Report No. 24475 United Republic of Tanzania Public Expenditure Review FY02 Report on Fiscal Developments and Public Expenditure Management Issues (An external evaluation) May 2002 Government of Tanzania Tanzania PER Working Group The World Bank PREM 2 Africa Region uE Document of the World Bank GOVERNMENT FISCAL YEAR FY02 = 2001/02 = July 1, 2001 to June 30, 2002 CURRENCY EQUIVALENTS Currency Unit = Tanzanian Shilling (Tsh.) Interbank Market mid-rate: US$1.00 = Tsh. 890 (July 2001) ABBREVIATIONS AND ACRONYMS AfDB African Development Bank MOH Ministry of Health AIDS Acquired Immune Deficiency Syndrome MSTHE Ministry of Science, Technology and BEMP Basic Education Master Plan Higher Education BOP Balance of Payment MTEF Medium Term Expenditure Framework BOT Bank of Tanzania NBC National Bank of Commerce CBMS Cash Budget Management Systems NFA Net Foreign Assets CFAA Country Financial Accountability NGO non governmental organization Assessment O&E organizational and efficiency (reviews) CG Consultative Group O&M operations and maintenance CSD Civil Service Department OC other charges CSRP Civil Service Reform Programme PE personnel emoluments DANIDA Danish International Development PER Public Expenditure Review Agency PHC Primary Health Care DEO District Education Officer PORALG President's Office - Regional Authorities DFID Department for International Development and Local Government DMO District Medical Officer PRBS Poverty Reduction Budget Support DMT District, Municipal and Town Councils PRGF Poverty Reduction and Growth Faciltiy ESP Education Sector Programme PRSC Poverty Reduction Support Credit ESRF Economic and Social Research Foundation PRSP Poverty Reduction Strategy Paper EU European Union RAS Regional Administration Secretariat FAO Food and Agriculture Organization REPOA Research on Poverty Alleviation FY Fiscal Year SAC Structural Adjustment Credit GDP gross domestic product SASE Selective Accelerated Salary Enhancement GNP gross national product SASP Structural Adjustment Support Program GOT Government of Tanzania SDC Swiss Development Cooperation HIPC Highly Indebted Poor Countries Debt SDP Sector Development Program Initiative SPA Special Program of Assistance HIV Human Immuno-deficiency Virus SSA Sub-Saharan Africa IDA International Development Association STD Sexually Transmitted Disease IFEM Inter-bank Foreign Exchange Market TANESCO Tanzania Electric Supply Company IFMS Integrated Financial Management System TAS Tanzania Assistance Strategy IMG Independent Monitoring Group TB Treasury Bill IMF International Monetary Fund TRA Tanzania Revenue Authority LA Local Authority UDSM University of Dar es Salaam LGRP Local Government Reform Program UNDP United Nations Development Program LPO Local Purchase Order VAT Value Added Tax MAC Ministry of Agriculture and Cooperatives MOEC Ministry of Education and Culture MOF Ministry of Finance Vice President: Callisto Madavo Director: Judy O'Connor Sector Manager: Frederick Kilby Task Team Leader: Benno Ndulu TABLE OF CONTENTS PREFACE ......................................................... I EXECUTIVE SUMMARY OF FINDINGS AND RECOMMENDATIONS ............ II I. OVERALL FISCAL PERFORMANCE: KEY ISSUES ......................................................... II II. FISCAL RISKS .......................................................... VI III. EXTERNAL DEBT MANAGEMENT AND COMMITMENT CONTROL ............................. IX IV. EXPENDITURE AND SERVICE DELIVERY TRACKING . .XII............................ X V. SUMMARY OF KEY RECOMMENDATIONS ......................................................... XVII VI. GOVERNMENT RESPONSE TO PER FY01 RECOMMENDATIONS ......... ................... XIX 1. INTRODUCTION ......................................................... 1 1.1 THE PER PROCESS IN TANZANIA ......................................................... 1 1.2 THE EXTERNAL EVALUATION OF PUBLIC EXPENDITURES .........................................3 1.3 PUBLIC SECTOR AND PUBLIC FINANCE REFORMS IN TANZANIA .............. .................3 2. REVIEW OF FISCAL DEVELOPMENTS ......................................................... 5 2.1 INTRODUCTION .......................................................... 5 2.2 AGGREGATE FISCAL PERFORMANCE ......................................................... 5 2.3 STRATEGIC RESOURCE ALLOCATION ......................................................... 17 2.4 VARIATION BETWEEN THE APPROVED BUDGET AND ACTUAL EXPENDITURES ....... 24 2.5 MOVING FROM THE CASH BUDGET TO CASH FLOW MANAGEMENT ........... ............. 32 2.6 INTEGRATION OF FOREIGN AID INTo THE BUDGET PROCESS .................................. 39 2.7 ASSESSMENT OF BUDGET EXECUTION BY THE CONTROLLER AND AUDITOR GENERAL ...................................................................................................................................... 47 3. EXPENDITURE AND SERVICE DELIVERY TRACKING ........................... 57 3.1 THE TRACKING CHAIN ............................................................. 57 3.2 OVERVIEW OF EXISTING TRACKING SYSTEMS IN TANZANIA .................................. 58 3.3 ALLOCATION AND TRANSFER MECHANISMS .......................................................... 59 3.4 SERVICE DELIVERY MECHANISMS .............................................................. 60 3.5 FINANCIAL AND SERVICE DELIVERY TRACKING IN PRACTICE ................................ 61 3.6. THE ROLE OF THE PER IN TANZANIA'S POVERTY MONITORING SYSTEM .. 66 3.7 RECOMMENDATIONS .............................................................. 68 4. FISCAL RISKS .............................................................. 70 4.1 MOTIVATION ............................................................. 70 4.2 SOURCES OF FISCAL RISKS IN TANZANIA'S PUBLIC SECTOR .............. .................... 70 4.3 CONTAINMENT OF FISCAL RISKS AND AGENDA FOR FURTHER WORK ......... ........... 74 5. EXTERNAL DEBT MANAGEMENT AND COMMITMENT CONTROL... 76 5.1 INTRODUCTION ........................................................... 76 5.2 SUMMARY OF FINDINGS ........................................................... 77 5.3 OVERVIEW OF TANZANIA'S DEBT .......................... ................................. 81 5.4 PERCEIVED RISKS OF VARIOUS DEBT CATEGORIES ................................................ 82 5.5 PERCEIVED RISKS OF INSTITUTIONAL ROLES AND IMPLEMENTATION ..................... 89 ANNEXES ......................................................................................... 98 ANNEX 1: REALLOCATIONS FROM AND TO THE MINISTRY OF FINANCE CONTINGENCY POSITION ......................................................... 99 ANNEX 2: DEBT RELIEF AND FISCAL DEFICITS ......................................................... 100 ANNEX 3: DEFINITION OF POVERTY-REDUCING EXPENDITURES IN THE BUDGET ....... 102 ANNEX 4: COMPOSITION OF SPECIAL EXPENDITURE ITEM UNDER VOTE 50 (FY00 - FY02) ......................................................... 103 ANNEX 5: TANZANIA DEBT MANAGEMENT AS ASSESSED BY THE jOINT IMF/IDA TEAM FOR HIPC COMPLETION POINT ......................................................... 104 ANNEX 6: QUESTIONNAIRE ON EXTERNAL DEBT MANAGEMENT IN HIPCs ................1 05 ANNEX 7: PUBLIC EXPENDITURE REVIEWS AND BUDGET PROCESS IN TANZANIA AND UGANDA ................................................. 124 ANNEX 8: PER FY02 - STATUS OF IMPLEMENTATION OF THE WORK PROGRAM (MAY 29, 2002) ........................................................... 143 ANNEX 9: AGENDA OF THE PER CONSULTATIVE MEETING, MAY 9-10, 2002 ........... 144 DATA ANNEX .......................................................... 146 List of Tables TABLE 2-1: FINANCING OF THE FISCAL DEFICIT (% OF GDP), FY96 - FY02. 6 TABLE 2-2: GOVERNMENT REVENUE (AS % OF GDP), FY96 - FY02 .8 TABLE 2-3: FOREIGN INFLOWS - GRANTS AND LOANS (AS % OF GDP), FY96 - FY01 .11 TABLE 2-4: HIPC DEBT RELIEF AS % OF GDP, FY00 - FY02 .12 TABLE 2-5: GOVERNMENT EXPENDITURES AS % OF GDP, FY96 - FY02 .14 TABLE 2-6: CIVIL SERVICE EMPLOYMENT, FY98-FYOI (DECEMBER OF EACH YEAR .15 TABLE 2-7: CIVIL SERVICE AVERAGE SALARIES, FY98-FYO1 .16 TABLE 2-8: COMPOSITION OF PUBLIC EXPENDITURES AS % OF GDP, FY96-FY02 .18 TABLE 2-9: SECTORAL RECURRENT EXPENDITURES (ACTUALS, AS A %AGE OF GDP ) .... 19 TABLE 2-10: SOCIAL SECTOR RECURRENT EXPENDITURES (ACTUALS, AS A %AGE OF GDP) .....................I...................................................................................................... I ...... 20 TABLE 2-11: SECTORAL DEVELOPMENT EXPENDITURES .................................................. 21 TABLE 2-12: EXPENDITURES FOR PRIORITY SECTORS, FY00 - FY02 ............................... 23 TABLE 2-13: CENTRAL GOVERNMENT OPERATIONS, ACTUAL AS A PERCENTAGE OF BUDGETED,FY97-FYO1 ....................................................... : 25 TABLE 2-14: CONTINGENCY ALLOCATION RETAINED BY MINISTRY OF FINANCE .............. 29 TABLE 2-15: REALLOCATIONS AND EXPENDITURE OUTrURN, FY01 ................................ 30 TABLE 2-16: DEVELOPMENT EXPENDITURES BY SECTOR, ACTUAL EXPENDITURES AS A SHARE OF BUDGETED EXPENDITURES, FY96-FYOO ................................... 31 TABLE 2-17: SHARE OF TECHNICAL COOPERATION AND INVESTMENT PROJECT ASSISTANCE CAPTURED IN THE BUDGET, FY01 ............................................. 44 TABLE 2-18: SUBMISSION OF ACCOUNTS AND PRODUCTION OF CAG REPORT, FY97 - FY00 ............................................ 49 TABLE 2-19: SUMMARY OF AUDIT CERTIFICATES ISSUED, FY96-FYOO ....................... 53 TABLE 2-20: SUMMARY OF QUERIES ISSUED AND PERCENTAGES REPLIED AND NOT REPLIED, FY96-FYOO ............................................................. 53 TABLE 2-21: UNVOUCHED AND IMPROPERLY-VOUCHED EXPENDITURES, FY96-FYOO . ..54 TABLE 2-22: CASH AND STORES LOSSES (IN TSH. MILLION), FY96-FYOO .54 TABLE 2-23: SUMMARY OF EXCESS SPENDING AND SAVING, FY96-FYOO .55 TABLE 2-24: ACCOUNTING FOR REVENUES - LOCAL AUTHORITIES, FY98-FYOO 56 TABLE 2-25: QUESTIONABLE PAYMENTS - LOCAL AUTHORITIES, FY99-FYOO 56 TABLE 3- 1: MONITORING MECHANISMS COVERING OUTCOMES, OUTPUTS, PROCESSES, AND INPUTS ............................................................. 67 TABLE 4- 1: THE FISCAL RISKS MATRIX ............................................................. 71 TABLE 4-2: SUMMARY OF CONSOLIDATED ARREARS CLEARED FOR PAYMENT BY CATEGORY JULY 1998 - DECEMBER 2000 (% SHARE ................................... 74 TABLE 5- 1: SELECTED DEBT INDICATORS FOR TANZANIA, UGANDA, KENYA, SUB- SAHARAN AFRICA, LATIN AMERICA, AND ASIA, 2000 .................................. 81 TABLE 5-2: TANZANIAN EXTERNAL DEBT OUTSTANDING, END-JUNE 2001 ..................... 82 TABLE 5-3: STATUS OF COMMERCIAL AND NON-PARIS CLuB LOANS .............................. 85 TABLE 5-4: ASSESSMENT OF NATIONAL DEBT STRATEGY VIS-A-VIS BEST PRACTICE AND OBSERVED PRACTICE ................................................... 91 List of Figures FIGURE 1-1: PER ACTIVmES AND THE BUDGET PROCESS .......................... ; ...................... 2 FIGuRE 2-1: KEY FISCAL DEVELOPMENTS, FY95-FY02 .................................................... 7 FIGURE 2-2: REAL GROWTH OF DOMESTIC REVENUE, FY95-FYO1 ................................... 9 FIGURE 2-3: FOREIGN DEVELOPMENT ASSISTANCE TO TANZANIA, FY95-FYO0 I............. 12 FIGURE 24: INCREMENTAL SPENDING ON PRIORITY AREAS COMPARED TO FY99 AND HIPC RESOURCES (IN PERCENT OF GDP) ............................ 24 FIGURE 2-5: COMPARISON OF ACTUAL AND POTENTIAL FISCAL BALANCE (AFTER GRANTS) .................................................................................................................................. 26 FIGURE 2-6: DEVIATION FROM BUDGET (AS A % OF GDP) OF FUNDS AVAILABLE AND ACTUAL EXPENDITURES ON O&M AND DEVELOPMENT .............................. 28 FIGURE 2-7: MONTHLY EXCHEQUER RELEASES AS A PERCENTAGE OF AVERAGE MONTHLY EXCHEQUER RELELEASES FOR A GIVEN YEAR, FY97-FYO1 ........................ 32 FIGURE 2-8: MONTHLY EXCHEQUER RELEASES AS A PERCENTAGE OF AVERAGE MONTHLY EXCHEQUER RELEASES FOR A GIVEN YEAR, FY98-FYO1 ........................... 33 FIGURE 2-9: AVERAGE ABSOLUTE MONTHLY DEVIATION FROM MEAN OF EXPENDITURE RELEASES FOR O&M ............... 34 FIGURE 2-10: MONTHLY CASH REQUIREMENTS FOR OC ACCORDING TO CASH FLOW PLANS, FY02 ............... 36 FIGURE 2-11: CASH FLOW REQUIREMENTS FOR OC FOR FY02 AND MONTHLY RELEASES FOR OC FOR FY00 AND FY01 ............... 38 FIGURE 2-12: OFFICIAL DEVELOPMENT ASSISTANCE TO SELECTED EAST AFRICAN COUNTRIES, 1999 ............... 39 FIGURE 2-13: OFFICIAL DEVELOPMENT ASSISTANCE BY TYPE OF ASSISTANCE, 1999 ............ 43 FIGURE 2-14: PROGRAM AND PROJECT SUPPORT AS A SHARE OF GDP, LOANS AND GRANTS, 1995-1999 ................................................................................ 45 FIGURE 2-15: DEVELOPMENT EXPENDITURE CAPTURED IN APPROPRIATIONS ACCOUNTS AS A SHARE OF DEVELOPMENT EXPENDITURE RECORDED IN THE BUDGET, FY96-FYO 1 ..................................................................................................................... 46 FIGURE 2-16: CAPTURING OF FOREIGN AID IN THE BUDGET ................................................... 47 List of Boxes Box 1: The recurrent deficit as a fiscal target ...................................................14 Box.2: Revenue Performance and GDP Growth Estimates .................................... 19 Box 3: Implications of expanded program support for fiscal management .................. 22 Box 4: Identifying Priority Expenditures in Tanzania .......................................... 31 Box 5: Audit Certificates ...................... 55 PREFACE Starting in 1997/98, Tanzania has embarked on an annual Public Expenditure Review (PER) process with the interrelated twin objectives of supporting the budget process and conducting an external review of fiscal developments. The Tanzania PER Working Group comprising representatives of the Government of Tanzania, the World Bank, other 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 of the agreed work program, and reviews all outputs. It also represents an important forum for discussion of public expenditure issues between government and a wide array of interested stakeholders in Tanzania. This report presents the findings of the external review of fiscal developments and public expenditure management issues, which previously have been shared with government and members of the PER working group as well as with a wider audience at the annual PER Consultative Meeting, which was held in May 2002 prior to the finalization of the Tanzanian government budget. The report is primarily based on the findings of a joint donor mission in November/December 2001 which was led by Benno Ndulu (mission leader, AFTP2) and consisted of Robert Utz, Sumana Dhar, Alema Siddiky (AFTP2), Philip Mpango, Hamisi Mwinyimvua (consultants, University of Dar es Salaam), Ben Tarimo (consultant), Jytte Laursen (Danish Embassy); Fiona Shera (DFID), Gerard Considine (Ireland Aid), Willem Bronkhorst (Netherlands Ernbassy), and Philip Courtnadge (UNDP). Frederick Kilby, (Sector Manager, AFTP2), also joined the mission from December 10 to 14. The report was written under the supervision of Fred Kilby, Sector Manager, AFTP2 and Peniel Lyimo, Deputy Permanent Secretary, Ministry of Finance, Government of Tanzania. David Bevan (Oxford University), Allister Moon (ECSPE) and Anand Rajaram (PRMPS) served as peer reviewers. Emmanuel Munganasi provided excellent research assistance and compiled the statistical appendix. Patrick Mamboleo was responsible for the word processing and the physical production of the report. EXECUTIVE SUMMARY OF FINDINGS AND RECOMMENDATIONS I. OVERALL FISCAL PERFORMANCE: KEY ISSUES Fiscal Stability 1. Overall fiscal discipline, defined in terms of recurrent deficit targets, has remained a key policy objective. After targeting and achieving recurrent surpluses during the period FY97 - FY99, the targets for FY00 and FY01 were set to provide scope for increased financing of priority sector activities under the Poverty Reduction Strategy and to accommodate increased foreign inflows in the form of program grants. The recurrent deficit in FY00 and FY01 was 0.5 percent and 1.2 percent of GDP, respectively, and the target for FY02 is a recurrent deficit of 2.5 percent of GDP. 2. Under conditions of sustained economic growth and continued high inflows of foreign assistance, especially in the formn of program support, even the relaxed deficits are well within the sustainability thresholds indicated by recent analyses of fiscal and debt sustainability and remain compatible with continued macro-economic stability. 3. The cash budget in conjunction with an increasingly sophisticated commitment control system remains the main instrument for achieving fiscal balance targets. However, in order to soften the negative impact of a cash budget system on allocative and operational efficiency, the government has adopted a number of important measures. These include the introduction of quarterly budget releases to the priority sectors since December 2000 and the introduction of cash flow planning to synchronize monthly and quarterly exchequer releases with cash requirements by the spending units. 4. With respect to the roll-out of quarterly exchequer releases, spending units report that quarterly reports have indeed significantly improved ability to implement their work programs in a rational and efficient way. It is recommnended that the quarterly release system for priority sectors is rolled out to all other spending units. The risk of piloting a general quarterly release system is limited by the provisions in the new public finance act that intra-vote reallocations between budget items require approval by the Ministry of Finance.' If a system of general quarterly releases were to be adopted, it would also be important to set appropriate incentives for spending units to act responsibly such as I Limiting the flexibility of vote holders to reallocate within votes reduces the scope for reallocating funds to items where the commitment control system is stringent and leaving items which are not full covered by the commitment control system, such as utilities, under-allocated which in consequence leads to the accumulation of arrears. making quarterly releases contingent on good public expenditure management and reporting practices in the spending units. 5. The introduction of cash flow planning is in its early stages. The quality of the cash flow plans by the spending units is of varying quality. At the aggregate level, exchequer releases according to cash flow plans would have important implications for the functioning of the cash budget system and the use of bridge financing mechanisms through the central bank. During the first nine months of FY02 recourse to central bank bridge financing was not necessary as sufficient funds were available from program support to fully cover requirements. However, the Ministry of Finance is encouraged to develop a strategy that would adjust the cash budget system so as to allow for accommodation of cash flow plans while limiting the risks to fiscal sustainability. 6. Another innovation introduced with the new public finance act is the requirement that all intra-vote reallocations between budget items have to be approved by the Ministry of Finance. Considered in the context of the aforementioned modifications of the cash budget system, this measure is expected to strengthen public expenditure management in two ways. Firstly, this requirement creates strong incentives for spending units to improve the quality of their budget and thus provide the Ministry of Finance and other stakeholders with improved and more reliable information on which allocative decisions can be made. The second positive impact is that this measure provides certain safeguards in the process of softening the cash budget system. At the same time, pre-approval of intra-vote reallocations impinges on the flexibility of vote holders to manage their budgets. However, on balance the benefits arising from improved budgeting at the sectoral level and the relaxation of the cash budget which is made possible through this measure, at present appear to outweigh the possible cost arising from decreased flexibility at the level of the spending unit. To minimize these costs it will also be important to monitor closely how requests for reallocations are handled by the Ministry of Finance. It would also be desirable if the Ministry of Finance prepared a summary report at the end of the fiscal year which would provide summary information on the reallocations requested, approved, and denied and the average duration of the process from request to a decision by the Ministry of Finance. Government Resources 7. Even though revenue collection as a percentage of GDP increased during FYO1 to 12.2 percent of GDP on account of the introduction of VAT on petrol imports and (transitory) higher oil prices, domestic revenue collection remains an area of concern. The concerns center on the relatively low level of revenue generation combined with relatively high marginal effective tax rates as the result of a narrow tax base. The constraints imposed by a low and stagnant level of revenue effort (as measured by the revenue to GDP ratio) are softened by the fact that due to Tanzania's relatively high economic growth the government's resource base has been expanding despite little improvement in the revenue to GDP ratio. Priority should thus be accorded to ensuring iii that the taxation system is equitable and supportive of economic growth rather than simply trying to raise the revenue to GDP ratio at all cost. 8. Foreign aid in the form of grants and concessional loans as shown in the Central Government Operations statistics has increased from 5.6 percent of GDP in FY99 to 6.8 percent in FY01. This increase is partly due to a recovery of project loans disbursements from FY99 to FY00 and partly due to interim debt relief provided under HIPC by multilateral donors. Available aggregate figures indicate that HIPC debt relief is indeed additional to other development assistance and is not offset by other forms of aid. In terms of the composition of aid, program support has been increasing while project support has slightly declined over the past three years. Indeed, during FY01 program support (3.5 percent of GDP) exceeded for the first time the amount provided as project support (3.3 percent of GDP). The shift by donors from project to program support is an important element of the government's strategy to bring more donor resources into the budget. It is also recommended to closely monitor efforts by the External Finance Department to improve its system of capturing development assistance in the budget. Expenditures 9. Total expenditure by the central government increased from 15 percent of GDP in FY99 to 17.2 percent of GDP in FY01. Most of the increase in expenditure related to increased allocations for non-wage, non-interest expenditure, while expenditure on wages and salaries and interest payments remained relatively constant at 4.1 and 1.7 percent of GDP, respectively. 10. Higher increases in non-wage as compared to wage expenditures reduce the disequilibrium between wage and non-wage expenditures which in the past has been a severe constraint on the effectiveness and efficiency of public spending. However, as this basic disequilibrium is being reduced, it becomes key that the Ministry of Finance reevaluates the relative priority given to increasing funding for non-wage expenditures and improving conditions of service in the public sector. 11. With respect to the wage bill, implementation of the medium term pay policy to strengthen the capacity of the public sector to retain key technical and managerial staff is gaining prominence. The Ministry of Finance is encouraged to consider the implementation of the Medium Term Wage Policy as a key element of its fiscal strategy rather than implementing wage increases as a residual decision in the course of budget implementation. 12. At the same time, it will be important to keep staffing levels and the skill mix under review and undertake necessary adjustments, whenever possible. For example, the introduction of new technologies in the public sector such as the Integrated Financial Management System has led to a change in the required skill profile for government accountants and auditors. iv 13. Development expenditures remain largely donor financed, with government providing only very limited resources in the form of counterpart funding. In this context, the adoption of sector wide approaches in the health and education sectors represent an important step towards better integration of donor resources into the budget. -Government is encouraged to pursue the development of sector programs in other sectors such as water and agriculture. Of particular concern are developments in the Road Sector, where the shift of the agreed on responsibilities from the Ministry of Works to TANROADS is proceeding slower than desired which prevents the adoption of a sector program for the road sector which would have the full support of all interested donors. 14. With respect to prioritization of poverty reducing expenditures, government's continued commitment to increase spending on the priority sectors and to protect allocations to the priority sectors from cuts in the process of budget implementation. Provision of quarterly instead of monthly budget releases to the priority sectors provides another important instrument to facilitate the implementation of poverty reducing expenditures. 15. Overall, expenditures on the key priority areas have increased by 1.5% of GDP in FY00 and by 3.1 percent of GDP in FY01 compared to expenditures in FY99, which exceeds the additional debt relief made available through the enhanced HIPC Initiative. 16. Unallocated funds retained by the ministry of Finance have increased from Tsh.45.5 billion in FY01 to Tsh. 232.4 billion in FY02, which represents more than one third of all government non-wage expenditure. Retention of such large unallocated amounts threatens to dilute the efforts to prioritize expenditures. It will be important that the Ministry of Finance adopt budget management practices that minimize the need to retain unallocated items. In addition to measures aimed at preventing arrears and contingent liabilities, which claim Tsh. 86 billion of the retained special expenditures, such measures would also include better integration of wage bill management into the budget process. With respect to items for which there is a genuine justification for retaining the allocation at the Ministry of Finance, it would be desirable to identify the intended purpose clearly in the budget books. Integration of donor.resources 17. Government efforts to achieve greater integration of donor resources have found positive resonance in the donor community. Progress has been made in three areas. Firstly, a greater share of donor support is now provided in the form of budget instead of project support. Secondly, government efforts to improve the reporting format for donor assistance is leading to improved coverage of project assistance in the budget. Finally, the development of sector programs and the strengthening of public sector financial management has resulted in a greater share of project assistance passing through the budget. 