100 99 95 100 36.0 Civil Service Comm. 83 98 111 100 100 150 90 37.0 PrimeMinisters Office 100 100 63 43 100 100 74 40.0 Judiciary 89 100 98 99 75 100 99 41.O Justice 77 100 106 90 86 99 96 42.0 OfficeOf The Speaker 102 99 99 101 100 100 97 45.0 Exchequer and Audit 97 99 99 97 100 99 100 50.0 Finance 64 96 84 102 98 100 66 5 1.O HomeAffairs 86 98 85 101 83 100 85 54.0 Radio Tanzania 79 100 100 93 96 74 82 55.0 Comm of HumanRights & GoodGov. 78 56.0 Regional Adminisstration and Local Gov. 44 315 100 99 57.0 Defence & National. 96 100 71 68 100 100 100 59.0 Law Reform. Commi. 61 103 102 99 94 99 115 60.0 Industrial Court OfTz. 35 100 101 100 88 100 87 61.0 ElectoralCommision 242 99 106 99 36 99 100 63.0 Local Govt. Servi. Comm. 59 100 100 100 99 97 112 64.0 Commercial Court 102 66.0 PlanningCommission 70 100 97 96 70 100 96 90.0 LandCourt 91.0 Anti-DrugCommission Sub Total 106 101 91 92 96 102 87 DEFENCEAND SECURITY 28.0 PoliceForce 101 100 87 100 92 100 97 29.0 PrisonServices 95 100 88 100 83 100 99 38.0 Defence 100 99 92 101 96 100 100 39.0 National Service 95 100 94 103 90 100 99 Sub Total 99 100 90 101 93 100 99 SOCIAL SERVICES 46.0 Education 100 99 110 100 98 100 96 49.0 Water, 81 100 90 100 100 91 94 52.0 Health 87 84 IO0 100 95 99 98 53.0 Comm.Dev.Wome.Aff. 92 100 100 103 62 100 100 65.0 Labour Youth Develop. 96 100 100 102 51 96 99 67.0 Teachers Service Comm. 76 100 98 100 100 100 97 68.0 Science, Tech.& HEd 96 91 92 100 95 100 99 70 & 89 Regions 91 99 99 107 94 100 94 Sub Total 92 96 99 104 94 99 95 ECONOMIC SERVICES 47.0 Works. 24 1004 73 128 76 100 95 48.0 Lands, Hous.,& Urb.Dev. 94 73 32 80 100 97 100 58.0 Energy andMinerals. 41 100 91 98 97 100 99 62.0 C o n " Transport 62 100 63 97 111 100 93 Sub Total 29 462 69 101 86 100 96 196 Table A8a: ActualRecurrentExpenditureas a Percentageof BudgetedRecurrentExpenditureby Vote (cont'd) VOTE VOTE HOLDER 1995196 1996197 1997198 1998199 1999100 2000101 2001102 PRODUCTIVE 24.0 Cooperatives & Marketing 98 43.0 Agriculture & Food Security 85 99 90 95 60 100 91 44.0 Industriesand Trade 120 99 100 99 93 100 89 69.0 Tour., Nat. Res. & Env. 98 99 82 101 85 86 96 Sub Total 92 99 90 97 74 94 94 CONSOLIDATEDFUNDSERVICE 20.0 State house 103 96 100 100 100 100 100 22.0 Public Debt 73 97 122 98 97 91 82 Sub Total 74 97 122 98 97 91 82 GRAND TOTAL 86 101 102 99 94 98 90 Source: Table 6a, 6b, and 6c 197 Table A8b: Actual DevelopmentExpenditure as a Percentageof BudgetedDevelopment Cxpenditure by Vote VOTE VOTE HOLDER 1995196 1996197 1997198 1998199 1999100 2000101 2001102 ADMINISTRATION 23.0 Accountant General 62 14 27.0 Registrar of Political Parties 30.0 Presidents Office 73 64 26 100 50 79 31.O 2nd Vice President 15 34 27 6 4 32.0 Civil Service Dept. 23 86 12 0 211 100 97 33.0 Ethics Secretariat 34.0 ForeignAffairs 0 0 35.0 Perm. Comm.Enq. 36.0 Civil Service Comm. 37.0 PrimeMinistersOffice 16 93 35 100 87 9 18 40.0 Judiciary 0 0 0 100 50 41.O Justice 25 95 8 2 52 42.0 Office Of The Speaker 100 69 37 99 45.0 Exchequer andAudit 9 10 50.0 Finance 2 26 9 0 25 15 4 5 1.O HomeAffairs 15 0 54.0 RadioTanzania 100 100 68 99 101 55.0 Comm o f HumanRights& GoodGov. 56.0 Regional Adminisstration andLocalGov. 0 48 83 57.0 Defence& National. 43 4 355 0 0 100 100 59.0 Law Reform. Commi. 20 60.0 IndustrialCourt OfTz. 61.O Electoral Commision 0 63.0 Local Govt. Servi. Comm. 64.0 Commercial Court 0 66.0 PlanningCommission 22 32 20 42 120 86 15 90.0 Land Court 91.O Anti-Drug Commission Sub Total 8 29 17 6 45 46 57 DEFENCEAND SECURITY 28.0 PoliceForce 23 33 55 100 29.0 Prison Services 11 12 0 38.0 Defence 0 39.0 NationalService Sub Total 19 25 34 100 SOCIALSERVICES 46.0 Education 9 81 73 94 69 90 42 49.0 Water, 1 15 7 62 28 37 38 52.0 Health 1 9 20 54 44 75 71 53.0 Comm. Dev.Wome. Aff. 9 0 0 65 51 90 98 65.0 LabourYouth Develop. 