18. Based on UNDP estimates of official development assistance to Tanzania, about 80 percent of inflows of project and program aid is reflected in the budget and about 50 v percent of that development assistance is reflected in the appropriation accounts. These estimates differ from those of earlier PER reports as they exclude extemal assistance not intended for the use by government such as support to NGOs, the private sector, or in kind emergency and disaster relief. II. FISCAL RISKS Sources of Fiscal Risks in Tanzania's Public Sector 19. One of the key reasons for the continued use of the cash budget system is government's concern about fiscal risks. In consideration of such potential risks, deficit targets are typically set at a much more stringent level than would be required on the basis of the fiscal sustainability analysis that has been carried out in previous years. The FY02 external review undertook a broad review of fiscal risks, including macro-economic risks arising from fluctuations in economic growth and the fiscal revenue base, risks arising from contingent liabilities, risks from the activities of agencies, local authorities, parastatals, etc. that are not directly under the commitment control system of the central government, or risks related to weaknesses in commitment control as identified under the FY01 PER. 20. The report subsequently focuses on two main sources of fiscal risks, appearing under special expenditure, namely: (i) public enterprise debts and retrenchment costs associated with privatization, and (ii) extra-budgetary commitments related to procurement: 21. Parastatal debts: One of the key problems observed is that most of the public enterprises that have been divested and those still in the pipeline for restructuring, have substantial liabilities which have to be taken over by the government in full or part. The parastatal liabilities that have fiscal implications for the government are estimated by the Task Force at Tsh. 280.47 billion and include suppliers' credits (23.8%), trade credits (20.8%), staff dues (19.9%), commercial loans (11.9%), unpaid utility bills (3.5%), and others (20.1%). This preliminary estimate by the Task Force is about 10 times higher than the Tsh. 28.3 billion parastatal debt figure recorded in the domestic debt report. Furthermore, 92.6% of these liabilities relate to public enterprises that are still in the pipeline for privatization and those that are not yet earmarked for privatization. 22. - In addition to parastatal debt, there are about Tsh. 172 billion of non-securitized liabilities including suppliers' arrears (49.4%) and compensation claims against nationalized assets (34.0%). Others include exchange losses accumulated in the BOT revaluation account and domestic guarantees. The timing, terms and form of payment of these debts is unclear but are very likely to have fiscal implications for the government in not a too distant future. 23. Risks from liabilities of independent companies in which the government has direct interest. In the formal agreements the GOT is protected from exposure of these vi liabilities. However, given that the GOT is a beneficiary of net income from the operations of such companies e.g. Meremeta, it would be enjoined in any proceedings of default. Hence, it would be useful for the GOT to get legal opinion on possible exposure to risk in international arbitration. Prudence would require subjecting debt contracting by any entity with government interest to the same requirements as public debt. 24. Retrenchment costs: Government policy regarding the financing of retrenchment is that if a public enterprise is capable of paying terminal benefits to its workers, it is allowed to pay a statutory payment. Golden hand shakes (ex-gratia payment) may be offered by a parastatal with approval by the Treasury Registrar, provided the parastatal can afford to pay such payments. Unfortunately, most of the parastatals that are in line for privatization are in poor financial position to be able to meet retrenchment costs from their own sources. This implies that the burden will have to be taken up by the government. Indicative costs estimates by PSRC for meeting the costs of retrenching all the 21,556 employees of five key parastatals (TRC, THA, ATC, DAWASA and TANESO) amnount to Tsh. 19.9 billion, covering statutory payments alone. Although no similar attempt has been made to estimate the retrenchment costs for the remainder of the public enterprises earmarked for restructuring, the implied retrenchment costs are bound to'be significant. 25. Commitment control and monitoring: Government has achieved significant improvements in the area of commitment control and monitoring of expenditure through the IFMS and application of provisions in the new Public Finance and Procurement Acts. Commitments made through the IFMS cannot exceed the funds budgeted for and released monthly or quarterly and for the purpose. On a daily basis, MOF is in a position to note any deviations/ contraventions to the norm. The IFMS also enables timely generation of reports for detailed scrutiny and analysis. As part of the government effort to improve on transparency, MOF has instituted quarterly disclosure of allocations to all spending units in the newspapers and directed these units to also display the same on their notice boards for public, scrutiny. Summary of Consolidated Arrears Cleared for Payment by Category, July 199 December 2000 (% Share) Vote Vote Holder Goods & Utilities Personal Total . _____________________ Services Emoluments . 28 Police Force 9.3 7.8 21.9 11.8 29 Prison Services 12.8 6.9 16.1 11.6 38 Defense 24.2 50.7 24.5 33.1 39 National Service 6.8 5.1 1.9 5.0 . Sub-total 53.1 70.5 64.4 61.5 46 Education & Culture 3.9 0.5 5.8 3.2 52 Health 1.1 0.0 1.9 0.9 All Other 41.9 29.0 27.9 34.4 ministries/depts./regions .. 26. However, it is worth noting that audited data on the stock of arrears (see table above) confirm the finding of the PER FY01 that the Police Force, Prisons Services, vii Defense and National Service alone accounted for about 2/3 of consolidated arrears. That is, 53.1% of debts for goods and services; 70.5% of debts for utilities and 64.4% of other debts. Debts on goods and services combined with utilities accounted for about 76.3% of the total debt cleared for payment. Thus, it is important to keep constant watch on possible occurrence of non-adherence to the centralized commitment system particularly for votes 28, 29, 38 and 39. Bold and prompt imposition of sanctions on Accounting Officers for extra-budgetary commitments is critical for inculcating fiscal prudence and containing future build-up of arrears, particularly for utilities. To close remaining loopholes in the commitment control system, expenditure items that are difficult to contain, such as food for prisoners and defense units should be made first claim charges. Containment of fiscal risks and agenda for further work 27. Government has embarked on quantifying the magnitude of the main categories of contingent liabilities which are associated with parastatal restructuring (public enterprise debts and retrenchment costs) and produced a preliminary set of estimates. Given the complexity of the, problem and incompleteness of data, it is recommended that government consider engaging a team of experts to do more detailed audit work to quantify and validate the true position of public enterprise debts and come up with options for their settlement. 28. On quasi-contingency items and provision for settlement of arrears, government is encouraged to work further to improve planning and budgeting, and especially by minimizing the lag between decision making and financing. It will be important for government to have a medium term target to reduce the quasi-contingency items currently lumped under special expenditure. These should instead be reassigned to the relevant votes and sub-votes and thereby remain with modest amounts under special expenditure for the truly unforeseen events, whose expenditure cannot be assigned ex-ante. For example, HIV/AIDS and anti-corruption campaigns, VAT refunds as well as salary adjustments should gradually be transferred to the respective ministry budgets. In addition, it will be prudent for the government to consider relegating public enterprises confirmed to have negative net worth to liquidation and writing off all uncollectible taxes of parastatals. 29. In terms of the agenda for further work in this area, there are other possible sources of fiscal risks in Tanzania, in addition to fiscal risks emanating from contingent government liabilities. These include macro-economic risks arising from fluctuations in economnic growth and the fiscal revenue base, and risks from the activities of agencies and local authorities that are not directly under the commnitment control system of the central government. For example, as noted in the PER FY01 report, the 1999 CAG report indicates the total stock of councils' creditors for 1999 was about Tsh. 5.5 billion. Over 90% of these payment arrears, which constitute part of the stock of contingent liabilities, are in form of suppliers credit while the remainder are statutory deductions not paid over. 30. It is therefore recommended that the proposed work to quantify and validate parastatal debts be extended to cover other sources of fiscal risks, including expected viii commitments and timing of privatization; and potential contingent liabilities of local authorities including possible insolvency of the Local Authorities provident Fund (LAPF). The proposed audit should also cover commitments made on behalf of the government / agreements signed by accounting officers outside the purview of MOF; the actuarial position of PPF and NIC; debt swaps and government guaranteed debts. Meanwhile, MOF needs to strengthen its monitoring and control function over LA liabilities by demanding and analyzing expenditure and revenue reports produced by LAs quarterly. III. EXTERNAL DEBT MANAGEMENT AND COMMITMENT CONTROL Context 31. The consideration of external debt commitment and. management was deemed timely in PER FY02 for three main reasons. (i) Tanzania is expected to be able to maintain a sustainable debt level and have a basis for a robust exit from recourse to further debt relief measures, having completed all required conditions and obtained debt relief from multilateral institutions under the enhanced HIPC initiative (amounting to US$ 3,000 million).2 (ii) President Mkapa has recently enjoined the Minister of Finance to take stock and review all sector loans initiated and existing by February 2002 as the first step towards forward planning of an appropriate debt management strategy. (iii) In adapting and adopting laws to streamline its public financial commitments the government of Tanzania in September 1997 prepared a National Debt Strategy; and, has in March 1999 revised and consolidated it (NDS: Part 1 - External Debt, March 1999) for guidance for proper debt management and its ultimate reduction to a manageable level. However, the legal foundation for external debt contracting in Tanzania remains the Government Loans, Guarantees and Grants Act, No. 30, of 1974. The Ministry of Finance (MOF) is working on proposals for revising this Act soon. Issues 32. Following the establishment of macroeconomic stability and the liberalization of prices in the first phase of its reforms, the government of Tanzania has now focused public policy reform explicitly to promote growth - led by the private sector - and poverty reduction.3 Problems in external debt management could derail the achievement of the poverty reduction targets and undermine sustainability of the stable macro-economic 2 IDA/IMF decision point was reached in April 2000; and the completion point was reached in November 2001 3 The PRSP (2000) and its recently completed first annual update (October 2001) embody steps to towards implementation of poverty alleviation objectives. It builds on the NPES (1997); and, ongoing processes adopted in the annual PERs, the MTEF consultations, and the CG meetings. ix environment. Hence a strong system of external debt management and commitment control is crucial to ensure integrity of the public financial systems. 33. The report reviews Tanzania's debt management practice relative to its own strategy adopted in 1999 and relative to good practice laid out in recently developed IMF/World Bank guidelines on debt management systems. It contains a matrix assessing current practice against good practice and existing strategy; and makes recommendations to address specific weaknesses. Particular attention was paid to (i) integrity/transparency of processes governing debt contracting; (ii) debt servicing and monitoring capacity (iii) ability of the fiscal authorities to track prudent debt contracting and servicing to avoid fiscal risks associated with debt contracting outside the purview of existing procedures or good practice. Summary of Findings 34. Under its PRGF/ESAF programs with the IMF, the government has agreed to adhere to borrow only on concessional terms. The international threshold for this benchmark is a grant element of at least 35 percent and all of Tanzania's recent borrowing has fulfilled this cut-off. In the stocktaking of debt exposure by reviewing current practices, it was found that the discipline and the controls around contracting debt have been standardized to a considerable extent. 35. Procedures for contracting multilateral (56.8% of debt stock) and Paris Club bilateral debt (25% of nominal stock) are most standardized, government oversight and donor requirements is adequate; and the DMC/DCC have the least time cost of preparation in pre-appraisal. 36. Procedures for contracting non-Paris Club bilateral debt (15% of nominal stock) and their system for financial management are not standardized. Many of the loans are old and mainly relate to Tanzania's financing arrangements during the socialist era. The process of scrutiny, as to the terms, is also complex. 37. Commercial Credit, mainly provided for Supplies/ Investment: These can originate from any of Tanzania's trade partners and include financing from private banks. These loans constituted 3.2 percent of the nominal debt in June 2001, and pose the most severe risk to debt sustainability despite their small overall share. Generally, the government has maintained the basic concessionality criteria of 35 percent grant element in incurring these loans, but by narrow margins. These loans have been used for financing of investment in large capital equipment; e.g. the air traffic control system. Moreover, the due process of debt contracting is largely by-passed in the initial negotiation and appraisal of these loans. Often the demand originates from MOUs entered into by sector ministries with suppliers and financiers of investment in their sector, with or without financing from foreign banks. Pressure is created on the central public oversight such that role of the MOF is reduced to rubber-stamping the legal agreement without adequate participation in the commitment/negotiation process, and moving away from being an anchor in external debt strategy. An important part of the x problem here is the absence of an institutional provision within MOF for dealing with commercial debt. 38. Short term credit: Another key risk area in the context of acquisition of goods and services by public entities is short term financing. It is mainly a problem of not enforcing procurement procedures. Ideally this expense should be budgeted for in the particular sector/departments plans. The National Debt Strategy does not cover public debt of periods less than 12 months, but if such short term debt obligations are not met the arrears can negatively impact the overall debt sustainability. 39. There are inconsistencies between the legal provisions under the Loans, Guarantees and Grants Act, no 30 of 1974 and specific Acts governing debt contracting authority of some sector ministries e.g. for National Defence and the National Service. These and others would review for ensuring the consistency and updating the Loans, Guarantees and Grants Act, no 30 of 1974 to bring the law into line with current strategy and practice. 40. For historical and capacity reasons the institutional roles for debt management have evolved in a manner that left assignment of responsibilities across departments in MOF and BOT unclear. Furthermore, the absence of a unit within MOF to manage the contracting and management of commercial debt exposes the country to fiscal risks resulting from such debt. There is also weak capacity for analyses of long term debt sustainability in Policy Analysis Department (PAD) of MOF. Recommendations 41. In looking forward the report makes recommendations in six broad areas. 42. Strengthening the National Debt Strategy 1999 in the area of Risk Identification and Provisioning for External Debt: It is recommended that the strategy should be supplemented with specific actions to be undertaken by the government in regular periodic assessment of structure of the debt portfolio and risks; and, the identification of mechanisms to monitor, evaluate, and prevent build-up of liquidity exposures, specially in foreign currency. This is essential for the Tanzanian external debt to remain sustainable; and, for its structure not to induce or propagate economic crises. Such reviews should be regular and should accompany budget presentations to Parliament. 43. Institutional Roles: It is recommended that the roles of various agencies as envisaged in the legal structure be adhered to strictly. Thus, the Minister of Finance is the sole signatory in all of URT's public debt contraction. The MOF entities responsible for his support are: Policy Analysis Department (PAD) - strategic manager of debt for sustainability and guidance to other entities; External Finance Department (EFD) - negotiator in specific debt contracting; and, Accountant General's Office (ACGEN) - custodian in charge of maintaining all legal agreements and debt obligation records, and, issuer of payment instructions against all creditor claims. The BOT is the banker to the xi government and responsible for executing payment instructions from the ACGEN; and, externalizing the foreign component of such payments. The DCC/DMC provides oversight on all proposed loans and government guarantees.4 44. Supplier Credit Arrangements: It is recommended that special scrutiny and oversight be exercised on the external debt obligations, especially from commercial sources, that arise from supplier credit arrangements for goods and services provided to sector ministries. The commitment control system should ensure that such credits are not secured outside the established procedures. 45. Short Term Debt and Non-Concessional Finance: It is recommended that special scrutiny and oversight be exercised on external debt obligations that arise from short term financing for sector ministries. Any credit from commercial sources, such as private banks, should receive special scrutiny by the DMC; and, no loans be accepted at grant element less than the government agreed 35 percent under the PRGF program (as has been the practice by GOT in recent years). As per the NDS, it should be demonstrated by the sponsoring sector ministry, and confirmed by EFD, that financing grants or credit at more favorable terms are not available. 46. Legal Provisions: It is recommended that laws for establishing public executive agencies, parastatal companies, and sector ministries that grant these entities special exemptions in external debt contracting - both for finance and for acquisition of goods and services - be scrutinized and revised for consistency with the Loans, Guarantees and Grants Act, no 30 of 1974 (and revisions to it being prepared by the MOF) and the National Debt Strategy 1999 (Part 1). In the medium term, such exemptions should be minimized and abolished. In addition, the public - especially the legislature - should be educated and sensitized on the legal roles and requirements; that prescribed sanctions be transparent; and stringency in enforcement of sanctions be instituted to deter future occurrences. 47. Independent Companies: There is concem that for independent companies in which the government is a beneficiary of net income, governnent could be enjoined in any proceedings of default. Hence, it would be useful for the GOT to seek a legal opinion on possible exposure to risk in international arbitration. As an example of the debt exposure and fiscal risks involved the report presents the case of Meremeta Ltd. It is recommended that: GOT being a shareholder of any such company, the scrutiny of its debt contracting should be on similar rigorous grounds as all other government or public guaranteed debts. The Minister of Finance should sign off on such company debt only when satisfied that all tests have been met. 4 The DCC, with members from GOT entities key in debt management, is given technical support on matching the loans with GOT priorities, the terms of loans, and the timing of repayments (bunching up) by their technical team, DMC. PAD functions as the secretariat for DCC/DMC. xii To the extent that government receives income from such companies, this should be brought into the overall expenditure allocation process to preserve the integrity of the MTEF. In cases where similar self-financing is practiced, the retention scheme is subject to the budget ceilings consistent with the overall priority allocations. IV. EXPENDITURE AND SERVICE DELIVERY TRACKING 48. The report takes stock of and assesses the existing expenditure as well as service delivery tracking system in Tanzania as an input into the government's plans for strengthening its own tracking system. This review dovetails with two related reviews undertaken in 2001, the CFAA and the IMF/World Bank assessment of capacity to track poverty-reducing expenditure. The two reviews dealt mainly with the capacity of the government to monitor allocative efficiency and the integrity of the transfer mechanisms across levels of government. This report assesses (i) the extent of coverage of the full tracking chain, (ii) the coherence of the tracking system, (iii) the institutional roles across layers of government, (iv) incentives for compliance to reporting, and (v) provisions for npalyzing in a coherent fashion the information flowing from the tracking system to inform policy and action plans. 49. The complete chain of tracking public service provision entails five main links: (i) Ensuring that allocation of resources for specific purposes - sector and activity- is according to an agreed prioritization framework and budget. Much of the traditional PER work to date has focused on this link in the chain. (ii) Ensuring the integrity and efficacy of the transfer mechanisms to facilities supplying services, through layers of government institutions and the banking system. The purpose here is to determine leakages, blockages and delays in the flow of resources to the facilities and find ways to resolve these constraints. The assessment critically depends also on the reverse flow of documentation and accounting for actual expenditures. (iii) Assessing the efficiency of the service delivery unit in converting allocated resources into real supply of services. Action plans are the main instruments mapping inputs to outputs, often in the form of contracts for delivery of specified services against budgets. (iv) Assessing the quality and availability of services delivered from the demand side - e.g. based on information from user score cards or service delivery surveys conducted by independent agencies. (v) Assessment of impact of the delivered services in terms of the effectiveness of the delivered services in improving outcomes - e.g. improving livelihood, reducing poverty in their various dimensions. xiii 50. The first two links in the chain, ( i and ii above) make up the domain broadly understood to be expenditure or financial tracking and is the main concern of the Ministry of Finance and the PER process to date. The rest of the links, (iii - v), track the efficacy and impact of resource application for development or welfare enhancement, the ultimate purpose for which they were availed. The Civil Service Department (CSD), the sector ministries/agencies, have been particularly active in these dimensions given their primary responsibilities for implementing policy, setting standards and monitoring outcomes. Main Findings 51. A wide range of administrative expenditure and service delivery monitoring systems exist across the various levels of government. These form an important base for developing a more coherent and integrated monitoring system. A major concern, however, is the wasteful duplicative monitoring/tracking systems in a context of weak and stretched managerial capacity, a result of history of multiple project and program management systems. The reports though substantive vary in quality across programs and districts. Some programs e.g. Health Basket Fund use such reports (covering the whole sector) and related reviews/audits as triggers for the next tranche release, tilting the accountability effort towards it. 52. Allocation and Transfer mechanism: The coverage of the first two links in the chain has been fairly comprehensive. The PER process and the introduction of IFMS has enabled MOF to build a transparent process for resource allocation at the macro level and transfer of funds to the regional and local authorities. Although there is substantial information available on expenditure at all levels, this is not systematically reported back up the chain to the Treasury. The flow of these reports back to the center has been incoherent and target specific sector ministries, PO RALG and specific programs - but not the main financial authority- MOF. The planned submission of standardized quarterly reports to be prepared by LAs to MOF could potentially fill this gap. 53. There have been achievements in the areas of allocation at macro level, primarily enabled by the introduction of IFMS at the central government level. Instrumental were transparency/disclosure of information associated with the participatory PER process and recent decisions to publicize allocations. Main weaknesses, as the recent IMF/WB review also notes, are associated with inadequate capacity for audit functions (internal and external), sub-national capacity for reporting spending and integration of external resources into the budget as well as the Exchequer system. 54. A key issue for priority expenditure monitoring relates to definition of priority sector spending (as broadly shown in Annex 3). Comments on whether priority sector allocations are broadly consistent with PRSP should be agreed as the framework for monitoring expenditure. However, this should be reviewed on an ongoing basis as part of the annual preparation of the Poverty Reduction Strategy progress report and MTEF. Particular issues to be addressed this year relate to identification of key spending areas in Justice system, review of Agriculture on the basis of ASDS/ASDP and ensuring fully capturing of HIV/AIDS expenditures including spending in non-priority sectors xiv 55. Service Delivery Mechanis: Three types of institutions track service delivery. These institutions have a primary responsibility, for implementing policy, setting and enforcing standards, and monitoring implementation as well as outcomes. The, institutions are CSD, sector ministries and related agencies, and the layers of institutions under PORALG. These provide coverage of service delivery in MDAs with performance budgets/contracts with CSD, and some coverage at the sector level. No mechanism exists to review service delivery at local level. However, PORALG is planning to establish an inspectorate unit to cater for this. 56. There are multitude of reporting mechanisms at local level due to different demands at the sector level and central government requirements. Difference in financial. years also unnecessarily complicates the reporting system. 57. Institutional Roles: There is lack of clarity on role of RAS in the tracking/reporting chain. There is also no overall body that pulls together information on financial performance, and on service delivery and linkages with outcomes. The issue is whether sector ministries and MOF currently get what they need to inform policy decisions, because of lack of consolidated reporting. 58. Incentives to reporting - in the like of those under health basket - are limited. The issue is whether there is scope for broadening and rationalizing this approach. Recommendations 59. The PRSP table (Annex 3) showing "Disaggregation of Recurrent Poverty Reducing Expenditure" should act as a framework for informing PRSP, PRBS, etc. It should be updated and reviewed as part of the PRS progress report and MTEF. 60. GOT wishes to internalize expenditure tracking system. It is recommended to review the existing tracking systems to identify information gaps and recommend how systems can be streamlined and refocused, and how information flows can be improved; and clarify roles and responsibilities of the key institutions in this process. This could be assessed as part of this year's PER expenditure tracking exercise. As one of its terms of reference this year's tracking study could also check whether the existing sectoral reports really reflect the assessment of service delivery/value for money; and whether harmonization of central government and local government fiscal years would help make rationalization of expenditure reporting systems more tractable and coherent. As of now reporting and tracking of expenditure has been unharmonious and complicated by the existence of the two different reporting years at local and central government level. 61. Special attention needs to be paid on both financial reporting and service delivery, and ways of integrating the two. It is recommended that there is a need for the government to take stock of the existing reporting/expenditure tracking and monitoring systems and use this information as much as possible in integrating the systems before a new system is put in place. xv 62. The government should also review how to pull together information on financial and service delivery and links to the poverty monitoring process. This could be examined under the expenditure tracking study and recommendations made on how well this should be institutionalized. 63. The establishment of an inspectorate unit in PORALG augurs well for greater cohesion among and rationalization of the wide range of tracking system at the local government level. Building capacity of financial departnents of LAs is essential and the new inspectorate unit could play an important role in this regard. Currently shortage of skills and training constrains implementation of systems e.g. reporting, auditing and IEFMS implementation. xvi V. SUMMARY OF KEY RECOMMENDATIONS Issues |Recommendation Page/para Overall Budget Management I. Inefficiency of cash budget Extend quarterly release system to all MDAs 34/ ____ system 2.66 2. Inefficiency of cash budget Improve cash flow-management through improved 35/ system cash flow plans by MDAs 2.68 3. Potential inefficiencies Monitor and report on intra-vote reallocation 17/ arising from required approvals by the Ministry of Finance 2.23 approval by MOF of intra- vote reallocations 4. Poor integration of wage Include medium term wage policy in MTEF and 16/ bill decisions into MTEF budget process 2.19 and budget 5. Weakened resource Adopt budget management practices that reduce 29/ allocation process through need to retain large amounts under lump-sum 2.47 retention of large amounts special expenditures at the Ministry of Finance (75/. for "special expenditures" 4.13) 6. Weakened transparency Improve transparency in reporting on special 29/ due to allocations to expenditure 2.47 unallocated "special expenditures" Fiscal Risks 7. Arrears on expenditures on Ensure full coverage of commitment control 74/ utilities, automatic catering, system including votes 28, 29, 38, and 39 4.10 etc. 8. Liabilities of public- Engage team of experts to undertake audit of 74 enterprises (including those public enterprise debts 4.12 under privatization) are not known with certainty 9. Liabilities of various Undertake work to quantify contingent liabilities in 75/ government linked the areas of the LAPF, PPF, and NIC 4.14 insurance and pension ____ schemes are not known . Debt Contracting and Management _ 10. Deviations from established Adhere strictly to institutional roles for debt 79/ procedures have led to contracting and debt management 5.11 incidences of dubious debt contracting 11. Transparency and oversight Publish annual public debt statement for both 79/ related to public debt is external and domestic debt with the budget 5.12 weak 12. Non-concessional and EDF staff should be assigned negotiating 79/ commercial loans are responsibility for non-concessional bilateral and 5.12 negotiated by the Bank of for commercial loans, with the support of the Bank Tanzania, with only limited of Tanzania inputs and oversight by the Ministry of Finance xvii =___ Issues Recommendation Page/para 13. Accountant General is Synchronize debt databases of Accountant General 79/ responsible for maintaining and the Bank of Tanzania 5.12 record on public debt, but BOT also maintains data base 14. Supplier credit Special scrutiny and oversight be exercised on the 83/ arrangements represent a external debt obligations, especially from 5.20 high risk for debt commercial sources, that arise from supplier credit contracting not in line with arrangements for goods and services provided to established procedures and sector ministries. The commitment control system the PRSP objectives should ensure that such credits are not secured outside the established procedures. 15. Short term financing Special scrutiny and oversight needs to be 84/ arrangements represent a exercised on external debt obligations that arise 5.22 high risk for debt from short term financing for sector ministries. contracting not in line with Any credit from commercial sources, such as established procedures and private banks, should receive special scrutiny by the PRSP objectives the DMC; and, no loans be accepted at grant element less than the government agreed 35 percent under the PRGF program (as has been the practice by GOT in recent years). 16. Public entities which are Laws for establishing public executive agencies, 76/ exempt from general debt parastatal companies, and sector ministries that 5.26 contracting procedures grant these entities special exemptions in external represent a high risk for debt contracting - both for finance and for contingent liabilities from acquisition of goods and services - should be external debt scrutinized and revised for consistency with the Loans, Guarantees and Grants Act, no 30 of 1974 (and revisions to it being prepared by the MOF) and the National Debt Strategy 1999 (Part 1). 17. Companies were For any such company, the GOT being a 88/ government is a beneficiary beneficiary of net income would be enjoined in 5.28 of net income expose any proceedings of default. GOT should seek a government to the risk of legal opinion on possible exposure to risk from contingent liabilities debt of any independent company where GOT is a beneficiary of net-income should the need for international arbitration arise. Capacity Building Measures 18. Introduction of cash flow Develop capacity for cash flow planning in management system sectoral minstries and central ministries 19. Implementation of debt Strengthen capacities for debt management strategy and adherence to institutional roles xviii VI. GOVERNMENT RESPONSE TO PER FYO1 RECOMMENDATIONS __Recommendation Government Response I Review the classification of first charge expenditures Guidelines prepared in 2002 (for 2002/03 budget) for completeness and ensure that the guidelines for instructed MDAs and regions to allocate adequate funds FY02 include instructions for fully funding these. for utilities and outstanding commitments, stating that arrears on utilities will not be accommodated in future. Cash flow plans were submitted by all MDAs. However, the quality of the plans varies and more needs to be done to ensure that better plans are prepared in future. 2. Budget Guidelines should require spending units to The 2002 budget guidelines required all MDAs to fill present a cash flow plan along with their annual new forms, designed to improve performance reporting. budgets, which conform to the respective aggregate The new forms include action plans for recurrent and ceilings but clearly show the time pattem of their cash development budget, cash flow plans, and performnance requirements. reporting (physical and financial) - in order to facilitate release of funds, periodic monitoring and evaluation of the budget performance. 3. Government should consider a cash flow smoothing There is now an understanding in principle that the Bank instrument that respects the overall budget ceilings for of Tanzania is prepared to provide bridge financing in the year (Either a cash reserve operated as a the form of cash advances or through marketable Government deposit that can be drawn down or built instruments such as treasury bills. The preference of the up depending on the net cash requirement position Bank of Tanzania is to use marketable instruments since versus available cash; or a prudent credit line with this would provide a safeguard against the possibility the BOT as advances subject to the same rule that the that bridge financing would turn into effective long-term net position as end year be zero. Central Bank financing of government expenditures. The Govermment remains concemed about the cost of such financing and would prefer to use instruments whose time structure corresponds to the temporary nature of the financing needs. In any case this instrument has not been used, to the extent that donor resources (under PRBS) and therefore allocations to priority sectors, are done quarterly and in advance. 4. A separate line item be provided for contingencies This line item exists under Vote 50 to minimize the stress from unforeseen expenditures on the overall functioning of the Government 5. Exchequer releases be issued on the stipulated date of Exchequer releases are made between 1 0d" and 1 5th of the 5th of each month so as to minimize uncertainty the month. The stipulated date of the 5th of each month on available cash. has still to be attained. 6. The Govemment should consider a relief program to This awaits the findings of a recently completed in-depth clear up the stock of debt of poor Local Authorities study commissioned to REPOA by the PER Working (Highly Indebted Poor Local Authorities - HIPLA) to Group; and also a detailed fiscal decentralization work give them a fresh chance under the local government to be undertaken in 2002. These studies will give reform program. recommendations for the way forward. 7. New payments arrears should be integrated into the To address the problem of arrears accumulation, the following year's respective spending unit's budgets as Government has committed itself (in Guidelines for first charge from their allocations. FY02/03 Budget) to: clearing all audited budgetary arrears in 2001/02; controlling the generation of LPOs through IFMS; and dishonoring any goods and services that are not accompanied by LPOs generated by IFMS. To date about 95 percent of verified arrears have been _ cleared. The rest will be cleared by end June 2002. xix Recommendation Government Response 8. To prevent abuse of extra-budgetary commitments: (i) To prevent abuse of extra-budgetary commitments, GOT enforce the centralized system for issuing LPOs; (ii) has committed itself (in FY03 Budget Guidelines) to: enhance enforcement of discipline by the accounting clearing all audited budgetary arrears in 2001/02; officers (AOs) to ensure that no invoices are held controlling the generation of LPOs through IFMS; and outside the financial reporting system; and (iii) dishonoring any goods and services that are not strengthen internal audit functions in ministries, accompanied by LPOs generated by IFMS. GOT has regions and local governments. To minimize collusion also made it clear to the private suppliers, through the with suppliers, publicize/make it very clear to the media (newspapers etc.) that any LPO issued outside the private suppliers that any LPO issued outside the central payment system will not be honored by the MOF. central payment system will not be honored by the Strengthening of internal audit functions in ministries, MOF. regions and local governments is being addressed through recruitment of new qualified staff, pay reform under PSRP, and in-service and other trainings. 9. Ensure that adequate supplies of printed LPOs is Connectivity to PLATINUM/EPICOR under IFMS has availed to all those in need of them. addressed this issue in MDAs and in sub-treasuries through stand alone systems. 10. Quickly review the IFMS roll out program to ensure All MDAs are now connected to the IFMS via EPICOR availability of functioning equipment, effective program. In practice this means that all expenditure supportive technical services, and adequate training to transactions are now being executed through the system. operating staff. Central government agencies located in regions and districts are served through the system in 19 Sub- Treasuries mainly via stand alone systems. 11. There is a need for a study to determine the extent of Attempts to address the problem are being made by the inequity of treatment and transparency in the criteria recently formed Fiscal Decentralization Task Force; but used for differential treatment of LAs in exchequer more scientific suggestions are envisaged from a releases and funding. detailed fiscal decentralization study. 12. There is a need for further training of accounts clerks The issue is being addressed by LGRP on a continuing and orientation of audit staff to the financial basis. accounting and management of LAs, as well as training of finance and audit staff in computer skills to cope with the introduction of the Platinum System. To help more effective supervision by the councilors, there is also a need occasional seminars for the councilors to sharpen their understanding of financial statements. 13. There is need to clarify the role of the Regional Fiscal Decentralization work, and PER PORALG study Secretariat vis a vis the sector ministries in monitoring commissioned by the PER Working Group, are expected and evaluation, and in enforcing national standards. to provide guidance on the way forward. 14. The report recommends that both financial and audit The Treasury has issued public notices on allocations reports (translated into Kiswahili) be displayed /made and LAs have begun doing the same. public at ward and village levels. 15. The practice existing up to the early 1980s of sending Circular has been sent out to this effect. copies of the quarterly reports to Treasury was very useful for accountability / feedback. This practice should be re-instituted and a desk officer designated in MOF to deal with such reports. xx 1. INTRODUCTION 1.1 THE PER PROCESS IN TANZANIA 1.1 In 1997, Tanzania introduced the annual Public Expenditure Review (PER) process as part of a comprehensive public sector reform program. The PER has two principal objectives: (a) providing support to the budget process and budget management and (b) providing feedback on public expenditure and management issues to government and other stakeholders through the external evaluation. Annex 7 provides a discussion of the evolution of the PER process since 1997 and compares it to a similar process that was launched in Uganda at the same time. 1.2 Key characteristics of the PER process in Tanzania are that it is government led but open to a wide range of stakeholders and its close integration with budget process. The objective of this close integration is to ensure that outputs from the PER process can feed in a timely manner into the PER process. Figure 1.1 provides a schematic overview of the key activities in the PER and the budget process. 1.3 The PER working group, chaired by the Ministry of Finance, provides overall leadership to the process. The group meets bi-weekly and includes representatives of central and sectoral ministries, the donor community, research and academic institutes, and NGOs. In addition, to the PER working group, a macro-group deals mainly with macro-economic fiscal issues. Sector representatives are members of the PER working group and have their own arrangements for PER work at the sectoral level. 1.4 The work program of the PER working group is developed at the beginning of the financial year and includes updates of sectoral PERs, studies of cross-sectoral issues, and the external evaluation. In addition to these specific pieces of work carried out by the PER working group, it also serves as an important forun for dialogue on budget issues throughout the year. For example, the working group provides comments on the draft budget guidelines prior to their finalization. Annex 8 provides an overview and status report of the activities carried out by the PER working group in FY02. 1.5 In addition to its bi-weekly meetings, the PER working group also organizes the annual PER consultative meeting which for the current PER took place on May 9-10, 2002. This meeting provides the forum for public expenditure issues to a wide range of stakeholders, including members of parliament, representatives of civil society, the private sector, and the donor community and was attended by around 300 people in FY02. Annex 9 provides the agenda of this years consultative meeting. 1.6 An important innovation of the FY02 PER process was the set up of a basket funding arrangement for the funding of commissioned studies and other expenditures of 1 the PER working group. While in the past individual studies or activities commissioned by the PER working group were funded by individual members of the working group, under the basket fund resources are pooled and the working group decides collectively on their use and the selection of consultants. Since not all working group members who provide funding were able to join the basket fund, at present, the basket funding arrangement coexists with previous funding arrangements. Figure 1-1: PER Activities and the Budget Process PER Activities Budget Process July Discussion of budget in August parliament September Development of PER WG work program_| October November Initial work on sector PERs to guideiWork on budget PER main feed into budget guidelines by budget mission to carry guidelines guidelines committee December out external evaluation Issuing of budget January gguidelines to MDAs Finalization of sector PERs, external evaluation and cross sectoral work to feed into MTEF -and budget Prepartion of sectoral March budget submissions and MTEFs April May | PER consultative meeting Preparation of budget June Presentation of budget | , ....to parlia.nent 2 1.2: THE EXTERNAL EVALUATION OF PUBLIC EXPENDITURES 1.7 The external evaluation of public expenditures is a key element of the PER process in Tanzania. Its objective is to provide feedback to government and other stakeholders on budget performance and public expenditure issues. The external evaluation work is led by the World Bank, but other members of the working group (except government) also participate and contribute significantly to the external evaluation work. 1.8 The external evaluation consists typically of two principal elements. The first is a review of fiscal developments with a focus on the previous fiscal year and the first nine months of the year during which the external evaluation is carried out. This component fulfills the function of monitoring progress and developments in public expenditure management, focusing on the areas of overall fiscal discipline, strategic resource allocation, and operational efficiency. 1.9' - The second component of the PER is the analysis of specific issues that have been brought up during the PER consultative 'meeting or identified in the course of the deliberations of the PER working group throughout the year. At the beginning of the annual PER cycle at the begin of the fiscal year, the PER working group agrees on issues to be addressed. For the FY02 PER, issues to be addressed included fiscal risks, debt contracting and-management, and a review of systems in place for expenditure tracking and output monitoring. 1.10 Activities of the external evaluation are sequenced so as to feed directly into the budget preparation process. The PER main mission, during which most of the initial analytical work is carried out took place in November and December 2001 and the findings' feed directly into the preparation of the budget guidelines. Subsequently, further analytic work is undertaken to refine and substantiate the preliminary findings of the main mission report.. The key findings of the external evaluation are then presented at the annual PER consultative meeting to a broad range of stakeholders. 1.11 ,A.. series of other assessments were also carried out during FY01 and FY02, including: the IMF's fiscal transparency module of the Review of Standards and Codes (ROSC), a joint World Bank - IMF HIPC expenditure tracking exercise which showed Tanzania's public expenditure systems to be among the most advanced in Africa, and a Country Financial Accountability Assessment that was supported by the World Bank and DFID and whose' recommendations are being niainstreamed in various public finance reform'priograms. Section 3.5 provides a discussion of how the PER external evaluation is related to other monitoring instruments in Tanzania. 1.3 PUBLIC SECTOR AND PUBLIC FINANCE REFORMS IN TANZANIA 1.12 The Public Expenditure Review is a key element of the Governments Public Finance and Public Sector Reform agenda. Most closely linked to the PER is government's Public Finance Management Reform Program which comprises five 3 components, i.e., policy reforms (which included the introduction of a new Public Finance Act and a new Public Procurement Act in 2001), the rollout of the Integrated Financial Management System, strengthening of tax administration, improvements in extemal resource management, and a strengthening of the National Audit office. 1:13 Civil service reforms are now entering their second phase which will increasingly focus on public sector performance after the first phase was mainly associated with a downsizing of the civil service and bringing the wage bill under control. A Local Government Reform Program aims at strengthening capacities of the local authorities which bear the primary responsibility for basic service delivery in priority sectors under the PRSP such as basic education, primary health, water, and rural roads. Following up on the recommendations of the Warioba report which documented deeply entrenched corruption in the public sector, a national action plan for the control of corruption as well as sector specific anti-corruption plans are being implemented. In addition, government has also agreed to implement a Legal Sector Reform Program and has set up a Good Governance Coordination Unit in the Chief Secretary's Office. 4 2. REVIEW OF FISCAL DEVELOPMENTS 2.1 INTRODUCTION 2.1 The review of fiscal developments focuses mainly on FY01 and the first nine months of FY02. This section provides an overview of recent fiscal developments and discusses some of the key issues that affect public expenditure management. The analysis focuses primarily on issues related to aggregate fiscal discipline and strategic resource allocation. Specific sectoral issues relating to the efficiency of public expenditures are analyzed in sector specific PER updates. 2.2 AGGREGATE FISCAL PERFORMANCE 2.2 The review of aggregate fiscal performance focused on developments of the broad fiscal aggregates as presented in the Central Government Operations tables prepared by the Ministry of Finance. This includes the review of fiscal policy in terms of setting and pursuing specific fiscal targets, the resource availability from domestic sources and from foreign aid inflows, and broad expenditure trends in terms of the economic composition of public expenditures. Key policy issues in this area include the setting of fiscal targets and mechanisms to achieve these targets, efforts to increase revenue performance and to integrate foreign aid more fully into the budget, and the allocation of resources between wage and non-wage expenditures. Box 1: The Recurrent Deficit as a Fiscal Target Although the budget speech and various other documents define fiscal targets in terms of the recurrent budget deficit, de facto with the operation of the cash budget and under PRGF commitments, the key fiscal target is zero net-domestic borrowing. The usual rationale behind choosing the recurrent deficit as a fiscal target is that it indicates what amount of domestic resources is available for funding of development expenditures. However, for countries that are heavily aid-dependent and where most of development expenditures and often some recurrent spending are financed through development assistance, the deficit after grants or net-domestic financing requirement is generally considered to be a more appropriate fiscal target or indicator. With the increased availability of donor program finance, both in the forms of credits and grants, the fiscal deficit becomes even less meaningful as a fiscal target. The reason for this is that with the availability of concessional financing the deficit before grants or the recurrent deficit are largely determined by the availability of foreign finance rather than being a meaningful policy target of fiscal policy. 5 Table 2-1: Financing of the Fiscal Deficit (% of GDP), FY96 - FY02 Central Government Operations FY97 FY98 FY99 FY00 FY01 FY02 FY02* Actual Actual Actual Actual Actual Budget Actual Govemment recurrent savings (checks issued) 1.0% 1.0% 0.5% -0.5% -1.2% -2.5% 0.0% Balance before grants -1.6% -2.8% -3.5% -4.4% -5.0% -6.6% -3.6% (checks issued or commitment basis) Balance after grants 2.0% 0.2% 0.5% -0.5% -1.1% -1.2% 0.0% (checks issued or commitment basis) Balance after grants 1.90/c 0.2% -0.9% -0.5% -1.5% -1.2% -0.2% (checks cleared or cash basis) Foreign Loans (net) -0.5% 1.0% 0.4% 0.8% 1.1% 1.2% 1.8% Balance After Grants and Foreign Loans 1.4% 1.2% 0.7% -0.3% -0.4% 0 1.6% (checks cleared or cash basis) Domestic (net) -1.4% -1.2% -0.7% -0.1% 0.4% 0.0% -0.9% Bank -0.4% -0.9% 0.0%/a -0.1% -0.2% 0.4% -1.4% Nonbank (net of amortisation) -0.3% 0.5% -0.1% 0.0% 0.2% 0.0%/0 0.5% Privatization Funds 0.3% 0.1% 0.2% 0.2% 0.4% 0.2% 0.0% Change in arrears -0.9% -0.8% -0.8% -0.2% 0.0% -0.6% -0.7% * annualized estimate based on estimate for July-March FY02 Source: IMF, Tanzanian authorities 2.3 Overall fiscal discipline, defined in terms of recurrent deficit targets5, has remained a key policy objective. Targets for these variables are defined in the Medium Term Expenditure Framework and in the budget and monitored under the IMF supported PRGF prograrn. After targeting and achieving recurrent surpluses during the period FY97 - FY99, during the period FY00 to FY02 the targets have been revised to provide scope for increased financing of priority sector activities under the Poverty Reduction Strategy and to accommodate increased foreign inflows in the form of program grants. The recurrent deficit in FY00 and FY01 was 0.5 percent and 1.2 percent of GDP, respectively, and the target for FY02 is a recurrent deficit of 2.5 percent of GDP.6 Work on fiscal sustainability that was undertaken in the past indicates that these deficit targets are well within the sustainability thresholds and consistent with continued macro- Note: FY02 Outturn Figures Figures on actual expenditures and revenues for FY02 presented in this report cover the period July 2001 - March 2002. In order to allow comparison with figures for previous years, these figures have been annualized when presented as a percentage of GDP by using three- fourth of the projected GDP for FY02 as the denominator. It needs to be kept in mind that these annualized figures do not present a projection of the likely outturn for FY02, since fiscal data contain important seasonality, development data are only fully captured at the end of the fiscal year, and expenditure releases during the fourth quarter of the fiscal year typically differ significantly from expenditure patterns during previous quarters as a result of the cash budget system. 6 economic stability. Given substantial aid inflows in the form of grants and concessional loans, the recurrent deficit has in the past been consistent with net repayments to the domestic banking sector. 