121 104 27 21 30 9 100 67.0 Teachers Service Comm. 68.0 Science, Tech.& HEd 8 56 14 44 31 100 78 70 & 89 Regions 12 18 25 34 0 63 Sub Total 12 35 25 54 31 66 61 ECONOMIC SERVICES 47.0 Works. 28 38 0 37 58 53 95 48.0 Lands, Hous.,& Urb. Dev. 3 0 0 0 98 100 58.0 EnergyandMinerals. 2 22 26 67 46 17 58 62.0 Comm.& Transport 0 97 100 100 100 100 100 Sub Total 20 44 34 53 60 53 92 198 Table ASb: ActualDevelopmentExpenditure as aPercentageof BudgetedDevelopment Expenditureby Vote (cont'd) VOTE VOTE HOLDER 1995196 1996197 1997198 1998199 1999100 2000101 2001102 PRODUCTIVE 24.0 Cooperatives & Marketing 43.0 Agriculture & FoodSecurity 100 3 95 41 42 78 92 44.0 Industriesand Trade 41 5 79 3 1 24 65 102 69.0 Tour., Nat. Res. & Env. 65 42 89 1 0 93 89 Sub Total 61 16 91 31 3 1 81 92 CONSOLIDATEDFUNDSERVICE 20.0 State house 22.0 Public Debt Sub Total GRANDTOTAL 14 32 28 46 42 58 61 Source: Table 6a, 6b, and 6c 199 Table ASc: Actual Total Expenditureas a Percentage of BudgetedTotal Expenditure ~yVote VOTE VOTE HOLDER 1995/96 1996/97 1997/98 1998/99 1999/00 2000/01 2001/02 ADMINISTRATION 23.0 Accountant General 98 90 27.0 Registrar of PoliticalParties 100 100 50 99 98 30.0 Presidents Office 99 99 97 99 99 85 95 31.O 2nd Vice President 91 100 38 51 38 19 38 32.0 Civil Service Dept. 82 100 69 91 160 100 96 33.0 Ethics Secretariat 100 99 98 83 I00 113 34.0 Foreign Affairs 112 113 114 106 97 120 95 35.0 Perm. Comm.Enq. 93 100 100 99 95 36.0 Civil Service Comm. 83 98 111 100 100 150 90 37.0 PrimeMinisters Ofice 57 97 45 43 95 59 64 40.0 Judiciary 89 99 98 99 75 100 97 41.O Justice 76 99 79 81 86 99 82 42.0 Office OfThe Speaker 102 98 92 101 100 100 97 45.0 Exchequerand Audit 97 99 99 97 100 78 70 50.0 Finance 26 44 31 76 89 83 48 5 1.O HomeAffairs 62 98 74 101 83 100 85 54.0 RadioTanzania 93 100 100 93 89 78 85 55.0 Comm of HumanRights & Good 78 Gov. 56.0 Regional Adminisstration and 44 73 55 87 Local Gov. 57.0 Defence& National. 59 48 155 56 70 100 100 59.0 Law Reform.Commi. 61 103 102 99 94 99 45 60.0 IndustrialCourt OfTz. 35 100 101 100 88 100 87 61.O ElectoralCommision 242 99 106 35 99 100 63.0 Local Govt. Servi. Comm. 59 100 100 100 99 97 112 64.0 CommercialCourt 72 66.0 PlanningCommission 35 52 28 77 90 92 83 90.0 LandCourt 91.0 Anti-Drug Commission Sub Total 66 64 50 80 84 85 78 DEFENCEAND SECURITY 28.0 PoliceForce 98 98 87 100 92 100 97 29.0 PrisonServices 92 97 87 100 83 100 99 38.0 Defence 100 99 92 96 100 100 39.0 NationalService 95 100 94 103 90 100 99 Sub Total 98 99 90 101 93 100 99 SOCIALSERVICES 46.0 Education 58 95 98 98 85 95 78 49.0 Water, 14 61 29 69 35 71 59 52.0 Health 53 71 60 83 74 88 85 53.0 Comm.Dev.Wome.Aff. 43 27 30 81 56 96 99 65.0 Labour Youth Develop. 112 102 46 64 43 54 99 67.0 Teachers Service Comm 76 100 98 100 100 377 97 68.0 Science, Tech.& HEd 75 87 78 93 86 91 97 0 & 89 Regions 77 91 88 92 80 100 90 Sub Total 68 87 78 88 75 92 88 200 Table A8c: Actual Total Expenditure as a Percentage ofBudgetedTotal Expenditure by Vote (cont'd) VOTE VOTE HOLDER 199996 1996/97 1997/98 1998/99 1999/00 2000/01 2001/02 ECONOMIC SERVICES 47.0 Works. 26 248 25 45 67 83 95 48.0 Lands, Hous.,& Urb. Dev. 57 35 20 80 88 97 100 58.0 Energy and Minerals. 22 28 29 70 53 26 88 62.0 Comm.