2.4 Although the fiscal framework foresees a recurrent deficit of 2.5 percent of GDP, spending during the first half of FY02 was significantly below budget and the recurrent deficit was in fact zero. Taking grant and concessional loan inflows into account, there was a surplus of 1.2 percent of GDP which was used to reduce outstanding domestic debt and arrears. In previous years, under-spending during the first half of the fiscal year has typically been partially offset by accelerated spending during the second half of the fiscal year. However, the consistency of tight fiscal policies with net-flows to the domestic banking sector remains of concern in a situation where key government functions appear to be significantly under-funded and public sector wages for technical and professional cadres have fallen significantly below wage levels in the private sector. An additional concern about governments tight fiscal policy arises from the fact that financial markets suffer from excess liquidity and real interest rates on treasury bills have turned negative. Figure 2-1: Key Fiscal Developments, FY95-FY02 Tanzania is developing a solid track record of fiscal but the recurrent fiscal deficit target has been discipline and macroeconomic stability ... gradually relaxed OaaU balance alttr grants (checks cleared or cash basis) Rse,.rast Deficit 3.0% 1.0 2.0% .5% 1.0% 0.5% ,_ ,._____ - Ia _______ ____ F9 Y7F9 Y9F0 Y1F0 0.45 % . I 0.0%. s -130% . -.0% - 2 30% -- ... :* - ' .;.%0 -2.0% -6.0% .~~~~~~~~~~~~~~~2.5% E 40 10 I FY98 FY97 FY98 FY99 FY00 FY01 FY02 to accomnodate the shift towards budget support and to create scope for the expansion of poverty by donors... redu ing expenditures Pmgramr Support (As P7 5.0% r Education *Heth I 450% __ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _10 Water Othes I 25% ~4 2.0%- 215%2 FY96FY97FY98FY99FY00 FY01 FY02 7*' 2.5 In Tanzania, the key instrument to achieve fiscal deficit targets is a cash budget system combined with a computer assisted commitment control system. A large contingency allocation retained at the Ministry of Finance provides the government 7 protection against fiscal risks. Even though expenditures under this contingency allocation are earmarked for specific purposes, they provide the Ministry of Finance with some flexibility to either reduce expenditures for these earmarked purposes (e.g. salary increases) or to shift some of that expenditure to the next fiscal year (e.g. payments related to the parastatal restructuring process). The key risks which justify the setting of fairly stringent fiscal targets and the maintenance of a sizable contingency allocation include (a) deviations of revenue performance from budgeted amounts, (b) unforeseen expenditure requirements, including contingent liabilities, and (c) poor budget management of spending units. As measures are taken to reduce these risks to the achievement of the fiscal targets and as improved budget management systems become firmly entrenched, it will be appropriate to review the fiscal targets. Domestic Revenue 2.6 After three years of continuous decline, total revenue as a percentage of GDP increased from 11.4 percent in FY00 to 12.2 percent in FY01. This increase is mainly attributable to the introduction of VAT on petrol imports combined with the sharp increase in international petrol prices. Loss of revenue due to the decline in international petrol prices during the first half of FY02 has been offset by improved compliance following the introduction of bio-marking of fuel, which is intended to curtail leakages of untaxed, transiting fuel into the Tanzanian market. Table 2-2: Government Revenue (as % of GDP), FY96 - FY02 Central Government Operations FY97 FY98 FY99 FY00 FY01 FY02 * FY02 * Actual Actual Actual Actual Actual Budget Actual Total Revenue 13.5% 12.0% 11.5% 11.4% 12.2% 12.2% 12.2% TaxRevenue 11.9% 11.00/0 10.3% 10.4% 10.9% 11.1% 11.0% Taxes on imports and exports 3.8% 3.5% 3.6% 3.4% 4.8% 5.3% 4.7% Sales and excise taxes on local goods 3.1% 2.7% 2.7% 2.6% 2.5% 2.5% 2.5% Income taxes 3.0% 2.9% 2.7% 3.1% 2.6% 2.2% 2.6% Othertaxes 2.1% 1.9% 1.2% 1.2% 1.1% 1.1% 1.2% Nontax revenue 1.6% 1.0% 1.2% 1.0% 1.3% 1.2% 1.1% Note:* annualized estimate based on preliminary outturn for July - March FY02. Source: Tanzanian authorities (MOF) 2.7 In FY02 government introduced VAT on government imports in order to curtail abuses of the government exemption from VAT on imports that have occurred in the past. However, it appears that revenue potential from this source has been seriously overestimated. While the budget estimate of Tsh. 56 billion for revenue from this source may have been to optimistic to begin with, subsequent exemption of all donor funded imports has also contributed to lower revenue collection than originally expected. 2.8 Other tax sources, including sales and excise taxes on local goods, income tax, and other taxes continued their decline during FY01 with little sign of improvement during the first half of FY02. Underlying the decline in tax revenue over the past few years is a streamlining of the tax system which involved substantial reductions in external 8 taxes and the elimination of nuisance taxes. In addition, other factors that contributed to the decline in the revenue to GDP ratio were relatively large tax incentives for new investments, the continued downsizing of the parastatal sector, and sluggish private sector growth which has not yet yielded enough revenue to compensate for lost revenue Figure 2-2: Real Growth of Domestic Revenue, FY95-FY01 from the shrinking parastatal sector. Only 15.0% - revenue from income taxes showed better l0.0% - ~. --, . . . than projected performance. This is due to increased withholding taxes for wages and 5.0% l salaries due to strong employment growth 0.0 °% - t I --l -- > + F in the mining sector. I-.%|'_.-.lltW;j . :/, 2.9 The constraints imposed by a low and stagnant level of revenue mobilization 10.0% !---o ".-+'':';'{s^T are softened by the fact that due to CD 0) 0 o O t Tanzania's relatively high economic growth I- U. U. U- U. the government's resource base has been expanding despite little improvement in the revenue to GDP ratio. (Figure 2.2) Priority should thus be on ensuring that the taxation system is equitable and supportive of economic growth rather than trying to raise the revenue to GDP ratio at all cost. The expectation is that the revenue to GDP ratio will increase as a consequence of a sustained supply side response to economic reforms, albeit with a lag. The program of strengthening tax administration is ongoing and expected to yield tangible results over the next few years. In addition to administrative improvements, streamlining of tax exemptions is another potential source for additional revenue generation. 9 Box 2: Revenue Performance and GDP Growth Estimates The most common indicator for measuring a country's revenue performance is the ratio ofl domestic revenue to GDP. Thus, the quality of GDP estimates as well as the structure of GDP and sectoral contributions to economic growth have an important impact on the magnitude and change of this measure of revenue performance. In evaluating Tanzania's revenue performance it is often claimed that the specific structure of Tanzania's GDP contributes to the relatively low revenue performance. With respect to the composition of GDP, it is important to note that 30 percent of Tanzania's GDP is non-monetary and does thus not contribute directly to the tax base. Using only monetary instead of total GDP as the basis for the calculation of the GDP to revenue ratio, this ratio would be 17.3 percent instead of 12.2 percent in FYOI. In comparison, in Kenya non-monetary economic activities account only for 2 percent of GDP and the revenue to GDP ratio is 25.3 percent. For Kenya, calculating the revenue to GDP ratio on the basis of monetary instead of total GDP would increase that ratio only marginally to 25.8 percent. Another factor often considered to be relevant for the evaluation of Tanzania's revenue to GDP ratio is the fact that growth in mining has been expanding rapidly during recent years (17 percent average annual real growth over the past four years), while due to various tax holidays mining activities do not contribute to government revenue. However, since the share of mining in overall GDP is still relative small (2.3% in 2000) and the contribution' of mining activities to economic growth is also relatively small, (0.2 percentage points in 1998, 0.1 in 1999, and 0.03 in 2000), the expansion of untaxed mining activities is not a major factor in explaining the performance of the revenue to GDP ratio. A final factor that needs to be considered in evaluating the revenue to GDP ratio is the possibility that subsequent to the revision of GDP figures in 1997/98 there was a consistent overestimation of economic growth. Although there is no evidence that published GDP figures either over or underestimate economic growth, we consider the hypothetical impact of a systematic overestimation of GDP growth by 1, 2 and 3 percent, respectively. Revenue to GDP ratio as a function of real GDP growth FY97 FY98 FY99 FY00 FYOI Actual GDP 13.5% 12.0% 11.5% 11.4% 12.2% 1 percent overestimation of economic growth 13.5% 12.1% 11.7% 11.70/o 12.7% 2 percent overestimation of economic growth 13.5% 12.2% 11.9% 12.0% 13.1% 10 Foreign Aid Inflows 2.10 Foreign aid inflows in the form of grants and concessional loans - the second major source of funds for government - have continued to.increase during the past three years. While grant financing was relatively stable at around four percent of GDP, loan disbursements recovered from a relatively low level of 1.6 percent in FY99 to 2.2 percent of GDP in FY00 and FY01. For FY02, the budget foresees a further increase in aid inflows. Grant aid is projected to increase to 5.4 percent of GDP and net concessional foreign loan inflows are projected to increase to 1.2 percent of GDP. During the first half of FY02, inflows of program grants showed indeed a significant increase compared to the previous year. Recorded inflows of project grants were significantly below the budgeted amount. However, this is a consequence that most inflows of project aid are only accounted for towards the end of the fiscal year. Table 2-3: Foreign Inflows - Grants and Loans (as % of GDP), FY96 - FY01 Central Government Operations FY97 FY98 FY99 FY00 FY01 FY02 * FY02 * Actual Actual Actual Actual Actual Budget Actual Grants and Loans 3.1% 4.0% 4.4% 4.7.% 5.0% 6.6% 4.1% Grants 3.6% 3.0% 4.0% 3.9% 3.9% 5.4% 3.6% Program 1.8% 0.7% 1.2% 1.7% 1.5% 2.4% 2.6% Project 1.8% 2.3% 2.8% 2.2% 1.6% 2.2% 0.3% HIPC Relief (IMF, WB and AfDB) 0.1% 0.7% 0.8% 0.6% Foreign loans (net) -0.5% 1.0% 0.4% 0.8% 1.1% 1.2% 1.8% Foreign loans (loan disbursements) 0.8% 2.1% 1.6% 2.2% 2.2% 2.8% 2.7% Program Loans (import support) 0.5% 1.3% 0.6% 0.8% 0.5% 1.5% 0.5% Development Project Loans 0.3% 0.8% 1.1% 1.4% 1.7% 1.3% 2.1% Amortization -1.3% -1.1% -1.2% -1.3% -1.1% -1.6% -0.9% * annualized estimate based on prel. outturn for July-March FY02 Source: Tanzanian authorities 2.11 The most significant event in terms of foreign aid was access to HIPC debt relief in FY00 and the reaching of the HIPC completion point in November 2001. HIPC debt relief reduces Tanzania's debt service obligations by 58 percent over the next twenty years. The Central Government Operations tables show only multilateral HIPC debt relief as a grant, while bilateral HIPC debt relief is directly reflected as a reduction in expenditures on interest payments and debt service (Table 2.3). Since some of the bilateral debt has not been serviced prior to HIPC, actual resources that can be redirected for spending on poverty reducing expenditures are thus less than the reduction in debt service obligations provided by HIPC. Since the level of official grants (including HIPC) has remained fairly constant over the period FY99-FYOI, HIPC debt relief was partly offset by a decline in other grant financing. However, it is worth noting that the multilateral debt fund (MDF), which prior to HIPC provided already some form of debt relief has been converted into the poverty reduction budget support (PRBS) facility which provides a common instrument for providing budget support by interested bi- and multilateral donors. Donors contributing to the PRBS include the European Union, DFID, Finland, Norway, Sweden, Denmark, Netherlands, Ireland, Japan, Canada, and Switzerland. For FY02, the PRBS is expected to disburse an amount of US$ 147 million, of which US$ 68 million have already been disbursed during the first half of FY02 comprising disbursements from the EU, Finland, Norway, and the U.K. Table 2-4: HIPC Debt Relief as % of GDP, FY00 - FY02 FY00 FY01 FY02 Total HIPC Debt Relief (% of GDP) 0.8 1.9 1.8 HIPC Debt Relief shown as grants (% of GDP) 0.1 0.7 0.7* Based on annualized outturn for the period July - March FY02 Figure 2-3: Foreign Development Assistance to Tanzania, FY95-FYOI Figure 3a: Project and Program Support, FY99-FYO1 Figure 3b: Grants and Loans, FY99-FYO 1 - ~~~~~~~~~~~~~4 0% la~ 2.0% ' _% _FY0t I0% 1.0% I'U¶I,I Iii ~~~~~~I!EI~~mj 1~.5% 0.0% F9 FY6 F9 FY8.d0.5% FY9 FY6 F97 YOO FY99 FY00 FY01 FY95 FY98 FY97 FY98 FY99 FY00 FY01 a Prawed UProgmn OHIPC |OGrants *LOns UHIPC| Figure 3c: Grants, FY99-FYOI Figure 3d: Loans, FY99-FYO 1 __________________________. 2 3.0% . . 20% 0.5% 1 __ 0.0% ol FE~E~ 0% __ FY95 FY96 FY97 FY98 FY99 FY00 FY01 FY95 FY96 FY97 FY98 FY99 FY00 FY01 OProqectGrants EPmgramGrant OHIPC | |PrOiCt LOOM UProgrLaLoans 2.12 Over the past three years, the composition of foreign aid has changed in a number of important ways. First, until FY99 project support was increasing and program support decreasing. This trend has been reversed in the past two years with a decline in project support and an increase in program support, which is even more pronounced when HIPC debt relief is taken into account as a form of program support. (Figure 2.3a). This change in the composition of aid is mainly due to a change in the composition in grant financing (Figure 2.3c) from project to program financing through the PRBS. Credit financed project support continued to increase in FY01 and credit financed program support has fluctuated between 0.5 and 0.8 percent of GDP during the last three years (Figure 2.3d). However, the World Bank, the major external provider of credit to Tanzania is also shifting its assistance towards the increased use of program finance modalities, such as the recently approved support to the education sector development program or the 12 Poverty Reduction Support Credit (PRSC) which is currently under preparation which should reinforce the shift towards program financing. Box 3: Implications of Expanded Program Support for Fiscal Management The shift from project to program support that is currently underway in Tanzania has a number of important implications for fiscal management in Tanzania: * Balancing of recurrent and development expenditures: In the past the provision of project aid has often led to an imbalance between development and recurrent expenditures, which frequently resulted in inadequate funding for operation and maintenance of donor financed investments. Program support provides the flexibility to fully integrate recurrent and development spending under the MTEF and it will be important that adequate capacity is developed to ensure proper allocation of both tax and program aid resources between development and recurrent expenditures. In particular, given the current situation where 80 to 90 percent of the development budget is financed from project aid, with more aid in the form of program support attention will have to be paid to proper funding of the development budget. * Additional risk of resource variability: In the past program support showed typically a higher degree of variability and unpredictability than domestic revenue. Given the high cost that resource variability imposes on economic management, it will be important to develop aid mechanisms that allow for stable and predictable aid disbursements and that prevent sudden changes in the volume of aid. * Program aid as buffer to domestic revenue variability: Aside from trying to minimize fluctuations both in terms of amounts and timing of program support disbursements, mechanisms to utilize program aid as a means to close temporary cash flow gaps arising from asynchronous monthly revenue inflows and expenditure requirements. * Substitute for domestic revenue: Funds from program support are fully fungible to funds from domestic revenue. Since increased program support lessens the pressure to increase domestic revenue, appropriate incentives should be put in place to ensure that an equitable and growth supporting tax system is in place, which in the medium to long term will have the capacity to generate sufficient funds to lessen the dependence on program support. * Exchange rate implications: To the extent that budget support finances a greater share of non-tradables (e.g., teacher salaries) than does project support, a shift towards program support leads to upward pressure (appreciation) of the exchange rate with potential implications for the international competitiveness of Tanzania. * Deficit targets: With most of development assistance provided as project support, the recurrent deficit was often used as a policy target. With the shift towards program support which can be used to finance both recurrent and development expenditures, recurrent deficit targets will need to be adjusted to allow for the use of program finance for recurrent expenditures. In addition, the deficit after grants and domestic net borrowing may be the more adequate targets for fiscal policy. 13 2.13 Second, while grant support (inclusive of HIPC) has remained relatively constant at around 4 percent of GDP, inflows from concessional loans net of amortization have increased continuously over the past three years from 0.4 percent of GDP in FY99 to 1.1 percent in FY01. While disbursements on program credits were about 0.5 and 0.8 percent over the past three years, disbursements on project loans have increased from about 1 percent of GDP to 1.7 percent of GDP. The increase in project disbursements indicates improved project implementation performance. Government Expenditures 2.14 Government spending increased in FY01 to 17.2 percent of GDP compared to 15.8% of GDP in FY00. For FY02, the budget foresees a further increase in total expenditure to 18.8 percent of GDP. The increase in expenditure is attributable to the increase in expenditures on goods and services and transfers (i.e. Other Charges), while expenditures on wages and salaries, interest, and development remained roughly at the level of FY00. Figures for the first half of FY02 show the typical pattern of significant under-spending on other goods and services, a reflection of the cash budget system but also of the general cautious fiscal stance of government. Typically, spending during the second half and particularly during the last few months of the fiscal year is typically significantly higher than during the first half of the fiscal year. (For a detailed discussion of these intra-year expenditure patterns see the section on the cash budget.) Table 2-5: Government Expenditures as % of GDP, FY96 - FY02 Central Government Operations FY97 FY98 FY99 FY00 FY01 FY02 * FY02 * Actual Actual Actual Actual Actual Budget Actual Total expenditure and net lending 15.1% 14.9% 15.0% 15.8% 17.2% 18.8% 15.7% Recurrent expenditure. 12.5% 11.00/o 11.00/o 11.9% 13.4% 14.8% 12.8% Wages and salaries 4.7% 4.3% 3.7% 4.2% 4.1% 4.3% 4.0% Interest payments 2.6% 2.3% 1.6% 1.6% 1.7% 1.4% 1.4% Domestic 1.7% 1.0% 0.6% 1.0% 1.0% 0.90/ 0.8% Foreign 0.90/h 1.3% 1.0% 0.5% 0.7% 0.6% 0.6% Other goods and services and transfers 5.1% 4.5% 5.7% 6.1% 7.7% 9.1% 7.4% Development expenditure and net lending 2.6% 3.8% 4.0% 3.9% 3.8% 4.1% 2.9% o/w Expenditure financed domestically 0.5% 0.5% 0.3% 0.3% 0.5% 0.6% 0.5% * annualized estimate based on preliminary outturn for July-March FY02 Source: Tanzanian authorities 2.15 After an increase in the wage bill from 3.7 percent of GDP to 4.3 percent of GDP in FY00, in FY01 and FY02 the wage bill as a share of GDP declined marginally to 4.1 percent. This is the consequence of both a reduction in the number of civil service staff and wage increases below the nominal rate of GDP growth. 14 Table 2-6: Civil Service Employment, FY98-FY01 (December of each year) Salary Scale FY98 FY99 FY00 FY01 FY99 FY00 FY01 Staff in absolute numbers Percentage change TGOS 35651 34806 35001 37530 -2.4% 0.6% 7.2% TGS 63787 63310 59994 56838 -0.7% -5.2% -5.3% TGTS 122215 118868 119566 122148 -2.7% 0.6% 2.2% TPSW + TGPSW 36190 34821 36448 39823 -3.8% 4.7% 9.3% OTHERS 12785 11381 8837 772 -11.0% -22.4% -91.3% TOTAL 270628 263186 259846 257111 -2.7% -1.3% -1.1% Target under 258543 249,000 -3.7% Medium-term Pay Reform Strategy TGOS - Tanzania Government Operational Service, TGS - Tanzania Government Service, TGTS - Tanzania Government Teachers' Service, TPS - Tanzania Protective Service, Tanzania Government Protective Service Source: CSD 2.16 After the adoption of a comprehensive civil service reform program in 1993, the size of the civil service (comprising central government, protective service workers, teachers, and some local government personnel) has been reduced from 355,000 in 1992 to about 270,600 by 1997. During the last three years, the civil service has continued to shrink, although at a much slower base than in previous years. During FY01 the total government workforce declined by 1.1 percent. This is significantly below the projected levels in the medium-term pay strategy which foresaw a decline in the civil service by 3.7 percent to 249,000. 2.17 The workforce of the central government civil service (TGS) has seen the largest reductions in the past two years and the number of staff declined by 10 percent. Some of this decline is due to the shifting of government responsibilities and staff to executive agencies. The size of the Tanzania government teachers' service (TGTS) was relatively stable over the past three years. However, the PRSP and the Education Sector Development Program foresee a significant increase in the number of teachers in order to be able to meet education sector objectives such as universal primary education. Tentative projections indicate that if enrollment rates increase as projected the number of teachers will need to increase by 10 to 15 percent over the next two to three years. The Tanzania policy force (TPSW and TGPSW) increased by around 4000 during the past three years. 15 Table 2-7: Civil Service Average Salaries, FY98-FYO1 Salary Scale FY98 FY99 FY00 FY01 FY99 FY00 FY01 Average monthly salary (Tsh.) Percentage Change TGOS 37703 38212 47220 56633 1.4% 23.6% 19.9% TGS 57313 57625 78067 85267 0.5% 35.5% 9.2% TGTS 56314 57073 80162 92670 1.3% 40.5% 15.6% TPSW + TGPSW 60506 61400 70925 75669 1.5% 15.5% 6.7% OTHERS 115878 108011 167861 84149 -6.8% 55.4% -49.9% TOTAL 54684 55387 74143 78878 1.3% 33.9% 6.4% Source: CSD 2.18 During the period 1993 to 1998, average real civil service salaries increased by 75 percent. However, this increase was partly due to the consolidation of allowances already included in the compensation package and a recent study on the government's medium- term pay reform strategy shows that real net monetary compensation had indeed declined for many salary grades. During 1999 government implemented the first phase of the salary reform with significant salary increases, especially for technical staff and higher cadres which led to an increase in average salaries by 55.4 percent. However subsequent to this one-time increase in salaries there were no significant salary increases during FY01 and FY02. 2.19 The steady increase in spending on non-wage expenditures is clearly a very positive development in Tanzania's budget management as under-funding of operations and maintenance has in the past severely impinged on the effectiveness and efficiency of government operations. However, as funding for non-wage expenditures increases there is also increasing urgency to pay greater attention to (a) the composition of non-wage expenditures, (b) the trade-off between increases in non-wage expenditures and the implementation of the medium term wage policy and selective accelerated salary enhancements. It is recommended that decisions on wage and employment levels are made an integralpart of the budget/MTEFprocess. 2.20 At present, the level of development spending is to a large extent determined by the availability of donor project finance. Central government operations statistics indicate that about four percent of GDP or a little more than one fourth of the overall budget is spent on development expenditures, most of which is foreign financed. The relatively low outturn figure for the first half of FY02, 1.8 percent of GDP, reflects current practice where most donor funded expenditures are reported only at the end of the fiscal year. Problems with the proper integration of development expenditures in the budget continue. As donors move from project to program support, it will be important that development expenditures are fully integrated into the MTEF process. It will also be necessary to closely monitor the balance between recurrent expenditure and development expenditure. To the extent that program financing replaces project financing, government will have to pay close attention to adequate funding of development expenditure from domestic revenue or budget support. (A more detailed discussion of donor finance into the budget is presented in a separate section of this report). 16 2.3 STRATEGIC RESOURCE ALLOCATION 2.21 After restoring overall fiscal discipline, the second key objective in the area of public expenditure management has been. to improve strategic resource allocation in the public sector. Since FY97, the government process for strategic resource allocation has been opened up to greater participation by all stakeholders both within and outside of government and there are now a number of important processes that feed into government's allocative process. These processes include the now well established PER/MTEF process which has as one of its key objectives the improvement of the information base on which budgetary decisions are being made and to solicit inputs and review from a wide variety of stakeholders. The other key process is the PRSP process, which aims at sharpening the focus on poverty reduction of public expenditure and public policy. The first PRSP was prepared in 2000 and the first annual PRSP update in 2001. The PRSP confirmed areas for poverty reducing expenditures which should receive priority in the process of budget formulation and execution and which had been adopted over the past few years in the framework of the PER/MTEF process. 2.22. This section reviews the broader sectoral composition of public expenditure in Tanzania and then more specifically public expenditure on the priority areas. Aside from the review of allocations and actual expenditures there are two issues that are of particular concern in this area. The first relates to fairly large contingencies retained by the Ministry of Finance and allocated during the fiscal year that will be discussed in more detail below. 2.23 The second relates to the provision in the new public finance act that intra-vote reallocations need to be approved by the Ministry of Finance. The public financial management revokes the authority of accounting officers to affect virements between budget items. Authority for such virements has to be provided by the Ministry of Finance. This rule limits the managerial flexibility of account-holders and provides the Ministry of Finance with additional power to control expenditure. In particular, the IFMS provides the Ministry of Finance with an effective tool to exert such control. The objectives of this new rule is to provide the Ministry of Finance with the powers to enforce spending priorities at the budget item level and prevent inappropriate spending patterns. In addition, making intra-vote virements more difficult should also provide an incentive to improved budget preparation by the spending units. However, this rule also bears considerable risks which will need to be monitored closely. Firstly, it opens the door to micro-management of sectoral expenditure programs by the Ministry of Finance which could increase transaction cost and open new doors for inefficiencies in the implementation of expenditure programs. In this context, it will be necessary to closely monitor the functioning of the new approval system and the time it takes from a submission of a request for reallocation by the spending unit to the time a decision is made. Secondly, meaningful implementation of such control, in particular decisions on whether intra-vote reallocations are justified, put considerable demands on already limited capacities in the Ministry of Finance. It is recommended that the Ministry of Finance prepare a quarterly report that details all requests for reallocations, the decision taken, and the time elapsed between receipt of the request and issuing of a decision. 17 Functional Analysis of Public Expenditure 2.24 This section presents the functional analysis of public expenditures for the period FY96-FYO 1 using data obtained from the appropriations accounts. Table 2-8: Composition of Public Expenditures as % of GDP, FY96-FY02 VOTE HOLDER FY96 FY97 FY98 FY99 FY00 FY01 FY02 FY02* Actual Actual Actual Actual Actual Actual Budget Actual Recurrent Expenditures 12.5 13.5 13.1 12.8 12.8 13.6% 16.6% 12.0/ Debt Service 3.7 5 4.9 3.8 4.2 3.4% 4.0% 2.80Y Supply Votes 8.8 8.5 8.3 9.0 8.6 10.1% 12.7% 9.30/Y Recurrent Central 6.5 6.2 6 6.9 6.3 7.4% 10.0% 6.80/ Recurrent Regions 2.3 2.3 2.3 2.1 2.3 2.7% 2.6% 2.50/ Development Expenditure 0.5 0.9 1.6 1.8 1.5 2.1% 3.5% 0.40Y Total Expenditure 13 14.4 14.7 14.6 14.3 15.7% 20.1% 12.40/ * Annualized outturn based on the period July - March 2002 Source: Appropriation Account (FY96 - FY01), Expenditure Release Reports (FY02) 2.25 Overall, recurrent expenditures as a percentage of GDP increased from 12.8 percent of GDP in FY00 to 13.6 percent of GDP in FY01. The budget for FY02 indicates a further increase in recurrent expenditure to 16.6 percent of GDP, although expenditure releases during the first half of FY02 were significantly below the budgeted amounts and even below expenditures during FY01 . As the share of recurrent expenditures used for debt service payments declined from about 4.2 percent of GDP in FY00 to 3.4 percent in FY01, a greater share of recurrent expenditures is used to fund ministerial and regional supply votes. Since FY96, recurrent expenditure by the central government has increased from 6.5 percent of GDP to 7.4 percent. Recurrent expenditures by the regions and transfers to the local authorities have increased at a much slower pace from 2.3 percent of GDP in FY96 to 2.7 percent in FY01 and a slight budgeted decline for FY02 to 2.6 percent of GDP. Since transfers to local authorities provide resources for the delivery of 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 is of 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, such as that for the provision of drugs, is still maintained by the sectoral ministry, while donor financed expenditures for service delivery at the local level are, if at all, typically captured in the development budget for the sectoral ministry. Transfers to decentralized levels of government show less fluctuation than expenditures by central MDAs, indicating that these expenditures have been effectively protected, but also a reflection of the fact that these transfers cover mostly wages and salaries. It is recommended that in line with the stated government 7The definition of recurrent expenditure in this section uses the government of 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 of recurrent expenditures. 18 priorities efforts should be made that transfers to the local authorities increase at least at the same pace as expenditures by the central govermment. In this context it is also important that government sets up an appropriate process that would allow policy reforms in the area of fiscal decentralization to catch up with the local government reform process and the shifting priorities in government spending. A joint government-donor review carried out in 2001 highlighted that unless fiscal decentralization is accorded a higher priority on the government reform agenda, the entire local government refonn program is likely to make little substantial progress Table 2-9: Sectoral Recurrent Expenditures (actuals, as a %age of GDP) Sector FY96 FY97 FY98 FY99 FY00 FYOI FY02 FY02* Actual Actual Actual Actual Actual Actual Budget Actual Administration 2.1 1.4 1.6 2.2 2.1 2.3% 5.0% 2.1% Defence and Security 2.4 2.5 2.2 2.2 2.0 2.2% 2.1% 2.0% Social Services 3.5 3.5 3.6 3.7 3.6 4.4% 4.5% 4.2% Economic Services 0.2 0.6 0.4 0.4 0.7 1.0% 0.8% 0.7% Productive Services 0.5 0.4 0.3 0.5 0.3 0.4% 0.4% 0.4% Supply Votes 8.7 8.5 8.2 8.9 8.6 10.1% 12.7% 9.2% Consolidated Fund Services 3.7 5 4.9 3.8 4.2 3.5% 4.0% 2.8% Total Recurrent Expenditures 12.5 13.5 13.1 12.8 12.8 13.6% 16.6% 12.0% * Annualized outtum based on the period July - March 2002 Source: Appropriation Accounts (FY96-FYO 1), Expenditure Release Reports (FY02) 2.26 Table 2.9 presents the functional classification of recurrent expenditures funded from the central government budget for the period FY96-FY02. The social sectors include education, health, water, and the ministries for community development and women's affairs and labor and youth development. Expenditures on the social sectors, which had remained fairly constant at around 3.6 percent of GDP over the period FY96 to FY00 increased to 4.3 percent of GDP during FY01 and is projected to increase to 4.5 percent of GDP during FY02. 2.27 Expenditures on administration have been rising faster than expenditures on the social sector during the past five years. However, a significant part of the increase in expenditures on the administrative sector is related to expenditure on the new civil service pension scheme, clearance of arrears to suppliers incurred by all ministries, one time expenditures for the preparation of elections, and increased expenditures for priority areas related to accountability in the public sector such as the OCAG, the Judiciary, the Civil Service Reform Department, the establishment of the Ministry of Regional Administration and Local Authorities, and the Ministry of Lands. The sharp increase in budgeted expenditures in FY02 is attributable to the retention of funds for special expenditures in the budget of the Ministry of Finance which in the magnitude of 2.8 percent of GDP. This allocation for special expenditures will cover VAT refunds for governnent purchases (Tsh. 56 billion), contingent liabilities related to parastatals (Tsh. 44 billion), payment arrears (Tsh. 41.9 billion), salary adjustments (Tsh. 42.7 billions), abolition of primary school fee (Tsh. 11 billions), health insurance (Tsh. 11 billions), 19 contingencies (Tsh. 9.3 billions), program aid to Zanzibar (Tsh. 6.3 billion), HIV/AIDS (Tsh. 4.4 billion), housing allowances for higher learning institutions (Tsh. 3.5 billion), the anti-corruption campaign (Tsh. .7 billion) and the airport service charge (Tsh. 1.6 billions). Expenditures on defence and security have declined from 2.5 percent of GDP in FY97 to 2.2 percent in FY0 1. 2.28 Taking all this into account, expenditure patterns are broadly in line with the objective of increasing social sector spending faster than other spending. Nevertheless, given that the under the extremely tight resource constraints government is not able to fund operating cost for many non-priority government agencies even at a minimum level, government is urged to critically review and prioritize among administrative functions and limit its engagement to those services that can be properly funded. Table 2-10: Social Sector Recurrent Expenditures (actuals, as a %age of GDP) Sector FY96 FY97 FY98 FY99 FYOO FYOI FY02 FYO2* Actual Actual Actual Actual Actual Actual Budget Actual Education 0.3 0.3 0.4 0.3 0.3 0.4% 0.4% 0.4% Health 0.3 0.4 0.5 0.6 0.5 0.5% 0.6% 0.5% Water 0 0 0 0.1 0 0.1% 0.1% 0.1% Science, Technology & Higher Education 0.5 0.5 0.4 0.5 0.4 0.6% 0.6% 0.6% Regions 2.3 2.3 2.3 2.1 2.3 2.7% 2.6% 2.5% Total Social Services 3.5 3.5 3.6 3.7 3.6 4.4% 4.5% 4.2% * Annualized outturn based on the period July - March 2002 _ Source: Appropriation Accounts (FY96-FYO I), Expenditure Release Reports (FY02) 2.29 Expenditures in the social sectors have increased from 3.6 percent of GDP in FY00 to 4.2 percent of GDP in FY01. Most of the increase in social spending occurs at the regional and district level, where the responsibility for the delivery of basic services such as primary education and health care lies. However, given the financing constraints which Tanzania faces, scarce resources will have to be focused on those areas identified in the Poverty Reduction Strategy Paper, which can be expected to have the biggest impact on poverty reduction and economic growth. Analysis of the benefit incidence of government programs should provide important information for the allocation of public resources. Government is encouraged to make full use of the household budget survey data to study the benefit incidence of government programs. This might be of particular importance for relative budget allocation to different levels of the education and health system but also to make inter-sectoral allocations. 20 Table 2-11: Sectoral Development Expenditures (actuals, as a %a e of GDP) _ Sector FY96 FY97 FY98 FY99 FY00 FYO IFY02 Actual Actual Actual Actual Actual Actual Budge Administration 0.1 0.4 0.4 0.0 0.3 0.4% 1.6% Defence and Security 0.0 0.0 0.0 0.0 0.0 0.0% 0.0% Social Services 0.2 0.2 0.4 0.9 0.5 0.8% 1.5% Economic Services 0.1 0.2 0.6 0.7 0.6 0.6% 0.2% Productive Services 0.1 0.0 0.2 0.2 0.2 0.2% 0.2% Total Development Expenditures 0.5 0.9 1.6 1.8 1.5 2.1% 3.5% Source: Appropriation Accounts (FY96-FYO1), Expenditure Release Reports (FY02) 2.30 Recorded development expenditures show a continuous upward trend increasing from 0.5 percent of GDP in FY96 to 2.1 percent of GDP in FY01. This increase in development expenditures is partly due to a greater share of donor funded expenditures being captured in the appropriation accounts, but also to improvements in project implementation performance. However, given the constraints on the side of recurrent financing of operations and maintenance, it will be imperative to properly take into account the recurrent cost implications of development projects to ensure their sustainability. Poverty Reducing Expenditures 2.31 A key characteristic of the Tanzanian budget process is the clear identification of priority areas. In the past, priority areas were identified in the budget guidelines and the PER process provided a framework for the discussion and monitoring of priorities by a wider group of stakeholders. Priority areas have evolved over time, including both a sharpening of priorities and the addition of new priorities. The Poverty Reduction Strategy Paper, which has become the overarching framework for Tanzania's poverty reduction efforts, reconfirmed the priorities that had been agreed on in the course of the previous years. At present the key priorities include: Education, Health, Water, Judiciary, Agriculture, and Roads, and HIV/AIDS. Within the priority sectors, the PRSP defines a set of priority budget items which pertain to the provision of Basic Education, Primary Health, Water, Rural Roads, the Judiciary, Agricultural Research and Extension, and HIV/AIDS. Expenditure targets in the PRSP for these areas refer mainly to recurrent expenditures, as development expenditures are mostly donor financed. 21 Box 4: Identifying Priority Expenditures in Tanzania The fiscal stabilization and rationalization efforts launched in 1997 with the introduction of the cash budget system and the adoption of a participatory annual PER process and a Medium Term Expenditure Framework were the starting point for the definition of pro-poor priority sectors and expenditures. From an initial rather broad focus on social sector expenditures the definition has over the years been sharpened to focus more on the delivery of basic services and an inclusion of expenditures linked to income generation and growth such as agricultural extension, rural roads, the judiciary, and anti-corruption efforts. An important issue that was discussed over the past few years was whether there should be ring- fencing of specific expenditure items and move towards a "poverty action fund" (PAF) similar to the one in Uganda or whether the main focus should remain on poverty reducing sub-sectors such as basic education and primary health. So far, the consensus has been that given the information constraints as well as the complexity of production functions in these sectors, it would be preferable to retain the focus on service delivery in the sub-sectors rather than to move to the earmarking of specific expenditure items. In addition, it was also felt that the move to a PAF might distract from the focus on overall strategic resource allocation and expenditure management issues and limit the discussion purely to PAF items, which was considered not to be desirable. The key actors in the definition and refinement of priority sectors are the budget guidelines committee and the PER working group. The budget guidelines committee (including representatives from central and sectoral ministries) is responsible for the preparation of the budget guidelines, which define budgetary priorities and sector ceilings which are key inputs to the budget process. The PER working group (including representatives from central and sectoral ministries, the donor community, NGOs, and academic and research institutes) provides support to the budget process and the definition through the commissioning of sectoral and cross-sectoral studies. The bi- weekly meetings of the PER working group provide an important platform for dialogue on prioritization issues and have been instrumental in shaping pro-poor expenditure priorities over the past few years. Finally, the PER consultative meeting, which is held annually, provides an opportunity for govemment to listen to the views of a wide range of stakeholders on prioritization issues. This feeds into the preparation of the budget as well as the revision of the medium term framework. The.PRSP process largely drew on this processes and broadly confirmed the prioritization that was already in place. However, while previously the focus on poverty reduction was implicit, the PRSP has given it a higher profile and established a framework to monitor the impact of this prioritization of expenditures. Government has developed a poverty monitoring master plan which is currently under implementation. 22 Table 2-12: Expenditures for Priority Sectors, FY00 - FY02 FY0O* FYOI* FY02** Education 23.8% 24.2% 25.7% Basic Education 14.4% 19.2% 18.1% Health 8.1% 9.0% 9.6% Primary Health 4.3% 6.8% 5.7% Water 0.8% 1.0% 1.6% Agriculture (Research and Extension) 2.0% 1.0% 1.7% Agric. Research and Ext. 0.7% 1.0% 1.0% Roads 6.3% 7.8% 7.4% Rural Roads 2.3% 2.7% 2.9% Judiciary 1.2% 1.1% 1.5% HIV/AIDS N/A N/A 0.8% Total Priority Sectors 42.2% 44.1% 48.3% Total Priority Items 23.8% 31.8% 31.6% *actual **budget estimate 2.32 Since the publication of the PRSP in August 2000, the share of priority sector expenditures in total discretionary spending8 has increased from 42.2 percent in FY00 to 44.1 percent in FY01 with a further increase to 48.3 percent budgeted for FY02. Similarly, expenditures on priority items within the priority sectors have increased from 23.8 percent of discretionary expenditure in FY00 to 31.8 percent in FY01. 2.33 One of the key motivations for the HIPC debt initiative was to allow countries to use resources that were previously used for debt service for increasing expenditure on poverty reducing expenditures. Given the fungibility of resources and annual fluctuation in the overall resource envelope of government, earmarking of HIPC resources for specific purposes is neither desirable nor analytically meaningful. What is of interest is government's overall reallocation of resources for priority expenditures. As is illustrated by Figure 2.4, the increase in expenditures on priority sectors has increased by a significantly larger amount than was provided by HIPC. 8 Discretionary spending is defined as total recurrent expenditures minus expenditures on consolidated fund services, i.e., debt service and pensions. 23 Figure 2-4: Incremental Spending on Priority Areas Compared to FY99 and HIPC Resources (in percent of GDP) 6 va. IC 2 FYOO FYOI FY02 | HPC relief * Priority Sector Spending Note: Priority Spending for FYOO and FYOI are actual and FY02 reflects the budget estimates 2.4 VARIATION BETWEEN THE APPROVED BUDGET AND ACTUAL EXPENDITURES 2.34 The credibility of the budget process depends to a large extent on whether decisions on allocations made during the budget process are followed through during the process of budget execution. If expenditure releases to spending units deviate significantly from budget allocations, the budget process is likely to loose validity in the. eyes of budget holders and consequently, informal processes to obtain resources during the fiscal year may emerge. These informal processes include bi-lateral negotiations during the process of budget execution with the Ministry of Finance to obtain releases of funds as well as efforts to secure funds outside the budget, e.g. from donor sources or through the accumulation of arrears. 2.35 There are three principal reasons for deviations of actual expenditures from budgeted amounts at the beginning of the financial year with respect to recurrent expenditures: * differences between projected and actual resources available which under a cash budget can lead to expenditure releases by the Ministry of Finance which are less than the budgeted amounts; and * reallocations between spending units and in particular reallocations from the contingency account at the Ministry of Finance in the course of the budget year; * under/overspending by Ministries compared to the approved budget. 24 A. Differences between projected and actual resources 2.36 Under Tanzania's cash budget system, actual expenditures are determined by actual resources available from domestic revenue and donor support. If actual resource availability differs from the projections at the beginning of the fiscal year, expenditures are adjusted accordingly rather than covering the difference through government borrowing. In addition to variations in resource availability, deviations between budgeted and actual amounts of expenditures on first claim expenditures, i.e. payments for wages and salaries and debt services, also impact on the level of expenditures on operations and maintenance and domestic development expenditures. Table 2-13: Central Government Operations, actual as a percentage of budgeted,FY97-FY01 Actual as a percentage of budgeted FY97 FY98 FY99 FY00 FY01 FY02* Total Resources 95% 90% 88% 94% 103% 93% Domestic Revenue 101% 92% 93% 95% 108% 99% Budget Support 69% 79% 66% 89% 86% 77% Less First Claim Expenditures 109% 103% 90% 103% 97% 89% Debt Service and Arrears 110% 110% 90% 103% 97% 82% Wages and Salaries 107% 97% 90% 103% 98% 94% Available Resources for Other Expend. And Dom. Dev. 80% 76% 86% 86% 109% 96% Total Other and Development 70% 70% 94% 87% 106% 75% Other Goods and Services 84% 85% 98% 89% 107% 76% Domestic Development 26% 25% 53% 61% 92% 61% Deviation of actual from budgeted (as a % of GDP) l FY97 FY98 FY99 FY00 FY01 FY02 Total Resources -0.8% -1.5% -1.8% -0.8% 0.4% -0.6% Domestic Revenue 0.2% -1.0% -0.9% -0.6% 0.9% 0.0% Budget Support -1.0% -0.5% -L.0% -0.3% -0.5% -0.5% Less First Claim Expenditures 0.8% 0.2% -0.8% 0.2% -0.2% -0.4% Debt Service and Arrears 0.5% 0.4% -0.4% 0.1% -0.1% -0.3% Wages and Salaries 0.3% -0.1% -0.4% 0.1% -0.1% -0.1% Available Resources for Other Expend. And Dom. Dev. -1.6% -1.8% -1.0% -1.0% 0.6% -0.2% Total Other and Development -2.4% -2.2% -0.4% -0.9% 0.4% -1.2% Other Goods and Services -1.0% -0.8% -0.1% -0.7% 0.5% -1.1% Domestic Development -1.4% -1.4% -0.3% -0.2% 0.0% -0.1% Domestic Financing -0.6% -0.4% 0.8% 0.6% 0.1% -0.6% * data for FY02 cover only the period July - December 2002 2.37 The following analysis has two objectives. The first objective is to show to what extent outturns on broad revenue and expenditure aggregates were in line with budgeted amounts. The analysis is carried out by showing actual outcomes as a percentage of 25 budgeted amounts. The second objective of the analysis is to assess the source of shocks which lead to cuts in expenditures on operations and maintenance and domestic expenditure on development. This analysis is carried out mainly by reviewing the difference between budgeted amounts and outcomes as a percentage of GDP. Table 2.13 presents the key indicators for this analysis for the period FY97-FY02. Data for FY02 cover only the first half of the fiscal year and provide only limited information since revenue collection and expenditures are usually not evenly distributed over the year. Subsequently, the analysis will also look at intra-year patterns of the budget aggregates discussed in this section. 2.38 Table 2.13 shows that with the exception of FYO1, the budget has typically overestimated resource availability from domestic and foreign sources9. With respect to domestic revenue, the deviation between budget estimate and actual revenue collection was particularly high in FY98 and FY99 at 1.5 and 1.8 percent of GDP, respectively. However, in the last three years the difference between budgeted and actual revenue has declined significantly, with actual revenue exceeding budgeted revenue in FYO1 by 0.9 percent of GDP. Resources available from donor budget support have been consistently overestimated during the past five years by 0.3 to 1 percent of GDP. Improving the projections of donor budget support and making such resource flows more predictable is a key challenge as donors move to the increased use of budget support. Given the relative large share of funding that is provided through donor support, unless resource flows can be reliably projected the budget process will remain flawed. Figure 2-5: Comparison of Actual and Potential Fiscal Balance (after grants) 3.00% .; 2.00% Wi ..f-- ; ; -. 1.00% o0.00%" 3 * _ -1.00% _ -3.00% - - . FY97 FY98 FY99 FYOO FY01 FY02 I Fiscal balance (after grants) . Potential fiscal balance (after grants) 9 Foreign sources include only program support in the form of loans and grants. Project support is not included in the current discussion since deviations in the availability of project support are offset by matching variations in foreign financed development expenditures. 26 2.39 Figure 2.5 compares the actual fiscal deficit (after grants) to a hypothetical potential fiscal deficit (after grants). The potential fiscal deficit is the deficit that would have occurred if government would have compensated for the resource shortfall from domestic and foreign sources through borrowing to protect budgeted expenditure levels. With the exception of FY99, when the potential deficit would have reached 2.7 percent of GDP, for the other years the potential deficit would have been remarkably stable at around 1.1 to 1.3 percent of GDP. Indeed, the potential deficit would have shown much less variation than the actual deficit under the cash budget system. While the analysis shows that compensating for resource shortfalls during the period FY97 to FY01 would not have led to a series of unsustainable fiscal deficit, it also needs to be kept in mind that abandoning the cash budget system would alter the incentives for realistic estimates of resource availability and could give rise to pressures for more optimistic resource projections, which in turn would facilitate higher expenditure levels. This reinforces the point that a condition for moving away from the cash budget system would be the maintenance of conservative resource estimates. 2.40 First claim expenditures are statutory payments that have to be paid in full, independent of the prevailing resource situation. They comprise in essence debt service payments and wages and salaries. With respect to debt service payments, the deviation between budgeted and estimated expenditures shrank from 10 percent during the period FY97-FY99 to only 3 percent during the period FY00 - FY01. It is noteworthy that with respect to debt payments, actual payments have significantly deviated from budget estimates. This points towards weaknesses in Tanzania's debt management system, which need to be urgently addressed. 2.41 Actual expenditures on wages and salaries match quite closely the budgeted amounts. Although wages and salaries are first claim expenditures, they also serve as a buffer to absorb shocks to resource availability. Salary increments are typically only implemented towards the middle of the fiscal year if the resource situation permits. This is the underlying reason for the relatively large discrepancy between budgeted and actual expenditures on wages and salaries observed in FY99. However, as projections for resource flows and debt service payments become more reliable, it would also be desirable to make employment and wage policy an integral element of the budget and MTEF process rather than having the implementation of the medium term wage policy act as a residual. 27 Figure 2-6: Deviation from Budget (as a % of GDP) of Funds Available and Actual Expenditures on O&M and Development 1.00% 0.50% 1 0.00% IL-0.50%- C3 T~~~~~~~~~~ - 200% - .. - -1.50% -L -2.50% .* . -3.00% FY97 FY98 FY99 FY00 FY01 FY02 OiAvailable Funds for O&M and Development I Expenditures on O&M and Deveiopment 2.42 Under the cash budget system deviations from the budgeted amounts for revenue and first claim expenditures are directly offset through changes in releases for non-wage expenditures. As a result of inaccurate budget projections of resource flows and first claim expenditures, expenditures for operations and maintenance and for domestically financed development expenditures were in the past significantly less than what was budgeted. However, as figure 2.6 shows, the situation has improved quite significantly between FY97 and FY02. While in FY97 actual expenditures on O&M and domestic development were only 70 percent of the budgeted amount, in FY02 actual expenditures were in fact above the budgeted amounts. Another important change noticeable over the past few years is that while in FY97 and FY98 expenditure cuts were even deeper than what would have been required to accommodate resource shortfalls, in subsequent years the fiscal stance has become slightly more accommodative with expenditure cuts being slightly less then the resource shortfalls. 2.43 In conclusion, it appears that over the past five years there were significant improvements in budget management at the aggregate level. This is apparent by smaller deviations between budgeted and actual amounts in recent years compared to earlier years and also the fact that under the cash budget system B. Reallocations 2.44 During the budget process not all projected resources are allocated to spending units. A contingency provision is allocated to the Ministry of Finance which during the process of budget implementation is then re-allocated to spending units, if the revenue situation permits. In the Finance Bill Parliament authorizes the Ministry of Finance to 28 undertake reallocations across votes throughout the fiscal year. At the end of the fiscal year, a statement of reallocation is presented to Parliament for approval. 