& Transport 7 98 93 98 106 100 96 Sub Total 25 143 43 63 72 75 94 PRODUCTIVE 24.0 Cooperatives & Marketing 99 43.0 Agriculture & Food Security 86 79 91 65 47 87 92 44.0 Industries and Trade 86 68 98 95 76 97 91 69.0 Tour., Nat. Res. & Env. 88 73 87 52 59 88 94 Sub Total 87 77 91 64 52 88 93 CONSOLIDATED FUNDSERVICE 20.0 State house 103 96 100 100 100 100 100 22.0 Public Debt 73 97 122 98 97 91 82 Sub Total 74 97 122 98 97 91 82 GRAND TOTAL 71 90 79 86 83 89 86 Source: Table 6a, 6b, and 6c 201 rable A9a: Recurrent Expenditure Regions (in Bill. Tsh.) - VOTE VOTE 1996/97 1996197 1997/98 1997/98 1998/99 1998/99 1999/00 1999/00 2000/01 2000/01 2001/02 2001/02 2002/031 HOLDER Budget Actual Budget Actual Budget Actual Budget Actual Budget Actual Budget Actual Budget 70.0 Arusha 6.3 6.3 7.6 7.1 7.7 8.4 12.2 11.7 15.1 15.1 18.2 18.1 11.3 71.0 Coast 3.1 3.1 3.7 3.6 3.8 4.1 5.9 5.7 7.1 7.0 9.5 8.6 10.3 72.0 Dodoma 5.2 5.2 6.1 6.1 6.1 6.5 8.9 8.9 10.5 10.5 12.9 12.8 13.3 73.0 Iringa 5.6 5.3 6.6 6.5 6.8 7.3 7.3 2.3 11.4 11.3 13.7 13.6 14.9 74.0 Kigoma 3.7 3.7 4.4 4.4 4.7 4.9 5.1 6.4 7.9 7.9 9.8 9.8 10.1 75.0 Kilimanjaro 7.4 7.4 8.6 8.6 8.8 9.3 12.7 12.6 15.2 15.2 17.7 17.1 18.8 76.0 Lindi 3.0 3.0 3.5 3.5 3.6 3.9 5.1 5.1 6.4 6.3 7.8 7.8 8.5 77.0 Mara 5.0 5.0 6.1 5.9 6.1 6.7 8.6 8.6 11.0 11.0 13.6 13.4 12.7 78.0 Mbeya 6.3 6.3 8.0 7.9 8.2 8.5 11.5 11.4 13.4 13.3 16.3 16.3 17.4 79.0 Morogoro 5.6 5.6 6.7 6.6 6.8 7.4 9.9 9.9 11.5 11.5 13.8 13.8 14.9 80.0 Mtwara 3.7 3.7 4.3 4.2 4.4 4.7 6.4 6.4 7.7 7.7 9.2 9.0 9.6 81.0 Mwanza 6.9 7.0 8.4 5.8 8.4 9.2 12.3 12.2 13.9 13.9 17.7 17.6 18.6 82.0 Ruvuma 4.4 4.4 5.2 5.1 5.2 5.6 7.6 7.6 8.8 8.8 11.5 11.5 11.1 83.0 Shinyanga 5.4 5.5 6.4 5.9 6.6 7.3 7.3 9.6 11.3 11.3 14.5 14.5 16.1 84.0 Singida 3.6 3.6 4.1 4.1 4.2 4.4 5.9 5.8 7.0 7.0 8.8 8.3 9.2 85.0 Tabora 4.4 4.3 5.1 5.0 5.3 5.5 7.1 7.1 8.4 8.4 10.9 10.9 11.4 86.0 Tanga 5.9 5.9 7.0 6.9 7.1 7.5 10.1 10.1 12.0 12.0 14.5 14.4 15.4 87.0 Kagera 5.2 5.2 6.2 6.1 6.3 6.8 9.0 8.9 10.8 10.7 14.1 13.2 14.2 88.0 D'Salaam 5.6 5.6 6.4 6.4 6.7 6.8 10.0 2.7 11.5 11.4 27.4 13.6 17.2 89.0 Rukwa 3.2 3.2 3.8 3.5 4.0 4.1 5.6 5.5 6.4 6.4 8.1 8.1 8.6 95.0 Manyara 8.9 Total 99.4 99.1 118.3 113.2 120.9 129.0 168.6 158.2 207.5 206.8 270.0 252.4 272.51 rote:* Data for FY03 are fromthe ExpenditureFlashReports(July 2002 December2002) - Source: MoF, Appropriations Accounts andExpenditureFlashReport 202 rableA9b: DevelopmentExpenditure Regions(in Bill. Tsh.) - VOTE VOTE 1995/96 1996/97 1996/97 1997/98 1997/98 1998/99 1998/99 1999/00 1999/00 2000/01 2000/01 2001/02 2001/02 2002/03I HOLDER Actual Budget Actual Budget Actual Budget Actual Budget Actual Budget Actual Budget Actual Budget 70.0 Arusha 0.0 1.6 0.0 2.9 0.1 3.5 0.1 3.5 0.0 3.4 2.6 3.1 2.7 71.0 Coast 0.4 0.5 0.1 0.8 0.1 0.5 0.3 0.5 0.3 0.7 0.7 1.6 1.2 1.51 1.O 72.0 Dodoma 0.0 0.8 0.0 0.5 0.1 2.4 0.1 2.4 0.0 2.6 1.o 0.3 0.3 0.3 73.0 Iringa 0.0 0.9 0.0 1.8 0.1 2.8 1.3 2.8 2.3 5.0 0.4 3.6 3.4 4.4 74.0 Kigoma 0.2 0.5 0.1 0.3 0.1 1.4 0.0 1.4 0.0 0.2 0.2 0.4 0.4 0.3 75.0 Kilimanja 0.0 0.2 0.0 0.2 0.1 0.3 0.0 0.3 0.0 0.4 0.4 0.4 0.4 1.o ro 76.0 Lindi 0.0 0.3 0.0 0.8 0.1 0.4 0.1 0.4 0.0 1.3 0.3 1.8 0.3 0.8 77.0 Mara 0.3 0.5 0.3 0.7 1.4 1.7 2.6 1.7 1.5 3.5 1.4 4.5 1.7 1.1 78.0 Mbeya 0.1 0.4 0.0 0.4 0.1 0.4 0.1 0.4 0.0 0.7 0.5 0.6 0.6 2.3 79.0 Morogoro 0.1 0.3 0.2 1.7 0.3 0.6 0.1 0.6 0.0 3.2 0.3 3.3 0.3 3.3 80.0 Mtwara 0.0 0.5 0.0 0.8 0.1 0.6 0.1 0.6 0.0 1.3 0.3 1.3 0.5 1.4 81.0 Mwanza 0.2 0.8 0.6 1.3 0.7 1.8 0.8 1.8 0.6 2.2 2.2 2.9 2.3 5.0 82.0 Ruvuma 0.1 0.2 0.1 0.3 0.1 0.4 0.2 0.4 0.2 0.5 0.5 0.3 0.4 1.1 83.0 Shinyang 0.0 0.3 0.0 I.6 0.1 0.3 0.0 0.3 0.0 0.3 0.3 