2.45 This contingency fulfills several purposes: * reduce the risk of having to borrow in case revenue collection and foreign inflows are less than projected since with respect to many of the items covered under the contingency government has some flexibility with respect to the amount and the timing of the payments; * keep aside funds for emergencies; * cover increases in the wage bill which are typically not determined prior to the approval of the budget; * retain funds for the implementation of policy decisions which have public expenditure implications but where the institutional mechanisms for their implementation have not yet been sufficiently specified at the time of the budget or for multi-sectoral or thematic expenditure allocation for which the distribution across spending units has not yet determined at the time of the budget, (e.g., funds for multi-sectoral HIV/AIDS activities, the implementation of sectoral anticorruption plans, or * cover financial transactions carried out by the Ministry of Finance whose amount is not yet precisely known at the time of the budget (e.g., funds for clearing liabilities of parastatals that are being privatized, clearing of arrears, transfer of 4.5 percent of incoming budget support to Zanzibar, VAT refunds on government purchases). 2.46 In addition to reallocations from the contingency budget allocation, the Ministry of Finance also undertakes special requisitions to enhance spending on consolidated fund services, comprising debt service payments and expenditures of the state house. Table 2-14: Contingency allocation retained by Ministry of Finance FY98 FY99 FYOO FYOI FY02 Tsh. Billion 38.0 29.3 66.7 44.5 232.4 YO of Supply Votes (Ministerial and 6.7% 4.7% 8.8% 5.1% 16.6% ICFS) I 2.47 In recent years, the amounts allocated for contingencies has varied considerably, ranging from Tsh. 29.3 billion in FY99 to Tsh. 232.4 billion budgeted for FY02 (see Table 2.14). Annex 4 shows the breakdown of the contingency for FY99 - FY02. 2.48 While it is recognized that in an environment of unpredictable resource flows and vulnerability to external shocks the current practice of retaining an unallocated contingency item in the budget of the Ministry of Finance has its justification, care should be taken to establish clear criteria for determining the magnitude of this item, to limit the degree of discretion in the use of this contingency, and to establish clear and transparent criteria for the use of this contingency allocation. Otherwise there is a threat that this contingency undermines the credibility and transparency of the budget process and of budget implementation. It is further recommended that the Ministry of Finance prepares 29 a report on the use of special expenditure at the end of the fiscal year which is submitted to Parliament. Table 2-15: Reallocations and Expenditure Outturn, FY01 Approved Reallocation Budget after Actual Approved Actual Budget Reallocation Expenditure Budget after Expenditure Reallocation as % of as % of Approved (Tsh. (Tsh. (Tsh. (Tsh. billion) Approved Budget after Billion) Billion) Billion) Budget Reallocation Ministry of Finance 51.8 -48.2 3.8 40.8 7% Administration 139.2 21.0 160.2 152.2 115% 95% Defence 164.6 6.7 171.3 75.2* 103% 96% Social Services 332.7 15.1 347.8 332.0 105% 95% Economic Services 73.7 15.3 89.0 74.2 121% 83% Productive Services 29.6 -5.0 24.6 27.8 83% 113% Total Supply Votes 791.6 4.9 796.7 702.2 101% 88% CFS 288.6 -6.2 282.4 263.0 98% 93% GRAND TOTAL 1080.2 1 1079.1 1054.7 109% 89% Source: Budget books, appropriation accounts, and statement of reallocation. 2.49 Table 2.15 shows the allocation of this contingency in FY01 to the various sectors (Details of reallocations to individual votes can be found in annex 1). It is worth noting that in addition to funds originally retained under the contingency item, funds originally allocated for debt service payments and for the productive sectors were reallocated to other sectors through the contingency account. 2.50 Table 2.15 also shows actual expenditures in Tsh. and as a share of approved budget estimates after reallocations. An interesting feature of the data on reallocations and expenditure is that in many cases reallocations to the sectors from the contingency item are offset by under-spending by the sectors, which leads to a situation where at the aggregate level actual spending deviates less from the originally budgeted amounts then from the amounts after reallocations. C. Implementation of the Development Budget 2.51 Most of the difference between approved development expenditures and development expenditures recorded in the appropriations accounts is due to problems in capturing donor funded projects. 30 2.52 Firstly, the problem of appropriately capturing planned donor project disbursements in the budget submitted to Parliament still persists, although there are indications that the situation has improved significantly during recent years. The Ministry of Finance is making continued efforts to capture donor funded projects in the budget and donors are generally responding very positively to these efforts. In addition, the recent adoption of sector development programs for health and education and the planned sector prograrns for roads and agriculture also are important tools for better integration of donor funding in the budget, 2.53 In addition, even for donor funded projects that are contained in the approved budget, expenditures are often not channeled through the exchequer system and consequently not captured in the appropriation accounts. This makes it difficult for government to keep track of the implementation of the development budget. The difference between approved development funds in the budget and recorded expenditures in the appropriation accounts may be either due to the fact that donor funds by-pass the exchequer system, contributions are provided as real goods and services without any recorded financial flows to Tanzania, or planned development projects are not implemented as planned. Table 2-16: Development Expenditures by Sector, Actual Expenditures as a Share of Budgeted Expenditures, FY96-FYOO FY96 FY97 FY98 FY99 FY00 FY01 ADMINISTRATION 8.3% 29.2% 17.5% 6.4% 101.5% 45.7% DEFENCE AND SECURITY 18.6% 25.0% 34.4% SOCLAL SERVICES 12.2% 35.3% 24.6% 54.1% 32.6% 62.7% ECONOMIC SERVICES 20.0% 1.5% 33.9% 52.7% 59.9% 50.0% PRODUCTIVE SERVICES 66.6% 16.4% 91.4% 31.1% 36.5% 80.1% GRAND TOTAL 14.0% 6.1% 27.5% 45.9% 52.2% 56.5% Source: Appropriation Accounts 2.54 Table 2.16 shows the ratio of budgeted to actual development expenditures for the period FY96-FYOI. In recent years there has been a significant increase in the share of budgeted development expenditures actually recorded as expenditures. Between FY97 and FYO, the ratio increased from 6.14 percent to 56.5 percent. This points towards improved capturing of donor funded assistance in the appropriation accounts. It is also likely to be due to improved portfolio implementation performance by the government and donors. (A more detailed analysis of the integration of donor funding into the budget is presented below.) 31 2.5 MOVING FROM THE CASH BUDGET TO CASH FLOW MANAGEMENT 2.55 To ensure fiscal stability the government of Tanzania has successfully employed a rigid cash budget system. While the cash budget helped to achieve aggregate fiscal discipline, it impinged severely on allocative and operational efficiency and the government is now implementing reforms to introduce more flexibility and to move cautiously from a strict cash budget to a more efficient cash flow management system. 2.56 In previous years, monthly cash releases were determined on the basis of the average revenue collection during the preceding three months plus any funds received from program support. Once the monthly ceiling was established by the Bank of Tanzania, first charge expenditure items such as statutory payments for debt service and pension payments as well as payments for wages and salaries for the month were deducted. In a next step, one twelfth of budgeted amounts for priority sectors was released plus any special lumpy expenditures for the priority sectors, such as expenditures for examinations. Whatever remained after deducting all these expenditures from the monthly expenditure ceiling was then allocated for OC for non-priority ministries and for locally funded development expenditures. 2.57 Figure 2.7 shows monthly exchequer releases for expenditures on O&M as a percentage of average monthly exchequer releases for expenditures on O&M for a given year, covering the period FY98-FYOI, i.e. the period since the cash budget is in effect. Figure 2-7: Monthly Exchequer Releases as a Percentage of Average Monthly Exchequer Releleases for a Given Year, FY97-FY01 250% 200% [3FY97 5 '0 ; ' FY98 200% -A... . a - ....- FYOO UFY01 50% -.oFO 0% I- W i S j _: 2.58 The graph reveals a number of interesting features: 2.59 Firstly, expenditure releases during the first two months of a fiscal year (July and August) are typically significantly below average monthly releases. This reflects the 32 seasonal pattern in revenue collections, which are generally below average during the first months of the fiscal year. Low expenditure releases during the first two months could also be related to disbursements on program aid that took place at the end of the foregoing fiscal year,. but which are not taken into account in the calculation of expenditure releases during the next fiscal year. Finally, low expenditure releases at the beginning of the fiscal year reflect also a generally very cautious stance by the Ministry of Finance. 2.60 The second observation relates to generally high expenditure releases during the last quarter of the fiscal year and in particular during the last month. These high expenditure releases at the end of the fiscal year are a direct consequence of the Ministry of Finance's cautious fiscal management. However, relatively large expenditure releases at the end are not likely to contribute to an efficient implementation of spending units expenditure programs. Either the funds cannot be fully used or they are used to cover arrears which had been accumulated in anticipation of the more generous releases at the end of the year. 2.61 Aside from the patterns observed at the beginning and the end of a fiscal year, monthly expenditure data do not seem to display any easily discernible pattern during a year or across years, reflecting the operation of the cash budget system, where expenditures on O&M act as a residual. The lack of a pattern across years indicates further that the aggregate of domestic revenues, external inflows, and first claim expenditure does not produce any seasonal patterns that would then be reflected in the expenditure releases. 2.62 Looking more specifically at the expenditure release pattern for FYO1, there appears to be a weakening of some of the effects observed in previous years. In particular, the beginning of year and end of year effects are less pronounced. The overall months to months fluctuation in expenditure releases for O&M are also lower than in previous years. Figure 2-8: Monthly Exchequer Releases as a Percentage of Average Monthly Exchequer Releases for a Given Year, FY98-FY01 200% - 160% - 140%/ 120% . iF9 80% .04. . >. r i .FY9, 40% I 0%-3 33 2.63 It is also interesting to compare exchequer release patterns before and after the introduction of the cash budget system. Figure 2.8 shows monthly exchequer releases for expenditures on O&M as a percentage of average monthly exchequer releases for expenditures on O&M for a given year, covering the period FY94-FY96, i.e. the period before the cash budget. Surprisingly, the observations made for the cash budget period seem also to be valid for the pre-cash budget years with low expenditure releases at the beginning of the year, high releases in the last month, and no consistent monthly pattems for most of the months of the fiscal year. Figure 2-9: Average Absolute Monthly Deviation from Mean of Expenditure Releases for O&M 45% rsi 40% C Q Y t -j 35% :30% - -. c. 25% E - tg¶4; 4r%tE. 20% Xl 15% 1 0% _ fi' i q 4 5T i 'W"E7 5% 0% r FY94 FY95 FY96 FY97 FY98 FY99 FYOO FY01 2.64 Figure 2.9 shows the average of the absolute deviations from the mean of expenditure releases for other goods and services for the period FY94 to FYO1. The variability of expenditure releases of O&M has almost halved over the past three years from 33 percent in FY99 to 16 percent in FYO1. Resource availability for spending units has thus become more predictable in FYO1, although the cash budget system in all likelihood still imposes severe cost in the form of reduced efficiency in the implementation of expenditure programs. 2.65 As documented in previous PERs, this system limited considerably the capacity of spending units to implement their work plans. First, the availability of funds for operations and maintenance has been unpredictable since it depended on short term resource availability. Second, spending units were constrained to undertake only expenditures that did not exceed one twelfth of their budget allocations and they were not able to allocate expenditures over the year according to their needs. 2.66 In order to alleviate these constraints, the Ministry of Finance has introduced quarterly budget releases for the priority sectors, where one fourth of their annual budget allocation is made available to the priority ministries at the beginning of every quarter. 34 The move from monthly to quarterly commitmnent and allocation for the priority sectors and a few other social sector ministries such as Community Development Women Affairs & Children; and Labour & Youth Development, has had a positive impact in two respects. First, it has resulted into time saving on the part of MOF in doing the actual allocations. Second, the move has permitted these sectors to plan their expenditures better and reduced uncertainty. Around 42 percent of government's discretionary spending (non-wage, non-interest - recurrent) spending benefits from quarterly exchequer releases. However, those spending units not covered by the quarterly releases continue to experience the unpredictability and high variability of resource flows under the monthly cash budget system. In fact, as a significant share of spending is protected and released on a quarterly basis, exposure to unpredictability and variability by the non-priority sectors has indeed increased. The fscal risk inherent in moving to a generalized quarterly releases system is estimated to be small and recommends the general move from monthly to quarterly releases for all spending units starting in FY03. The upper limit of this risk would be one sixth of the annual allocation for discretionary spending to non-priority sectors, i.e. around 0.5 % of GDP. 2.67 In addition, ministries were also requested to prepare cash flow plans which would allow to match monthly or quarterly expenditure releases to the expenditure requirements of the spending units. While all ministries have prepared such cash flow plans for FY02, the Ministry of Finance considers these plans to be not yet of sufficient quality to make them the basis for exchequer releases. 2.68 Figure 2-10 shows the monthly cash flow requirements for a sample of eight ministries. The graphs show that while some ministries indicate fairly even distribution of cash requirements over the year, others such as education indicate that cash requirements differ significantly from month to month. A uniform monthly allocation of one twelfth of the annual budget allocation is thus clearly not optimal for the rational implementation of the ministries' work programs. One of the obvious weaknesses of current submissions of cash flow requirements is that some of the ministries only indicate quarterly expenditure requirements. Since the objective of the cash flow plans is not only to provide a basis for exchequer releases, but also to improve aggregate cash flow management by the Ministry of Finance, it is important that spending units are urged to estimate monthly cash requirements, even when they receive quarterly exchequer releases. 35 Figure 2-10: Monthly Cash Requirements for OC According to Cash Flow Plans, FY02 Presidents Office & The Cabinet Foreign Affairs 0.95 I 50 0.90 - *I i 4.0011 0.85 -7_ 0 0 0.80 2iH,FI1 j .~ 00D co ~ ~ ~ ~ ~ ~ ~ M o9 q 0.70 Defence Agriculture & Food Security 3.82 4 2.50 X382 f i 2.00 . 3.2382 ~ . ~~ 382 -~~~~~~~~ ~~~ 1.50 ;U ~382 ~3.82 ~~05 _ _ _ _ _ _ 3.82 .. -0 Education Water, 2.50 2.00 2 200 1i.5 200 1. ~150 1 m 10 '~. *~~~~ 100 C"~~~~~~~10 III - Finance Health 3.54 - c .00 4 =0 3.54 .. ~6.00 7, . *E 3.54 4.00 3.5 CO) 2.00 36 2.69 There is clearly urgent need to improve the cash flow planning of ministries which might require more hands on support by the Ministry of Finance. Nonetheless, it would be desirable if government aimed during the preparation of the FY03 budget to improve the preparation of the cash flow requirement by spending units and base budget releases during FY03 on these improved cash flow plans. It is recommended that efforts be undertaken to assist spending units in the preparation of adequate monthly cash flow requirements which then could form the basis from moving from the current rather rigid cash budget system to a system were expenditure releases match more closely actual requirements of the spending units. 37 2.70 Figure 2-11 shows the aggregate cash flow requirements for OC for FY02 as indicated in the cash flow plans prepared by spending units and compares them to the actual pattern of OC releases in the two previous years. Actual monthly expenditure releases for FY00 and FY01 show the typical large month to month variation which are the result of releases for other charges being determined as a residual under the cash budget. It is evident, that the requirements according to the cash management plans differ quite significantly from month to month expenditure patterns observed in previous years, which indicates that the expenditure releases according to cash management plans are likely to allow for greater efficiency in the implementation of expenditure plans than was previously possible. 2.71 In order to be able to Figure 2-11: Cash Flow Requirements for OC for FY02 implement a system of and Monthly Releases for OC for FYOO and FY01 quarterly releases and to release funds according to the cash flow plans rather than 80.00 - - . purely according to monthly 70.00 - . Cash Flow resource availability, the 60.00 Requirement Ministry of Finance will need FY02 cover monthly cash flow gaps Monthly through short term borrowing or 40.00 Releases drawing down of reserves. Last 3. . FY01 year's PER suggested that such a 2 ; * ' Monthly reserve could be replenished leases I from donor program finance. 10.00 l lO ..O. Re0ae l Alternatively and probably - . i7. l administratively more straight- ' iw forward, the Bank of Tanzania is cn Z prepared to offer short time bridge financing of cash flow gaps which would be covered from the next external program finance inflows. The comparison in figure 2-11 shows that given that the cash flow needs differ significantly from historical cash flow patterns and there might thus be need for such bridge financing. However, during the first four months of FY02 sufficient funds were available from program support and bridge financing was thus not necessary. 38 2.6 INTEGRATION OF FOREIGN AID INTO THE BUDGET PROCESS 2.72 Foreign aid plays an important role in Tanzania's economy and in the financing of the government budget. UNDP and OECD estimates indicate that Tanzania receives foreign aid in the magnitude of about 11.3 percent of its GDP or about US$ 30 for every Tanzanian. This puts Tanzania in the middle range of aid recipients in Eastern Africa. While in some countries such as Malawi (ODA is 23 percent of GDP), Mozambique (22 percent) and Zambia (21 percent) foreign aid plays an even bigger role than in Tanzania, other countries such as Kenya (2.9 percent) are for various reasons less aid dependent (Figure 2-12). Figure 2-12: Official Development Assistance to Selected East African Countries, 1999 60 30 20 - 0 I Z.amb&a MoZam&qw, Malawi Taa.ania L gpnd'| E1hi! |UODApru (USS) 63.1 46.5 41.3 30.1 e o ^ j lOoDAas % LGDP 212 22 23 -1i3 .Ž R2| .J0.2 Source:OECD, 2001 Developn-nt Cooperation Report 2.73 The significance of foreign aid becomes also very clear if one compares the level of foreign aid to domestic revenue generation. Tanzania's domestic revenue was between II and 12 percent of GDP during recent years, which implies that resources provided in the form of foreign aid are of a similar magnitude than Tanzania's domestic revenues. Such a large amount of aid has serious implications for the government budget process at all stages and integration of foreign aid into the budget process is one of the key issues if the efficient use of both domestic and foreign resources for poverty reduction is to be enhanced. 2.74 Improved integration of foreign aid is best seen on a continuum that ranges from an improved flow of information between government and donors over capturing foreign aid in the appropriations bill to foreign aid passing through the exchequer system to budget support, where foreign aid resources are fully integrated in the budget. The PER 39 process provides an important platform for government and donors to work jointly towards better integration of donor assistance in the government budget. Work under the PER has focused both on the qualitative and quantitative improvements of the inclusion of foreign aid in the budget process. Progress has been made across ali aspects of integration although it is often difficult to measure this progress. This section analyzes the impact of foreign aid on the various stages of the budget process and documents progress achieved in the recent past. Budget Preparation 2.75 The large amount of donor funding affects both the relevance as well as the quality of budget preparation. One of the key problems in the process of budget preparation is adequate and commonly shared information on budget strategy and donor support. Government requires reliable and disaggregated information on likely donor disbursements over the budget/MTEF period to be able to make efficient allocative decisions. Similarly, donors, especially as budget support gains in importance, increasingly expect and demand credible information on the government's medium term expenditure plans to assess the likely impact of aid and to facilitate the design their aid programs. In this area significant progress has been made on many fronts. The PER process as well as the PRSP process have been instrumental in sharpening government's budget strategy through a participatory process that yielded a widely agreed on set of budget priorities. The three year MTEF lays out the government's budget strategy and indicates the use of domestic and donor resources over the MTEF time horizon. In parallel to this process, the development of the Tanzania Assistance Strategy (TAS) yielded a frarnework for donor assistance of government's priorities. The PER process has facilitated the collection of detailed information on planned donor support for the MTEF period which then could be used in the budget process to effect resource allocations that cover both donor and government's own resources. 2.76 A second, maybe even more severe problem arising from a large volume of aid in comparison to governments own resources concerns the relevance of the government budget as the key resource allocation mechanism. The relevance of the budget and consequently attention to budget preparation by sector ministries was in the past diminished by the fact that domestically financed budget resources covered essentially salaries of public sector employees with relatively little funds available for operations and maintenance and even less for domestically financed investment. Thus, sector ministries expand considerable time and effort on their interactions with donors to secure project financing and relatively less attention to the preparation of the sector budget. This issue is of particular relevance with respect to the development budget which is almost exclusively donor financed. In this area, bilateral negotiations with donors are they key resource allocation instrument. While efforts to capture a greater share of donor finance in the budget are laudable, this inclusion serves mainly to document allocative decisions that have been made outside the budget process. The shift towards budget support represents the most important step towards restoring the primacy of the government led budget process by ensuring that a greater share of domestic resources is allocated through the budget process rather than through other processes. Other aid mechanisms that 40 contain elements of budget support, such as sector programs, still revolve mainly around the interaction between sector ministries and donors and usually limit the contestability of these resources among competing priorities. In addition, capacity constraints in the Ministry of Finance often limit MOF's effective participation in the design and application of these aid mechanisms. 2.77 A further problem that arises from large aid flows refers to the sustainability and recurrent cost implications of donor funded projects. Both on account of limited information on donor flows as well as on account of decisions on project funding being made outside the budget process, the recurrent cost implications of projects are often not fully taken into -account. This problem also persists with the adoption of sector development programs, whose full integration into the budget process has been highlighted as one of the key priorities in the last PER. Budget Execution 2.78 Large amounts of donor assistance also affect the process of budget implementation. Problems arise because of the unpredictability of resource availability as well as the distortion that arise from parallel project implementation mechanism for donor supported projects. 2.79 Unpredictability of the availability of donor resources, especially budget support, presents a main problem during the budget execution stage. These problems arise mainly from the fact that the release of donor resources is generally contingent on government meeting certain conditions, rather than the cash flow needs of government. The uncertainty introduced by conditionality based budget support disbursement is exacerbated by the cash budget system, where monthly expenditure releases are determined by cash availability with wide variations in the amounts of cash available on a month to month basis. This unpredictability is further exacerbated by the uncertainty surrounding the timing of program support disbursements or their complete failure to materialize. In order to reduce this additional uncertainty, the main instrument would be the move from the current cash budget. system to a more standard system where expenditure releases are based on budgeted amounts rather than on resource availability with appropriate safeguards in the form of expenditure and commitment controls and conservative projections on resource availability. It also should be noted that governments cautious move towards a cash flow management system with the possibility of using bridge finance from the central bank to cover monthly cash deficits which then would be covered from expected program support disbursements is an important step towards reducing unpredictability of resource availability for spending units. 2.80 Another problem area in budget implementation arises from the distortion in incentives that is often created by project support. This problem arises from the fact that donor financed activities often afford better working conditions for civil servants, e.g. in the form of an improved working environment, better availability of working materials, or the topping up of salaries, which bias the efforts of civil servants towards donor supported activities and away from activities that are funded from government's own 41 resources. This potentially leads to a distortion of the priorities arrived at during the budget process which may not necessarily be reflected in expenditures on inputs put rather in the outcomes of public sector activities. Accountability 2.81 Donor funding typically comes with its own accountability requirements, which are either parallel or in addition to government accountability mechanisms. However, foreign aid has important implications for domestic accountability processes for the use of public resources in Tanzania. 2.82 The most immediate effect concerns the fact that poor integration of donor assistance in the budget and the financial management processes of government makes it difficult for government to account for the use of all resources. With respect to the activities of the Controller and Auditor General, problems arise from the fact that project documentation is not always made available to the OCAG, even is specific project audits are carried out by private audit firms. In addition, project related documentation is frequently in the language of the donor rather than in English. Furthermore, cases were donor support is either included in the budget but not channeled through the exchequer system or where donor support is not included in the budget but during the financial year availed to government lead to the reporting of over or under-spending in the OCAG reports. 2.83 Aside from the direct impact of donor financed projects on government accountability, there are also important indirect effects. Foremost are the demands on limited domestic auditing capacity posed by donor projects. With donor projects usually being properly resourced for audits by private auditors, the existence of auditing requirements outside the government system contributes to the demand for qualified auditors outside of government and adds to the problems of the OCAG to attract and retain qualified audit personnel. Similar problems exist also in the areas of financial management were the most qualified personnel frequently can be found in project management units with attractive conditions of work rather then in the general government system. Types of Official Development Assistance and Integration into the Budget Process 2.84 The principal forms of official development assistance as classified by UNDP are technical cooperation, investment project assistance, program/budgetary aid or balance of payments support, food aid, and emergency and relief assistance. UNDP estimates indicate that total aid flows to Tanzania amounted to US$ 990 million in 1999, the latest figure available. About 25 percent of this assistance are in the form of technical cooperation, 40 percent are investment project assistance, 25 percent are balance of payments support, and 10 percent are emergency and relief assistance. 42 Figure 2-13: Official Development Assistance by Type of Assistance, 1999 E6ergacywrnd Reief Assistarre Food Ad hnvestsnt Project Assistwe Techical Cooperation 39% 25% Barcenof Paymnws Sipport 25% Source: UNDP, Tanzania Development Cooperation Report, 1999 2.85 The various types of development assistance differ systematically with respect to the degree of their integration into the budget. Food aid (1 percent of total assistance) and emergency and relief assistance (10 percent of total assistance) are generally outside the budget for various reasons. Much of this type of assistance is provided on short notice in reaction to emergencies, the assistance is often provided in kind, and food aid and emergency and relief assistance is frequently channeled through NGOs rather than through government. 2.86 Technical cooperation (25 percent of total assistance) as classified in the UNDP Development Cooperation Report comprises a large variety of activities, ranging from scholarships to support to NGOs to contributions to basket funds for public sector and health sector reforms. A significant part of technical cooperation support is outside government and benefits NGOs or the private sector. Another leakage arises from the fact that a significant amount of technical cooperation goes to sub-national governments is not captured in the central government budget (and generally also not in the budgets of local authorities.) However, even technical cooperation passing through the central government is only partially captured in the budget. One of the reasons for the limited integration of technical cooperation activities into the budget is that much of this support is in the form of experts that are sent from donor countries and directly paid by the donor. 2.87 Project support, which according to UNDP accounts for about 40 percent of development assistance to Tanzania, should ideally be captured in the development budget, although a small part of project assistance is also likely to go to NGOs or the private sector. 2.88 To estimate what share of project assistance provided to Tanzania is actually included in the budget, it is instructive to compare the UNDP estimate for investment 43 project assistance and technical cooperation with the figure for foreign financed development expenditure in the budget. For 1999, the UNDP estimate for investment project assistance and technical cooperation was about US$ 640 million while the amount of foreign financed development expenditures in the government budget was around US$ 270 million. However, as indicated above, a significant share of technical cooperation goes to NGOs and the private sector. Depending on what share of technical assistance is in fact provided to the central governnent, the share of donor funding of technical cooperation and investment project assistance captured in the budget is estimated to lie between 51 and 64 percent of that assistance provided to the central govermnent. Table 2-17: Share of technical cooperation and investment project assistance captured in the budget, FY01 Share of donor funding for TC and IPA provided Share of technical cooperation provided to central government to the central government captured in the budget 75 % of TC outside the government system 64% 50 % of TC outside the government system 57% 25 % of TC outside the government system 51% Note: TC technical cooperation, IPA investment project assistance Source: authors' calculations 2.89 Balance of payments support, with the exception of IMF supportl°, typically represents budget support and as such is fully captured and integrated in the government accounts. Any shift from project to program support implies thus automatically that a greater share of donor assistance is captured in the budget. 2.90 Over the past few years, govermnent has been keen to integrate donor assistance into the budget. The strategy to achieve this includes two principal elements. First is to target a shift from project to program support. Both UNDP and fiscal data indicate that significant progress has been made in this area. The shift is particularly pronounced in the area of grant support, where the adoption of budget support modalities by bi-lateral donors, originally through the multilateral debt fund and more recently through the Poverty Reduction Budget Support Facility, have significantly tilted the balance in favor of program support. With respect to development assistance in the form of credits, the shift is less pronounced since the scaling up of World Bank support during recent years was mainly in the form of project support. However, more recently the Bank has committed significant funds for the support of sector programs and is currently preparing a Poverty Reduction Support Credit with the intention of having the PRSC evolve as the main vehicle for providing Bank support to Tanzania. IO IMF support is typically provided to the Bank of Tanzania which in turn can on-lend funds to the govenmment. 44 Figure 2-14: Program and Project Support as a Share of GDP, Loans and Grants, 1995- 1999 3.0% I 2 0% 2.5% % p 2.0% 1 5% 73 e 0I e|M 00% ___ ____ ___ ____ ___ ____ ___ ____ ___00% FY95 FY96 FY97 FY98 FY99 FY00O Y0 FY95 FY96 FY97 FY98 FY99 FYOO FY01 F jo -Gts gram Gran | Projec Loar UProgram Low, I 2.91 The second element of government's strategy to achieve better integration of donor support into the budget was to request that donors provide information on planned project assistance that could then be reflected in the development budget. Modes of providing donor project support through the budget 2.92 In addition to capturing donor assistance at the stage of budget preparation in the budget, government is also keen to improve the integration of donor support into government process during budget execution. There are three principal ways in which donor support provided to the government of Tanzania is linked to government's financial management system. Development funds are classified as D-funds, C-funds or R-funds where D stands for direct to project, C stands for cash and R stands for re- imbursement. 2.93 D-funds go directly to the projects in the form of equipment, technical assistance, etc. The ministry with the project informs the Accountant General of the amount of assistance received by the end of the year (although formally ministries are supposed to inform the Accountant General on a quarterly basis). The Accountant General then informs the budget section in the Ministry of Finance which records the funds received as dummy fund (for accounting purposes) and issues a release warrant to the Accountant General. Then, the Accountant General issues an exchequer notification and an exchequer dummy (for accounting purposes) to the ministry concerned. Most of the donor support received by Tanzania is in the form of D-funds. For expenditure to be captured in government's appropriation accounts, it is such essential that donors provide adequate information to the ministry implementing the project. In the past, information flows have been often incomplete which was one of the causes for development expenditure reported in the appropriation accounts at the end of the year typically being significantly lower than development expenditure shown in the appropriations bill. The practice of providing information on project assistance received to the Accountant General only at the end of the year rather than at a quarterly basis makes the tracking of progress with respect to the implementation of the development budget difficult. 2.94 C-Funds are deposited into the Accountant General's development revenue account (where all development funds are deposited). The donor who deposited the money/cash issues notification indicating the amount and purpose of the money. The 45 Accountant General informs the budget section which records it (not recorded as dummy but actual cash) and issues release warrant to the Accountant General who in turn issues an exchequer notification and exchequer release to ministry concerned. Donor support provided in the form of C-funds includes general budget support, contributions to sector programs in health and education as well as contributions to basket funds for public and legal sector reform. 2.95 R-Funds represent donor assistance where the government covers expenditure for a donor funded project initially from its own funds and is later reimbursed by the donor. However, due to the tight budgetary situation this mode of donor support is currently not used. Figure 2-15: Development Expenditure Captured in Appropriations Accounts as a Share of Development Expenditure Recorded in the Budget, FY96-FY01 60% 50% I40% 30% 20% 10%n 10% FY96 FY97 FY98 FY99 FY00 FY01 2.96 An indicator of the extent to which development assistance passes through the government accounts is the ratio between development expenditure recorded in the appropriations accounts (which are audited by the Controller and Auditor General) and development expenditure recorded in the development budget at the beginning of the fiscal year. As figure 2-15 shows, the share of project assistance reflected in the appropriation accounts at the end of the year has increased significantly over the past six years. The fact that this ratio has improved over the past few years is however not only due to improved reporting on project expenditures, but it also reflect improved project implementation and more realistic development budget estimates. 2.97 Figure 2-15 summarizes the combined effect of the shift towards budget support and better capturing of project support in the budget and the appropriation accounts. The upper and lower estimates of the amounts captured assume that 25 percent and 75 percent 46 of technical cooperation are intended for the central government and should thus be reflected in the budget. According to theses estimates, between 62 and 73 percent of project and program assistance estimated to be provided to Tanzania is reflected in the budget and about between 47 and 55 percent of that assistance is captured in the appropriation accounts. Figure 2-16: Capturing of Foreign Aid in the Budget 120% - , 100% 100%. 100% - 80% , - 5 60%^ ,; : . , ; ;:, , 47uh d r62% 60%- 20% 0% Inflow s of program and Project and program aid in Project and pogram aid in project aid the budget the appropriations accounts Co upper estmmte bow er estsnt| 2.7 ASSESSMENT OF BUDGET EXECUTION BY THE CONTROLLER AND AUDITOR GENERAL Introduction Box 5: Audit Certificates The Controller and Auditor General issues for 2.98 The Office of the Controller and each audited vote an audit certificate which Auditor General (OCAG) has the indicates whether the supply and development mandate to audit the accounts of central vote, revenue statement, miscellaneous deposit government and local authorities. accounts, and statement of assets and liabilities Separate audit reports are produced for (a) present fairly the financial position of the (a) ministries, departments and regional respective vote (clean audit certificate), (b) directorates and (b) for local authorities. present fairly the financial position of the Current audits carried out by the OCAG respective vote subject to comments made by are essentially compliance audits, but the Controller and Auditor General (qualified the new Public Finance Act No. 6 of audit certificate), or (c) do not present fairly the ls newmpublic Fiane ACotrNo.le 6aof the financial position of the respective vote 2001 also empowers the Controller and (adverse audit certificate). Auditor General (CAG) to carry out I_(adverse_audit_certificate). value for money audits. Nonetheless, CAG audit reports provide important information on the efficiency of budget execution and the integrity of the government accounting system and the extent of leakages. 47 2.99 This section analyses the extent of compliance to prudential budget management procedures and regulations by the Government as indicated by the CAG report covering FY00 and is an update to the previous analyses done for PER FY99 and PER FY00. As in the previous PERs, the analysis of the CAG reports is intended to: establish the prevalence of qualified or adverse audit opinions issued by CAG; identify the main causes of these unfavorable audit opinions; assess the extent of responsiveness to audit queries and follow up action by vote holders; and analyze and explain the patterns for excess spending or under-funding of votes." Main Findings 2.100 Section 25 (2) of the Public Finance Act No. 6 of 2001 requires each accounting officer to prepare and transmit to the Controller and Auditor General an Appropriation Account and various statements for audit purposes, within a period offour months after the end of each financial year. Section 35 (1) of the same Act requires the Controller and Auditor General, upon receipt of the above-mentioned accounts, to (on behalf of the National assembly) examine, enquire into and audit them within a period of nine months or such longer period as the National Assembly may by resolution appoint after the end of the year to which the accounts relate. Similarly, Section 45(4) of the Local Government Finances Act 9 of 1982 states that "As soon as possible after the close of thefinancial year of a local government authority, the authority shall submit its accounts for audit to the auditors, who should complete the audit not later than six months after the close of the financial year." Clearly, this section of the local government law is vague since it does not specify the time, from the close of the financial year, within which the final accounts should be submitted to CAG for auditing. As such LAs have only been guided by the Local Authority Financial Memorandum of 1997, which requires them to submit the accounts to the Finance Committee of the Council for adoption within three months after the close of the fiscal year. From the Finance Committee the accounts are then forwarded to the full Council, where they are approved and signed before submission to CAG for audit. Moreover, the way it stands, the law gives the auditors six months to complete the work of auditing all the LA accounts. However, since LAs need about three months from the close of the financial year to submit their accounts to the auditors, the auditors remain with only three months to do their work, although in practice they have utilized their statutory six months. Despite the existence of various procedures and provisions in the law, production of final audit reports, especially for the central government, has been hampered by delays. According to Assistant CAG in-charge of Regions and Local Government accounts no delays have been experienced in the production of audited reports for LAs in the past four years, although the FY98 report bears year 2000 as its publication date while the FY00 report bears no publication date. " The Country Financial Accountability Assessment (CFAA) carried out in 2001 provides a detailed analysis of the strengths of financial accountability processes in Tanzania, covering both the public and the private sectors, including an assessment of the government accounting and audit system. 48 2.101 The delays in production of audit reports have emanated either from late submission of the accounts to CAG by the accounting officers or late examination and auditing of accounts by CAG, or both. Table 2.18 shows the required and actual dates on which the CAG reports were produced during FY97 - FY00. It is clear from these dates that the delay of production of CAG report ranges from 3 months to 6 months for the central government and from zero to four months for the local government. Lack of capacity at the OCAG is one of the reasons for the delay, but the other reason is delay in submission of accounts to CAG by the accounting officers, as shown in Table 2.18. The number of central government accounts not submitted by the stipulated deadline (3 1st October), though has declined in recent years from 43 percent of total submissions in FY98 to 33 percent in FY00, is still high. Looking at the evidence from CAG reports for FY97 - FY00, it is noticeable that most of the delayed accounts are submitted a month late (about 69 percent on average), a number of them are submitted two months late (17 percent) and few of them are submitted three or more months late (14 percent). Similarly, the number of local authorities' accounts not submitted by the stipulated deadline (31st March) has been declining, from 31 percent of total submissions in FY97 to 14 percent in FY00. As commonly remarked by CAG in various reports, more effort is required to submit appropriation accounts by the stipulated deadlines of 315t October for the central government and 3 Is' March, every year. Table 2-18: Submission of Accounts and Production of CAG Report, FY97 - FY00 Preparation date FY97 FY98 FY99 FYOO Central Government Required July 98 July 99 July 00 July 01 Actual October 98 October 99 December January 02 00 _ _ _ _ _ _ Delay 3 Months 3 Months 5 Months 6 Months No. of Accounts submitted on time 37 36 38 43 No. of Accounts submitted late 25 27 26 21 Late submission as % of total 40 43 41 33 submissions Local Authorities Required October 98 October 99 October 00 October 01 Actual October 98 February 00 October 00 October 01 Delay No delay 4 Months No delay No delay No. of Accounts submitted on time 61 88 86 101 No. of Accounts submitted late 27 23 26 16 Late submission as % of total 31 21 23 14 submissions Source: CAG Reports (various) 2.102 In FY00, audit certificates of Central Government (ministries & regions) accounts and statements issued by CAG continued to be predominantly adverse (Table 2.19). About 58% of all accounts received adverse certificate, 18% received qualified certificates and the remaining 24% received clean certificates. Among ministries, the share of votes which received a clean certificate increased from 25 percent in FY99 to 31 49 percent in FY00; those which received a qualified certificate decreased from 40 percent to 26 percent; while those that received an adverse opinion also increased from 35 percent in FY99 to 43 percent in FY00. For regions, there was a substantial decrease in the percentage of accounts which received an adverse opinion from 91 percent in FY99 to 82 percent in FY00. Regional accounts which received clean certificate increased from 9 percent to 13 percent, while accounts which received qualified opinion rose from 0 percent in 1999 to 5 percent in 2000. In FY00, audit certificates of Central Govemment (ministries & regions) accounts and statements issued by CAG continued to be predominantly adverse (Table 2.19). About 58% of all accounts received adverse certificate, 18% received qualified certificates and the remaining 24% received clean certificates. Among ministries, the share of votes which received a clean certificate increased from 25 percent in FY99 to 31 percent in FY00; those which received a qualified certificate decreased from 40 percent to 26 percent; while those that received an adverse opinion also increased from 35 percent in FY99 to. 43 percent in FY00. For regions, there was a substantial decrease in the percentage of accounts which received an adverse opinion from 91 percent in FY99 to 82 percent in FY00. Regional accounts which received clean certificate increased from 9 percent to 13 percent, while accounts which received qualified opinion rose from 0 percent in 1999 to 5 percent in 2000. 2.103 During the year 2000 final accounts of 117 local authorities were audited. These local authorities (LAs) include 25 urban councils and 92 district councils. The state of the LA accounts slightly improved in year 2000 with the share of councils that received clean certificates increasing from 8 percent in 1999 to 14 percent in 2000. The share of councils which received qualified certificates decreased from 41 percent in 1999 to 22 percent in 2000, while those that received adverse opinions increased from 51 percent in 1999 to about 64 percent in 2000 (Table 2.19). Overall the status of the local government accounts in 2000 was still poor with the majority (about 64%) receiving adverse certificates. 2.104 The response to CAG's audit queries showed an improvement for both ministries and regions in FY00. Satisfactory replies and subsequent redress by vote holders to queries from OCAG continued to be a very small proportion of the total queries issued. Overall about 26 percent of queries issued were replied to in FY00. This is a significant increase over the period FY96 - FY99 when overall an average of about 11 percent of issued queries were replied to (Table 2.20) and it is an indication that some action seems to have been taken by accounting officers to rectify the recurring weaknesses in expenditure control and other compliance issues. However, with about 74 percent of audit queries not replied to in FY00, more actions need to be taken to rectify the situation. The implication of this situation is 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 of non-responsiveness to the CAG's queries is basically the absence of the will to enforce accountability and a lack of appropriate legal instruments for enforcement. Unfortunately, as noted in the CFAA, the new Public Finance Act, 2001 does not contain any means of ensuring that the CAG findings and queries are followed up. In addition, aside from legal changes and strengthened capacities, improvements in financial management will also require changes 50 in the incentive structure and operative culture to more accountable and transparent behavior in the civil service and by bringing such accountability increasingly into the public domain. Involvement of the press, donors, political institutions, and non- governmental organizations in the review of public expenditure effectiveness will, thus, continue to serve as strong stimulants of prudence and accountable behavior. 2.105 The FY00 was marked by the continued prevalence of expenditures that are not vouched or improperly vouched. However, incidences of unvouched expenditures and improperly vouched expenditures decreased. Improperly vouched expenditures as a share of total expenditure decreased from 11 percent in FY99 to 6 percent in FY00 and the share of unvouched expenditures to total expenditure decreased from 2 percent in FY99 to 0.3 percent in FY00 (Table 2.21). Overall, the share of non-vouched and improperly vouched expenditures to the total expenditure decreased to 6.5 percent in FY00 from 12.5 percent in FY99. 2.106 A rise in incidences of embezzlement of cash and stores was noted in FY00. The magnitude of cash and stores losses increased from 0.07 percent of total expenditure in FY99 to 0.16 percent in FY00 (Table 2.22). Embezzlement of cash and stores remain high mainly on account of non-production or non-completion of bank reconciliation statements and poor procurement and stores management. With the IFMS and the new Procurement Act now in place, this problem is expected to decline over time. 2.107 The report of the CAG also shows a significant increase in excess spending and savings with respect to recurrent expenditures in FY00. The level of excess spending on the development account also increased, while savings on the development account remained high, even though they showed a small decline. 2.108 Excess spending occurs in specific votes due to either additional spending on a line item in excess of the budget allocation or unauthorized reallocations across specific line items. Excess spending on the recurrent account increased from Tsh. 9.3 billion in FY99 to Tsh. 25 billion in FY00. The relative large amount of excess spending on the recurrent account under the cash budget system, where overall spending levels are tightly controlled, implies relatively large deviation of actual expenditure patterns from the budget. With the introduction of the requirement that intra-vote reallocations need to be pre-approved by the Ministry of Finance, excess spending should decline. The increase of excess spending on the development account from Tsh. 0.6 billion to Tsh. 6.5 billions is likely to represent donor funding of projects where actual funding exceeded the funding levels indicated in the budget. 2.109 Savings occur if expenditure on a certain budget item is below the budgeted amount. On the recurrent account, savings increased from Tsh. 33.2 billion in FY99 to Tsh. 66.3 billion in FY00. This reflects the fact that under the cash budget system expenditure releases in FY00 were significantly below budget due to the fact that domestic revenue collection was also significantly below the budgeted amount. Savings on the development have been a perennial problem due to poor integration of donor project finance into the governnent accounting system. However, in this area the CAG 51 reports show a significant improvement over the past two years. Savings decreased from Tsh. 205.8 billion in FY98 to Tsh. 125.4 billion in FY99 and further to Tsh. 116.2 billions in FY00. Savings on the development account occur mainly because of: unrealistic estimates; projects, which face poor implementation due to under-funding or other implementation bottlenecks; unrecorded expenditures; non-recording or non- charging of aid expenditure for 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 of goods, materials and equipment. 