0.4 0.4 0.4 a 84.0 Singida 0.0 0.4 0.0 0.5 0.1 0.4 0.1 0.4 0.0 0.9 0.4 0.8 0.6 2.0 85.0 Tabora 0.0 0.2 0.0 0.2 0.1 0.2 0.1 0.2 0.0 0.3 0.3 0.4 0.4 0.4 86.0 Tanga 0.0 0.8 0.0 1.4 0.1 1.5 0.1 1.5 0.9 0.9 0.9 1.6 1.4 1.6 87.0 Kagera 0.2 0.7 0.1 2.2 1.2 5.6 3.5 5.6 3.8 9.3 7.7 8.4 4.9 5.2 88.0 D'Salaam 0.0 0.7 0.0 3.1 0.1 4.0 0.1 4.0 0.0 0.2 0.2 0.6 0.6 0.1 89.0 Rukwa 0.7 0.4 0.2 0.3 0.2 0.3 0.1 0.3 0.0 0.3 0.3 0.3 0.3 95.0 Manyara Total 2.4 10.9 1.9 21.7 5.4 29.2 iote: Datafor FY03 are fromthe Expenditure FlashReports (July 2002 December2002) * - 9.8 29.2 9.7 37.1 20.9 36.6 23.1 35.61 Source: MoF,Appropriations Accounts and Expenditure Flash Report 203 rable A9c: Total ExDenditure Regions (in Bill. Tsh.) - VOTE VOTE 1996197 1996197 1997198 1997/98 1998/99 1998199 1999/00 1999/00 2000/01 2000/01 2001/02 2001/02 2002/03 ~ HOLDER Budget Actual Budget Actual Budget Actual Budget Actual Budget Actual Budget Actual Budget 70.0 Arusha 7.9 6.3 10.5 7.2 11.2 8.5 15.7 11.7 18.6 17.8 21.3 20.8 12.8 71.0 Coast 3.6 3.2 4.5 3.8 4.3 4.4 6.4 6.0 7.8 7.7 11.1 9.8 11.3 72.0 Dodoma 6.0 5.2 6.6 6.2 8.5 6.6 11.4 8.9 13.1 11.5 13.2 13.1 13.6 73.0 Iringa 6.4 5.3 8.5 6.6 9.6 8.7 10.1 4.6 16.4 11.7 17.3 17.0 19.3 74.0 Kigoma 4.2 3.8 4.7 4.5 6.1 4.9 6.5 6.4 8.1 8.1 10.2 10.2 10.5 75.0 Kilimanjaro 7.6 7.4 8.8 8.7 9.0 9.3 13.0 12.6 15.6 15.6 18.1 17.5 19.8 76.0 Lindi 3.3 3.0 4.3 3.6 4.0 4.1 5.5 5.1 7.7 6.6 9.6 8.1 9.3 77.0 Mara 5.5 5.3 6.8 7.3 7.9 9.3 10.3 10.0 14.5 12.4 18.1 15.1 13.8 78.0 Mbeya 6.7 6.3 8.4 8.0 8.7 8.6 11.9 11.4 14.1 12.0 16.9 16.9 19.7 79.0 Morogoro 5.9 5.8 8.4 6.9 7.4 7.5 10.6 9.9 14.7 8.0 17.1 14.1 18.2 80.0 Mtwara 4.2 3.7 5.1 4.3 5.0 4.8 7.0 6.4 9.0 14.2 10.5 9.5 11.0 81.0 Mwanza 7.7 7.6 9.7 6.6 10.2 10.0 14.0 12.9 16.1 16.1 20.6 19.9 23.6 82.0 Ruvuma 4.6 4.5 5.5 5.2 5.6 5.8 8.0 7.7 9.3 9.3 11.8 11.9 12.3 83.0 Shinyanga 5.7 5.5 8.0 6.0 6.9 7.3 7.6 9.6 11.6 11.6 14.9 14.9 16.4 84.0 Singida 4.0 3.6 4.6 4.2 4.6 4.5 6.3 5.8 7.9 7.4 9.6 8.9 11.2 85.0 Tabora 4.6 4.4 5.3 5.1 5.6 5.6 7.4 7.1 8.8 8.8 11.3 11.3 11.7 86.0 Tanga 6.7 6.0 8.4 7.0 8.5 7.6 11.6 11.0 12.9 12.9 16.1 15.8 17.0 87.0 Kagera 5.8 5.3 8.4 7.3 11.9 10.3 14.6 12.6 20.1 18.4 22.5 18.1 19.4 88.0 D'Salaam 6.3 5.6 9.5 6.4 10.7 6.9 14.0 2.7 11.7 11.6 28.0 14.2 17.4 89.0 Rukwa 3.7 3.4 4.2 3.7 4.3 4.2 5.9 5.5 6.6 6.6 8.4 8.4 8.9 95.0 Manyara .. 11.0 Total 110.3 101.0 140.0 118.6 150.1 138.8 197.8 167.8 244.7 228.4 306.6 275.5 308.11 tote:* DataforFY03 are fromthe ExpenditureFlashReports (July 2002 December2002) - Source: MoF, AppropriationsAccounts andExpenditureFlashReport 204 Table AlOa: Actual Recurrent Expenditure as a PercentageofBudgetedRecurrent Expenditure - legions VOTE VOTE HOLDER 1995/96 1996/97 1997/98 1998/99 1999/00 2000/01 2001/02 (In percent) 70.0 Arusha 91 100 93 108 96 100 99 71.0 Coast 80 100 98 108 97 99 91 72.0 Dodoma 93 100 99 106 100 100 99 73.0 Iringa 78 95 99 108 32 100 99 74.0 Kigoma 87 100 99 104 124 100 100 75.0 Kilimanjaro 110 100 99 106 99 100 97 76.0 Lindi 46 100 99 109 99 99 100 77.0 Mara 94 100 98 110 100 100 99 78.0 Mbeya 97 99 99 104 99 99 100 79.0 Morogoro 93 100 99 110 99 100 100 80.0 Mtwara 91 100 99 105 100 100 98 81.0 Mwanza 90 102 70 109 100 100 99 82.0 Ruvuma 96 100 98 106 99 100 100 83.0 Shinyanga 91 100 92 112 132 100 100 84.0 Singida 93 100 99 106 99 99 94 85.0 Tabora 95 99 99 102 99 100 100 86.0 Tanga 92 100 98 106 99 100 99 87.0 Kagera 105 100 99 109 99 99 94 88.0 DSalaam 98 100 99 101 27 99 50 89.0 Rukwa 88 100 92 104 98 100 100 95.0 Manyara Total 91 100 96 107 94 100 93 ource: Table 9a. 205 Table AlOb: Actual DevelopmentExpenditureas aPercentageof BudgetedDevelopment Expenditure- tegions VOTE VOTE HOLDER 1995196 1996197 1997198 1998199 1999100 2000l01 2001l02 (In percent) 70.0 Arusha 1 1 4 3 1 77 87 71.0 Coast 93 18 19 64 56 100 73 72.0 Dodoma 1 1 23 4 0 39 92 73.0 Iringa 1 2 5 47 83 7 94 74.0 Kigoma 21 14 38 0 100 113 75.0 Kilimanjaro 1 17 58 2 100 102 76.0 Lindi 1 12 13 25 2 23 17 77.0 Mara 52 66 192 149 85 40 38 78.0 Mbeya 13 14 25 23 9 71 101 79.0 Morogoro 15 57 18 17 0 10 9 80.0 Mtwara 7 4 12 18 1 22 37 81.0 Mwanza 42 73 56 46 37 100 78 82.0 Ruvuma 24 67 34 58 41 100 139 83.0 Shinyanga 3 10 7 0 100 106 84.0 Singida 2 5 21 25 1 50 76 85.0 Tabora 4 8 41 44 2 100 108 86.0 Tanga 1 3 7 7 62 100 87 87.0 Kagera 8 18 56 62 67 83 58 88.0 DSalaam 2 3 2 2 0 100 100 89.0 Rukwa 46 40 46 36 2 100 98 95.0 Manyara Total 12 18 25 34 33 56 631 ource: Table 9b. 206 Table AlOc: Actual Total Expenditure as a Percentageof Budgeted Total Expenditure - tegions VOTE VOTE HOLDER 1995196 1996197 1997198 1998199 1999/00 2000/01 2001102 (In percent) 70.0 Arusha 78 80 69 75 75 96 98 71.0 Coast 82 88 84 103 93 99 88 72.0 Dodoma 74 87 93 77 79 88 99 73.0 Iringa 59 83 78 90 46 71 98 74.0 Kigoma 72 90 95 80 97 100 100 75.0 Kilimanjaro 100 98 99 103 97 100 97 76.0 Lindi 29 91 84 100 91 86 85 77.0 Mara 90 97 108 118 97 86 84 78.0 Mbeya 87 95 96 100 96 85 100 79.0 Morogoro 86 97 83 102 94 54 82 80.0 Mtwara 87 89 85 94 91 158 90 81.0 Mwanza 88 99 68 98 92 100 96 82.0 Ruvuma 87 99 94 103 96 100 101 83.0 Shinyanga 83 96 75 106 126 100 100 84.0 Singida 79 90 91 99 92 94 93 85.0 Tabora 88 94 97 100 96 100 100 86.0 Tanga 72 89 83 89 95 100 98 87.0 Kagera 26 91 87 86 87 92 80 88.0 D'Salaam 73 89 68 64 19 99 51 89.0 Rukwa 74 92 88 99 93 100 100 95.0 Manyara Total 77 92 85 92 85 93 90 ource: Table 9c. 207 rable A1la: Sectoral RecurrentExpenditureBefore and After Reallocation - VOTE VOTEHOLDER 1999/00 1999/00 1999/00 1999/00 1999/00 1999/00 i Approved Reallocation Budget After Actual Budget After Actual Expenditure Budget (in Bill.Tsh.) Reallocation Expenditure Reallocation as a % of Budgeted (in Bill. (in Bill. Tsh.) (in Bill. Tsh.) as a % of Expenditure After Tsh.) Approved Reallocation ADMINISTRATION 23.0 Accountant General 24.0 Cooperative and Marketing 47.0 Works. 38.0 38.0 4.6 100 12 61.0 ElectoralCommision 16.2 .o 16.2 5.9 100 36 50.0 Finance 128.4 4.5 132.9 48.5 104 37 27.0 Registrar of Political Parties 2.7 2.7 1.3 100 50 65.0 Labour Youth Develop. 2.8 .2 3.0 1.5 108 51 43.0 Agriculture 15.7 -3.4 12.3 7.4 78 60 53.0 Comm. Dev.Wome.Aff 2.2 .2 2.3 1.5 107 62 66.0 PlanningCommission 1.6 .3 1.9 1.3 120 70 40.0 Judiciary 7.9 .o 7.9 5.9 100 75 33.0 Ethics Secretariat .2 .o .2 .1 107 77 51.O HomeAffairs 5.4 .5 5.8 4.8 109 83 29.0 Prison Services 19.4 19.4 16.2 100 83 69.0 Tour., Nat.Res.& Env. 8.3 .4 8.7 7.4 105 85 41.0 Justice 1.o .1 1.1 .9 105 86 60.0 IndustrialCourt OfTz. .2 .2 .1 100 88 54.0 RadioTanzania 1.2 .o 1.2 1.1 102 89 39.0 National Service 10.1 2.0 12.1 10.9 119 90 28.0 Police Force 34.9 34.9 32.2 100 92 44.0 Industries and Trade 2.8 .8 3.6 3.4 130 93 59.0 Law Reform.Commi. ,1 .1 .1 100 94 52.0 Health 31.6 2.2 33.8 32.0 107 95 38.0 Defence 77.1 6.1 83.2 79.0 108 95 35.0 Perm.Comm.Enq. .2 .o .2 .2 107 95 68.0 Science, Tech.& HEd 30.9 1.6 32.5 31.0 105 96 58.0 Energyand Minerals. 2.2 .8 3.0 2.9 134 97 22.0 Public Debt 265.4 "28996 294.4 286.7 111 97 46.0 Education 18.0 3.0 21.0 20.7 116 98 30.0 PresidentsOfice 14.4 5.2 19.6 19.3 136 99 70 89 Regions - 134.1 41.1 175.2 172.9 131 99 63.0 Local GoW. Servi.Comm. .2 .o .2 .2 103 99 32.0 Civil Service Dept. 2.5 .9 3.4 3.4 137 100 67.0 TeachersService Comm. .2 .o .2 .2 116 100 42.0 Office Of The Speaker 5.2 .7 5.9 5.9 113 100 48.0 Lands, Hous.,& Urb. Dev. 3.6 ,1 3.6 3.6 103 100 36.0 Civil Service Comm. .2 .o .2 .2 126 100 57.0 Defence& National. 2.0 .3 2.3 2.3 114 100 26 & 31 Vice President .5 ,1 .7 .7 124 100 62.0 Comm.& Transport 10.7 1.2 11.9 11.9 111 100 49.0 Water, and Livestock 2.8 .6 3.4 3.4 121 100 Development 20.0 Statehouse .9 *I50 1.0 1.o 105 100 45.0 Exchequer and Audit 1.0 .1 1.1 1.1 106 100 25 & 37 PrimeMinistersOffice 2.5 2.2 4.7 4.7 187 100 56.0 MORALG 17.4 17.4 17.6 100 101 34.0 ForeignAffairs 11.3 '2,575 13.9 16.6 123 120 55.0 Commof HumanRights& Good Gov. 64.0 CommercialCourt I Note: * Retention Scheme Funds (Approved Estimates) *' Rea. no. TY/G/86/111/100 ** Special Requisition Source: MoF, Appropriations Accounts and Expenditure FlashReports for vote 38, 68, 73 & 75. 208 rableAllb: Sectoral Recurrent Expenditure Before and After Reallocation,FYOl - VOTE VOTE HOLDER 2000/01 2000/01 2000/01 2000/01 2000/01 2000/01 i Approved Reallocation Budget After Actual Budget After Actual Expenditure as a Budget (in Bill. Tsh.) Reallocation Expenditure Reallocation YOof Budgeted (in Bill. (in Bill. Tsh.) (inBill. Tsh.) as a YOof ExpenditureAfter Tsh.) Approved Reallocation Budget ADMINISTRATION 56.0 MORALG 14.7 -10.8 3.9 3.9 26 100 50.0 Finance 74.5 -33.5 41.0 40.9 55 100 66.0 PlanningCommission 4.1 -1.8 2.3 2.3 56 100 59.0 Law Reform Commi. .2 .o .2 .2 87 99 51.O Home Affairs 6.3 -.8 5.6 5.6 87 100 48.0 Lands, Hous.,& Urb.Dev. 4.5 -.2 4.3 4.2 95 97 36.0 Civil ServiceComm. .5 .o .5 .5 96 100 29.0 PrisonServices 21.6 -.7 20.9 20.8 97 100 33.0 EthicsSecretariat .2 .o .2 .2 97 100 22.0 Public Debt 292.6 -6.3 286.4 261.9 98 91 67.0 Teachers ServiceComm. 1.5 .o 1.5 1.5 98 100 69.0 Tour.,Nat. Res. & Env. 13.7 -.2 13.5 11.6 98 86 53.0 Comm.Dev.Wome.Aff. 2.0 .o 2.0 2.0 99 100 52.0 Health 40.5 -.3 40.1 39.9 99 99 54.0 RadioTanzania 1.6 .o.o .o 1.6 1.2 100 74 27.0 Registrarof Political Parties 2.8 2.8 2.7 100 99 20.0 Statehouse 1.1 1.1 1.1 100 100 68.0 Science, Tech.& HEd 44.8 .1 44.9 44.7 100 100 60.0 Industrial Court Of Tz. .3 .o .3 .3 101 100 44.0 IndustriesandTrade 4.1 .o 4.1 4.1 101 100 63.0 LocalGovt. Servi. Comm. .3 .o .3 .3 101 97 42.0 Office OfThe Speaker 6.4 .1 6.6 6.6 102 100 65.0 LabourYouth Develop. 3.1 .I 3.2 3.1 103 96 38.0 Defence 86.2 3.1 89.2 89.2 104 100 46.0 Education 27.6 1.0 28.7 28.6 104 100 70 89 Regions - 198.9 8.7 207.5 206.8 104 100 62.0 C o n " Transport 10.0 .5 10.5 10.4 105 100 39.0 National Service 11.9 .6 12.5 12.5 105 100 41.O Justice 1.4 .1 1.5 1.5 106 99 61.0 ElectoralCommision 29.1 1.7 30.9 30.6 106 99 45.0 ExchequerandAudit 1.7 .1 1.8 1.8 107 99 30.0 Presidents Office 22.7 2.0 24.7 24.1 109 100 35.0 Perm Comm. Enq. .3 .o .3 .3 109 100 28.0 PoliceForce 38.3 3.7 42.0 41.9 110 100 40.0 Judiciary 8.0 .8 8.8 8.8 110 100 25 & 37 Prime Ministers Office 4.5 .5 5.0 5.0 112 100 26 & 31 Vice President 1.2 .2 1.4 1.4 116 100 43.0 Agriculture 10.2 1.9 12.1 12.1 119 100 58.0 Energy andMinerals. 3.0 .6 3.7 3.7 121 100 32.0 Civil ServiceDept. 3.5 .9 4.3 4.3 125 100 49.0 Water, andLivestock 4.3 1.6 5.9 5.4 138 91 Development 23.0 Accountant General 16.2 6.3 22.5 22.4 139 100 34.0 ForeignAffairs 13.9 6.6 20.5 24.5 147 120 47.0 Works. 34.8 21.1 56.0 55.9 161 100 57.0 Defence& National. 2.5 1.6 4.1 4.1 162 100 24.0 CooperativeandMarketing 55.0 Comm of HumanRights & GoodGov. n Note: * Retention Scheme Funds (Approved Estimates) *' Req. no. TY/G/86/111/100 ** Special Requisition Source: MoF, AppropriationsAccounts andExpenditureFlashReportsfor vote 38,68,73 & 75 209 Table A1IC: Sectoral RecurrentExpenditure Before and After Reallocation,FY02 - VOTE VOTE HOLDER 2001/02 2001/02 2001/02 2001/02 2001/02 2001/02 Approved Reallocation BudgetAfter Actual BudgetAfter Actual Expenditure Budget (in Bill.Tsh.) Reallocation Expenditure Reallocationas a as a % of Budgeted (in Bill. (in Bill. Tsh.) (in Bill. Tsh.) % of Approved ExpenditureAfter Tsh.) Budget Reallocation ADMINISTRATION 50.0 Finance 280.9 (192.5) 88.4 58.3 31 25 & 37 PrimeMinistersOffice 5.8 5.3 11.1 8.2 192 55.0 Commof HumanRights 0.9 0.0 0.9 0.7 100 & GoodGov. 54.0 RadioTanzania 1.6 0.1 1.7 1.4 104 560.0 Home Affairs 1.O 6.9 0.0 6.9 5.9 100 IndustrialCourtOfTz. 0.3 0.0 0.3 0.3 100 44.0 Industries and Trade 5.0 0.8 5.8 5.2 115 36.0 Civil Service Comm. 0.6 0.0 0.6 0.5 100 43.0 Agriculture 9.2 0.4 9.6 8.7 105 23.0 Accountant General 19.0 19.7 98.7 90.3 519 62.0 Comm.& Transport 10.3 1.7 12.0 11.2 117 70 89 Regions - 221.6 48.4 270.0 252.6 122 49.0 Water, and Livestock 9.2 0.3 9.5 8.9 103 Development 32.0 Civil Service Dept. 4.0 1.8 5.8 5.5 146 34.0 Foreign Affairs 19.2 10.1 29.3 27.8 153 47.0 Works 48.2 3.4 51.6 49.0 107 69.0 Tour.,Nat. Res.& Env. 14.6 0.2 14.8 14.1 101 41.0 Justice 1.5 0.9 2.4 2.3 160 46.0 Education 34.4 3.0 37.4 36.0 109 66.0 PlanningCommission 14.0 0.0 14.0 13.5 100 42.0 Office OfThe Speaker 7.3 1.4 8.7 8.4 119 67.0 Teachers Service Comm. 1.7 0.0 1.8 1.7 102 28.0 Police Force 39.8 3.3 43.1 41.9 108 27.0 Registrar of Political 3.9 0.0 3.9 3.8 100 Parties 52.0 Health 46.4 2.7 49.1 48.2 106 24.0 Cooperativeand Marketing 40.0 Judiciary 11.5 0.5 12.0 11.8 104 29.0 Prison Services 22.5 0.9 23.4 23.1 104 39.0 National Service 13.9 2.5 16.4 16.2 118 68.0 Science, Tech.& HEd 52.7 3.2 55.9 55.3 106 65.0 Labour Youth Develop. 4.3 0.0 4.3 4.3 100 58.0 Energy and Minerals. 3.5 19.3 22.8 22.6 650 56.0 MORALG 20.5 0.0 20.5 20.4 100 22.0 Statehouse 1.4 0.0 1.4 1.4 100 38.0 Defence 96.6 3.4 100.0 99.7 104 45.0 Exchequer and Audit 1.9 0.0 1.9 1.9 98 30.0 PresidentsOffice 28.3 2.8 31.1 31.1 110 53.0 Comm. Dev.Wome. Aff. 2.6 0.3 2.9 2.9 110 26 & 31 Vice President 2.1 0.3 2.4 2.4 115 57.0 Defence&National. 2.9 0.5 3.4 3.4 117 61.0 ElectoralCommision 0.4 0.3 0.7 0.7 157 48.0 Lands, Hous.,& Urb. 4.6 0.1 4.7 4.7 102 Dev. 64.0 Commercial Court 0.5 0.0 0.5 0.5 100 63.0 Local Govt. Servi. 0.4 0.0 0.4 0.4 100 Comm. 33.0 Ethics Secretariat 0.3 0.0 0.3 0...3 101 113 59.0 Law Reform.Commi. 0.2 0.0 0.2 0.2 100 115 ote: RetentionScheme Funds (ApprovedEstimates) * *' Req.no.TYIGl86lll M O O ** SpecialRequisition Source:MoF, Appropriations Accounts and ExpenditureFlashReports for vote 38,68, 73 & 75 210 Table A12a: Regions Recurrent Expenditure Before and After - Reallocation, FYOO 1999100 1999100 1999/00 1999100 1999/00 1999100 VOTE VOTE Approved Reallocation Budget After Actual BudgetAfter Actual HOLDER Budget (in Bill. Tsh.) Reallocation Expenditure Reallocation as a Expenditureas (in Bill. (in Bill. Tsh.) (in Bill. YOof Approved a YOof Budgetes Tsh.) Tsh.) Budget Expenditure After Reallocation 70.0 Arusha 9.3 2.9 12.2 11.7 131 96 71.0 Coast 4.9 .9 5.9 5.7 119 97 72.0 Dodoma 6.5 2.4 8.9 8.9 137 100 73.0 Iringa 74.0 Kigoma 5.1 1.3 6.4 6.4 126 99 75.0 Kilimanjaro 9.7 3.0 12.7 12.6 131 99 76.0 Lindi 4.2 .9 5.1 5.1 123 99 77.0 Mara 6.4 2.2 8.6 8.6 135 100 78.0 Mbeya 8.8 2.6 11.5 11.4 130 99 79.0 Morogoro 7.6 2.3 9.9 9.9 131 99 80.0 Mtwara 5.1 1.3 6.4 6.4 126 100 81.0 Mwanza 9.1 3.2 12.3 12.2 135 100 82.0 Ruvuma 6.0 1.6 7.6 7.6 126 99 83.0 Shinyanga 7.3 2.7 10.0 9.6 137 96 84.0 Singida 4.8 1.1 5.9 5.8 123 99 85.0 Tabora 5.5 1.6 7.1 7.1 129 99 86.0 Tanga 7.9 2.2 10.1 10.1 129 99 87.0 Kagera 6.9 2.1 9.0 8.9 130 99 88.0 DSalaam 7.5 2.5 10.0 2.7 134 27 89.0 Rukwa 4.1 1.5 5.6 5.5 137 98 Total 126.8 38.6 165.3 155.8 130 94 Source:MoF, Appropriations Accounts 211 ITable A12b: Regions Recurrent Expenditure Before and After Reallocation,FYOl - 2000/01 2000/01 2000/01 2000/01 2000/01 2000/01 VOTE VOTE Approved Reallocation BudgetAfter Actual Budget After Actual Expenditure HOLDER Budget (in Bill. Tsh.) Reallocation Expenditure Reallocation as a as a YOof Budgeted (in Bill. (in Bill. Tsh.) (in Bill. Tsh.) YOofApproved Expenditure After Tsh.) Budget Reallocation 70.0 Arusha 14.3 .8 15.1 15.1 106 100 71.0 Coast 6.9 .2 7.1 7.0 103 99 72.0 Dodoma 10.1 .5 10.5 10.5 105 100 73.0 Iringa 11.1 .3 11.4 11.3 102 100 74.0 Kigoma 7.4 .4 7.9 7.9 106 100 75.0 Kilimanjaro 14.4 .9 15.2 15.2 106 100 76.0 Lindi 6.2 .2 6.4 6.3 103 99 77.0 Mara 9.9 1.1 11.0 11.0 112 100 78.0 Mbeya 13.3 .2 13.4 13.3 101 99 79.0 Morogoro 11.2 .3 11.5 11.5 103 100 80.0 Mtwara 7.5 .3 7.7 7.7 103 100 81.0 Mwanza 13.4 .5 13.9 13.9 104 100 82.0 Ruvuma 8.7 .1 8.8 8.8 102 100 83.0 Shinyanga 10.9 .4 11.3 11.3 103 100 84.0 Singida 6.9 .1 7.0 7.0 101 99 85.0 Tabora 8.3 .1 8.4 8.4 101 100 86.0 Tanga 11.6 .4 12.0 12.0 103 100 87.0 Kagera 10.1 .6 10.8 10.7 106 99 88.0 D'Salaam 10.4 1.1 11.5 11.4 111 99 89.0 Rukwa 6.2 .1 6.4 6.4 102 100 rota1 198.9 8.7 207.5 206.8 104 100 Source:MoF, Appropriations Accounts 212 ITable A12c: Regions RecurrentExpenditure Before and After Reallocation,FY02 - 2001/02 2001/02 2001/02 2001/02 2001/02 2001/02 VOTE VOTE Approved Reallocation Budget After Actual Budget After Actual HOLDER Budget (in Bill.Tsh.) Reallocation Expenditure Reallocation as a Expenditure as a (in Bill. Tsh.) (in Bill. Tsh.) (in Bill. Tsh.) YOof Approved YOof Budgeted Budget Expenditure After Reallocation 70.0 Arusha 15.9 2.3 18.2 18.1 115 99 71.0 Coast 8.9 0.6 9.5 8.6 107 91 72.0 Dodoma 11.1 1.8 12.9 12.8 116 99 73.0 Iringa 11.7 2.0 13.7 13.6 117 99 74.0 Kigoma 8.6 1.2 9.8 9.8 114 100 75.0 Kilimanjaro 15.7 2.0 17.7 17.1 113 97 76.0 Lindi 7.0 0.8 7.8 7.8 112 100 77.0 Mara 10.7 2.9 13.6 13.4 127 99 78.0 Mbeya 14.2 2.1 16.3 16.3 115 100 79.0 Morogoro 12.0 1.8 13.8 13.8 115 100 80.0 Mtwara 8.2 1.o 9.2 9.0 112 98 81.0 Mwanza 14.8 2.9 17'7 17.6 120 99 82.0 Ruvuma 9.9 1.6 11.5 11.5 116 100 83.0 Shinyanga 12.1 2.4 14.5 14.5 120 100 84.0 Singida 7.5 1.3 8.8 8.3 118 94 85.0 Tabora 9.6 1.3 10.9 10.9 114 100 86.0 Tanga 12.6 1.9 14.5 14.4 115 99 87.0 Kagera 12.4 1.7 14.1 13.2 114 94 88.0 D'Salaam 11.7 115.7 27.4 13.6 234 50 89.0 Rukwa 7.1 1.0 8.1 8.1 115 100 Total 221.6 48.4 270.0 252.4 122% 93% Source:MoF, Appropriations Accounts 213 - w IO 00 9 N ? b 00 z 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 ~ ~ 0 0 0 0 0 0 0 0 0 0 0 0 6 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 w 0 0 ~Nmo m I O w N d N w w w m m w 0 0 e N - w d m I O m h 1 0 0 w - 0 - 0 0 - Q \ I O IO 2 z - g 0 0 6 Q\ g ~ L ~c m w N O m N 0- N - N N w m - - 0 0 m m Q\ w 6 2m N O - 0 0 0 0 d y m N IO .- U w w w o o o o o o 0 0 W W d O m d m b 0 0 0 0 0 d d o m z d d w cn W 4 1 b - T W Z d d m d N 2 o o r 4 o o o o o o 0 0 - 0 0 0 0 2 0 0 O P P 0 0 O N W - m w v ) r 4 o w d m Z Z E d N ggz?;m" 2 Z c n S - P O O w O N m ~ C I 0 0 0 d z w a m P m m o o w m o o m o 0 0 O O N O 0 0 0 0 c n - m m Z OnO 0 0 " d P d 2 m a m c n -cm vw) N 0 0 0 0 0 0 0 0 0 0 0 0 N W W d w r 4 o w N m d - w m -