2.110 Local authorities own revenue which was not properly accounted for in 2000 was about 19 percent of total actual own revenue, a decrease from 30 percent noted in 1999 (Table 2.24). The main reason for this shortfall is that revenue estimates of local authorities are typically overoptimistic. (This over-optimism is likely to be a combination of over-estimation of the revenue base, overestimation of ability to collect revenue, and underestimation of leakages from revenue collection to the transfer of collected funds into the public coffers.) In addition, non-banking of revenues collected failure to produce revenue books also contribute to the large share of local authorities' revenue which is not properly accounted for. With respect to expenditures of local authorities, a slight decrease in questionable payments within local authorities' accounts was noted during 2000 (Table 2. 25). 2.111 Poor certification and losses are mainly a result of weak controls, extra budgetary expenditures and irregularities. In particular, the immediate and underlying causes of the most frequent audit queries include: incidences of excess votes; failure by the accounting officers to reconcile bank accounts; inadequate supporting documentation of 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 of the above-mentioned causes of audit queries. 52 Table 2-19: Summary Of Audit Certificates Issued, FY96-FYOO Accounts and Statements awarded Accounts and Statements awarded Accounts and statements issued an Clean Certificate Qualified Certificate Adverse Opinion FY96 FY97 FY98 FY99 FY00 FY96 FY97 FY98 FY99 FY00 FY96 FY97 FY98 FY99 FY00 Ministries 69 48 40 16 37 10 4 11 25 32 55 82 43 22 52 as a % of all 51% 36% 43% 25% 31% 7% 3% 12% 40% 26% 41% 61% 46% 35% 43% Ministries Regions 31 20 0 7 10 8 10 0 0 4 41 46 60 73 63 as a % of all 39% 26% 0% 9% 13% 10% 13% 0% 0% 5% 51% 61% 100% 91% 82% Regions Ministrie & 100 68 40 23 47 18 14 1I 25 36 96 128 103 95 115 Regions as a % of all 47% 32% 26% 16% 24% 8% 7% 7% 17% 18% 45% 61% 67% 66% 58% Ministries and Regions 1996 1997 1998 1999 2000 1996 1997 1998 1999 2000 1996 1997 1998 1999 2000 Las 8 20 23 10 16 9 30 20 50 25 23 37 65 62 72 asa%ofLAs 20% 23% 21% 8% 14% 23% 34% 19% 41% 22% 58% 43% 60% 51% 64% Table 2-20: Summary of Queries Issued and Percentages Replied and Not Replied, FY96-FYOO Ministries/De FY96 FY97 FY98 FY99 FY00 pt. & Regions .__ _ .__ _ _ _ _ _ Numbe Not Replied Number Not Replied Numbe Not Replied Number Not Replied Number Not Replied r of Replied (percent) of Replied (percent) r of Replied (percen of Replied (percen of Replied (percen Queries (percen Queries (percen Queries (percen t) Queries (percent) t) Queries (percent) t) t) t) t) Total 1790 92.2 7.8 1640 90.1 9.9 1824 80.9 19.1 2043 83.0 17.0 1739 68.3 31.7 Ministries Total Regions 1590 92.2 7.5 1434 90.2 9.8 1265 96.8 3.2 1147 91.5 8.5 790 85.3 14.7 Grand Total 3330 92.3 7.7 3074 90.1 9.9 3089 87.4 12.6 3190 | 86.0 L 14.0 2529 1 73.6 [ 26.4 53 Table 2-21: Unvouched and Improperly-Vouched Expenditures, FY96-FYOO YEAR UNVOUCHED IMPROPERLY- TOTAL UNVOUCHED & TOTAL VOUCHED IMPROPERLY-VOUCHED EXPENDITURE Absolute Percent of Absolute Percent of Absolute Values Percent of Values Total Values Total Total Expenditure Expenditure Expenditure FY96 1,344 0.3 13,903 3.3 15,247 3.6 423,310 FY97 6,005 1.1 25,676 4.5 31,681 5.5 571,948 FY98 5,465 0.8 40,363 5.9 45,828 6.7 682,437 FY99 15,690 2.0 82,192 10.5 97,882 12.5 780,563 FYOO 2,235 0.3 54,866 6.2 57,100 6.5 881,193 Table 2-22: Cash and Stores Losses (in Tsh. million), FY96-FYOO Year Total Expenditure Total Loss of Cash Total Loss as Percent of and Stores the Total Expenditure FY96 423,310 73 0.02 FY97 571,948 971 0.17 FY98 682,437 751 0.11 FY99 780,563 567 0.07 FYOO 881,193 1,387 0.16 54 Table 2-23: Summary of Excess Spending and Saving, FY96-FYOO Excess (Tsh. Million) Ministries/De Recurrent Development pt & Regions FY96 FY97 FY98 FY99 FYOO FY96 FY97 FY98 FY99 FYOO Total 13,753 0 3,939 4,569 24,095 495 91 0 642 6,528 Ministries Total Regions 0 197 16 4,635 0 0 0 675 0 0 Grand Total 13,754 197 3,956 9,205 24,095 496 92 675 642 6,528 Saving (Tsh. Million) Recurrent Development FY96 FY97 FY98 FY99 FYOO FY96 FY97 FY98 FY99 FYOO Total 33,615 56,930 42,293 16,432 64,015 98,359 70,141 188,792 106,121 96,673 Ministries Total Regions 8,713 1,454 665 16,792 2,261 17,983 8,787 17,012 19,307 19,541 Grand Total 42,329 58,383 42,958 33,224 66,276 116,342 78,928 205,804 125,428 116,214 55 Table 2-24: Accounting for Revenues - Local Authorities, FY98-FYOO Category Number of Amount Amount Shs. Amount Change of Councils Shs. millions Shs. 2000 over (Year 2000) millions (Year 1999) millions 1999 (Year (Year 1998) .__ _ 2000) Revenue not accounted for 59 372 516 532 -28% Revenue collected but not 10 212 208 279 2% banked Short falls in Revenue 114 12602 18639 5566 -32% collections Revenue books not produced of Fixed Fees Receipts 106 992 1266 1595 -22% Missing bicycle 21 47 55 49 -14% stickers Mikokoteni Stickers 1 0.2 TOTAL 14225 20683 8022 -31% LAs own Revenue not properly accounted 19% 30% 12% for as % of LAs own Actual Total Revenue Source: CAG Report (various issues) Table 2-25: Questionable Payments - Local Authorities, FY99-FYOO Category Number of Amount Tsh. Amount Change of 2000 Councils million 2000 Tsh. million over 1999 (Year 2000) 1999 Unauthorised Expenditures 0 0 745 -100% Unvouched Expenditure 75 1335 2257 -41% Payments not supported by proforma 51 434 631 -31% invoices Improperly vouched expenditure 102 5337 3864 38% Recoverable overpayments 2 5 17 -65% Irregular payments 16 401 240 67% Withholding Tax Evaded 6 3 14 -82% Statutory deductions not paid over 9 427 332 28% Irregular transfer of funds 23 679 659 3% Nugatory Expenditure 4 10 TOTAL 8623 8760 _ Source: CAG Report (various issues) 56 3. EXPENDITURE AND SERVICE DELIVERY TRACKING 3.1 In previous years, the PER working group had commissioned various expenditure tracking studies, which traced the flow of funds from the Ministry of Finance to the districts and to service delivery units. The information provided by these studies was instrumental to highlight weaknesses in the system underlying the flow of funds in the government system and has also given rise to a number of important reforms, including the regular publication of amounts of money released by the Ministry of Finance to districts for specific purposes such as health, education, or road maintenance. These tracking studies have also provided the impetus for government's plan to strengthen its own capacity for expenditure tracing by setting up an expenditure tracking system that would provide regular and timely information as to whether funds released by the Ministry of Finance to the districts are indeed used for the intended purpose. Being able to control and properly monitor the flow of funds and their use at the district level is important since government grants to the district finance the bulk of basic service delivery in the education, health, water, roads, and agriculture sector, which have been identified as key priorities under the PRSP. In addition, with the increased availability of program support funds and the debt relief under HIPC, Tanzania's development partners also rely on information on the proper use of budgetary funds.- 3.2 Thus, one of the main objectives of the external evaluation was to take stock of and assess the existing expenditure as well as service delivery tracking system in Tanzania as an input into the government's plans for strengthening its own tracking system. This review of expenditure tracking systems follows two related reviews undertaken in 2001, the CFAA and the IMF/World Bank assessment of capacity to track poverty-reducing expenditure. The two reviews dealt mainly with the capacity of the government to monitor allocative efficiency and the integrity of the transfer mechanisms across levels of government. This section assesses (i) the extent of coverage of the full tracking chain, (ii) the coherence of the tracking system, (iii) the institutional roles across layers of government, (iv) incentives for compliance to reporting, and (v) provisions for analyzing in a coherent fashion the information flowing from the tracking system to inform policy and action plans. 3.1 THE TRACKING CHAIN 3.3 The complete chain of tracking public service provision entails five main links: (i) Ensuring that the allocation of resources for specific purposes - sector and activity is according to an agreed prioritization framework and budget. Much of the traditional PER work to date has focused on this link in the chain. 57 (ii) Ensuring the integrity and efficacy of the link between allocations and releases and transfer mechanisms to facilities supplying services, through layers of govermnent institutions and the banking system. The purpose here is to determine leakages, blockages and delays in the flow of resources to the facilities and find ways to resolve these constraints. The assessment critically depends also on the reverse flow of documentation and accounting for actual expenditures. (iii) Assessing the efficiency of the service delivery unit in converting allocated resources into real supply of services. Action plans are the main instruments mapping inputs to outputs, often in the form of contracts for delivery of specified services against budgets. (iv) Assessing the quality and availability of services delivered from the demand side - e.g. based on information from user score cards or service delivery surveys conducted by independent agencies. (v) Assessment of impact of the delivered services in terms of the effectiveness of the delivered services in improving outcomes - e.g. improving livelihood, reducing poverty in their various dimensions. 3.4 The first two links in the chain, (i) and (ii) above, make up the domain broadly understood to be expenditure or financial tracking and is the main concern of the Ministry of Finance and the PER process to date. The rest of the links, (iii) - (v), track the efficacy and impact of resource application for development or welfare enhancement, the ultimate purpose for which they were availed. The Civil Service Department (CSD) and the sector ministries/agencies have particular interest in these dimensions given their primary responsibilities for implementing policy, setting standards and monitoring outcomes. The Poverty Monitoring System now being established will presumably establish the needed coherence by integrating financial and service delivery information and make assessment of the impact of public service provision on poverty reduction and livelihoods of Tanzanians. 3.5 The Ministry of Finance has decided to internalize and organize regular tracking of block grants to the local authorities and their utilization. Draft terms of reference have been prepared for a baseline assessment of the adequacy of the government procedures and channels for disbursing funds and capacity for reporting on pro-poor expenditure at all levels of govermnent. Private consulting firms will be retained to conduct the country- wide review, itself divided by zones. Treasury has issued a circular (June 2001) to all accounting officers in regions and local authorities requiring them to submit quarterly spending reports following a format to be circulated early 2002. The latter is being issued to standardize the reporting fornats, which are currently diverse. 3.2 OVERVIEW OF EXISTING TRACKING SYSTEMS IN TANZANIA 3.6 The existing tracking system follows the two broad categories identified above - financial and service delivery. Past reviews of the capacity to track expenditure focused on the first category. These include expenditure tracking carried out under the PER process, the CFAA conducted in 2000/2001, and more recently the IMF/World Bank 58 assessment of capacity to track poverty-reducing expenditure. Similar reviews for tracking service delivery have not been carried out. 3.7 A wide range of administrative expenditure and service delivery monitoring systems exists in Tanzania across the various levels of government. These form an important base for developing a more coherent and integrated monitoring system. A major concern, however, is the wasteful duplicative monitoring/tracking systems in a context of weak and stretched managerial capacity, a result of a history of multiple project and program management systems. The reports though substantive vary in quality across programs and districts. Some programs e.g. the Health Basket Fund use such reports (covering the whole sector) and related reviews/audits as triggers for the next tranche release, tilting the accountability effort towards it. 3.3 ALLOCATION AND TRANSFER MECHANISMS 3.8 Reviews under the PER process are a regular feature of the government-led annual process. On the one hand, they focus on assessing the allocation of public resources through the budget and MTEF to ascertain whether they were in line with agreed priorities for reducing poverty. Such priorities are sector and activity specific and those aimed at addressing cross-cutting issues deemed important for overall achievement of PRSP targets e.g. HIV/AIDS, gender-sensitivity, governance and environmental concerns. The external evaluation component of the PER process is the main vehicle for this undertaking. On the other hand, specific PER studies are also commissioned by the PER Working Group to track resource flows down to the service delivery units. They intend to establish the integrity of the transfer mechanism and the efficacy of the system to target intended activity. 3.9 The recent IMF/World Bank review of the capacity to track pro-poor expenditure was part of a broader initiative to assess such capacity in 25 HIPCs. It was based on 15 benchmarks of good practice for expenditure management. The benchmarks covered budget formulation, execution and reporting. The details included the allocation of public spending at the macroeconomic level and the efficacy of the transfer mechanism for targeting poverty reducing spending at the sub-national level. The review also assessed the flow of information from sub-national entities to the center on the use of public funds at the local level. Tanzania met 8 of 15 benchmarks, and requires some upgrading work, which is well underway. Out of the 24 countries'2 assessed Tanzania was categorized in the top group of eight with scores ranging between 8 and 9 (for Uganda only). 3.10 According to that assessment, the strengths of the Tanzanian public expenditure management system are largely in the areas budget formulation, execution at the central level, supported by the introduction of the Integrated Financial Management System 12The countries included Benin, Bolivia, Burkina Faso, Cameroon, Chad, Ethiopia, The Gambia, Ghana, Guinea, Guyana, Honduras, Madagascar, Malawi, Mali, Mauritania, Mozambique, Nicaragua, Niger, Rwanda, Sao Tome and Principe, Rwanda, Senegal, Tanzania, and Uganda. 59 (IFMS) at the central government level which facilitates expenditure control and monitoring and the PER process which supports budget formulation. Also instrumental were transparency/disclosure of information associated with the participatory PER process and recent decisions to publicize allocations. The main weaknesses identified by the review were associated with inadequate capacity for audit functions (internal and external), sub-national capacity for reporting spending and integration of external resources into the budget as well as the Exchequer system. The recommended measures for addressing remaining weaknesses were consistent with those identified under the CFAA and reflected in the CFAA action plans. The action plan is comprehensive, and given a wide-range of ongoing and planned external TA projects, no major gap in donor assistance is indicated. 3.11 As for priority expenditure monitoring, a key issue relates to definition of priority sector spending (as broadly shown in Annex 3). As part of the PRSP and PER process, government has developed a list of priority expenditures which are monitored under the PRSP process. However, the coverage of priority expenditures will need to be reviewed on an ongoing basis as part of the annual preparation of the Poverty Reduction Strategy progress report and MTEF. Particular issues to be addressed this year relate to the identification of key spending areas in Justice system, review of Agriculture on the basis of the agricultural sector development strategy and plan, and ensuring fully capturing of HIV/AIDS expenditures including spending in non-priority sectors 3.4 SERVICE DELIVERY MECHANISMS 3.12 Three types of institutions track service delivery. These institutions have a primary responsibility for implementing policy, setting and enforcing standards, and monitoring implementation as well as outcomes. These are CSD, sector ministries and related agencies, and the layers of institutions under the Presidents Office Regional Administration and Local Governments (PORALG). They provide coverage of service delivery in MDAs with performance budgets/contracts with CSD, and some coverage at the sector level. No mechanism exists to review service delivery at local level. However, PORALG is planning to establish an inspectorate unit to cater for this. 3.13 The CSD through its performance improvement program enters into contractual agreements with ministries, departments and independent public agencies (MDAs) for the delivery of specified services against budgets for financing the same. The MDAs prepare and agree with CSD on annual action plans and a performance budget. To date, MDAs that have entered into such contractual arrangements include: CSD, Ministry of Water, Ministry of Agriculture, Ministry of Education, Ministry of Health, Ministry of Finance, Ministry of Water and Livestock Development, and President's Office Planning and Privatization. The MDAs are required to prepare reports of performance based on these agreements for review by the CSD. A performance improvement fund is being set up to provide incentives for the enhancement of service delivery capacity in the MDAs. In addition, the CSD has conducted the first independent service delivery surveys to assess the impact of actions taken by the MDAs on the quantity and quality of public services. The survey, conducted by Price Waterhouse Coopers in November 2000, covered the 60 CSD itself, the Ministry of Agriculture and the Ministry of Water, i.e., the three MDAs earmarked for piloting the performance improvement fund. The survey focused on the client assessment of services offered by the three MDAs. 3.14 All regions and districts prepare annual action plans and budgets and submit these to PORALG. All regions and local authorities are also required to submit quarterly performance reports. These reports cover quarterly financial expenditure and in the case of health and education also the status of implementation of various activities delineated in the action plan. The health sector maintains a management information system covering service delivery and some impact of services on disease prevalence delivery of service. In the case of health services quarterly independent audits are conducted by a joint team from PORALG, the sector ministry and the regional secretariat. 3.15 Vis-a-vis local authorities, sector ministries are responsible for policy, setting standards, monitoring and evaluation, and supporting capacity building in all entities delivering specific services. While in the past these ministries were also conduits of financial resources to the sub-national entities delivering basic services, changes are under way to channel resources from Treasury directly to the local authorities. The review of tracking systems paid particular attention to the system for tracking service delivery for two sectors - health and education. 3.5 FINANCIAL AND SERVICE DELIVERY TRACKING IN PRACTICE Resource Transfers 3.16 After resources are allocated for a specific purpose, there are two ways through which such resources are transferred from the central government down to service delivery units. The first mode of resource transfer is through cash transfers from the Treasury to local authorities, either directly or indirectly through the sector ministries. In the case of block grants, resources are transferred directly from the Treasury to respective LA accounts in the National Microfinance Bank (NMB) accompanied by a note on how such resources should be utilized. Block grants are intended to cater for wage (PE) and non-wage (OC) expenditures. Grant money from the Treasury is first deposited into an collection account of the local authority. Once verification has been done, the respective amounts are sent to separate departmental accounts ready for appropriation. Out of the available funds, salary payments are effected automatically while expenditures for supplies and services follow the standard government process initiated -by the department concerned. The chain for the transfer of donor resources (for example the Health Basket Fund) used to be long, starting with the transfer of resources from Treasury to PORALG and then to a special LA account (Health Basket account) in the NMB. This chain resulted in long delays of up to four months before resources were received by LAs. Beginning in 2001, however, this chain has been shortened with resources now flowing directly from the Treasury to the LA Health Basket accounts. The utilization of such resources is guided by a specific LA Health Basket action plan as agreed between the LA, donors and MOH. 61 3.17 The second mode of resource transfers from the center to local authorities and service delivery units are in-kind transfers which involve a combination of cash transfer at the level of central government and transfer of in-kind materials (drugs and text- and exercise- books) purchased by the central government (MOH and MOEC, respectively) to the LAs. 3.18 In the case of medical supplies, upon receipt of funds (warrant of funds) from the Treasury, the MOH issues a cheque to the Medical Stores Department (MSD) accompanied by credit allocations to health facilities. The allocations cater for drugs, equipment and kits. On receipt of the cheque and credit allocations, MSD deposits the money in its bank account and credits the respective health facilities' accounts maintained by MSD. Upon receipt of the allocation list from MOH, spending units/health facilities place an order for the supply of drugs/medical supplies/equipment not exceeding the credit amount available to them. Whereas hospitals place their orders for purchase of drugs or equipment on a demand basis ("pull" system), health centers and dispensaries have no such a choice - they receive standardized drug kits on a monthly basis under the "push" system. An "indent system", currently under pilot in the Morogoro region, has proved to be an improved system. The system is demand-based and thus allows LAs to choose the type of drugs that are in demand depending on disease pattern of a given location. 3.19 Beginning in 2001, the MOEC has shifted the procurement of exercise and textbooks to LAs. Under the previous system, procurement was carried out centrally by the MOEC, which then distributed the school materials to school facilities all over the country. Under the new system procurement is managed by each LA, which upon receipt of funds from the Treasury, floats the tender and competitively chooses the supplier among various bidding publishers. Reporting 3.20 Central Government: There are distinct reporting systems in place, catering for each level of resource transfer and/or use. At the Treasury all resource transfers are recorded and reports generated automatically using the IFMS on a daily, monthly and quarterly basis. To the extent that all MDAs and regions are connected to the IFMS, any transfers of resources undertaken by these institutions to other levels of the government are also captured in the IFMS. All central government accounts are audited internally by government internal auditors and also externally by the CAG. 3.21 Local Government: LAs compile expenditure reports on a monthly and quarterly basis. The reports provide the status of budget performance including grant receipts and own revenue collections. The monthly reports are discussed in the finance committee of the council, while quarterly reports are discussed by the full council. After discussion and agreement at the council level, copies of the quarterly reports are provided to the Regional Secretariat and PORALG. For funds under the Health Basket Fund the report on utilization of funds is sent to PORALG and the group of donors contributing to the Fund. Normally, such reports are prepared on a timely basis, because they have to act as 62 a trigger for the next disbursement of funds. Occasionally, PORALG working closely with the sector ministry and Regional Secretariat (mainly Economic and Planning Division) visits samples of the councils to discuss the report and follow-up to the issues arising. In the case of education, quarterly reports on the utilization of various education funds are sent to MOEC with a copy to PORALG. Generally, the practice of sending copies of the LA quarterly reports to Treasury existing up to the early 1980s stopped in the 1990s but is now being re-instituted and will probably become more effective beginning early 2002 after the standard reporting format is circulated. As of now, in the absence of physical tracking of resource flows from the Treasury down to the service delivery units, MOF has no other mechanism to provide feedback on whether resources are being utilized as per intended purpose. 3.22 LAs have internal audit departments, whose function is to help management discover any irregularities early and to improve and safeguard accounting practices. The internal audit department is supposed to be independent and have access to accounting records all the time. However, in most LAs the internal audit units have only been introduced recently, are poorly manned, and the majority of them are managed by newly recruited internal auditors with little experience in the accounts and financial management systems of LAs. The accounts of LAs are also externally audited. On the basis of legal stipulations, local authorities prepare annual financial accounts for submission to the Controller and Auditor General (CAG) within three months after the closing the financial year. The accounts are then audited within nine months after the closing of the financial year, and an audit report prepared by CAG indicating, among others, the quality of accounts of the for each LA (whether clean, satisfactory or adverse). 3.23 To facilitate transparency and expenditure tracking, PORALG is emphasizing participatory planning, monitoring and evaluation. In this regard, a national framework for participatory planning and a planning tool (O & OD) has been put in place. Moreover, PORALG has instructed all LAs to announce on notice boards all funds disbursed by the Treasury and the councils. Copies of the disbursements are also to be made available to the Members of Parliament (MPs), Regional Commissioners (RCs) and District Commissioners (DCs) for them to be able to track the expenditures being made by the LAs under their administrative jurisdiction. 3.24 Facilities: Various reports are also prepared by service delivery facilities for purposes of reporting the utilization of resources and or in-kind materials. For example, on a monthly basis and on demand, the MSD sends to health facilities and the MOH a statement of accounts showing an itemized listing of all credit allocations to that unit, the value of drugs/supplies furnished to that unit during the period and any balance of funds available for future spending. The statement to MOH shows money received from them and the utilization of that money (credit allocations) and any balance remaining thereof. Health facilities on their part prepare reports indicating the utilization of both cash (their allocation of block grants, health basket funds, cost sharing funds) and in-kind materials (for health centers and dispensaries), the number of patients attended and the ten most prevalent diseases during the month/quarter. The reports, which are prepared following the format provided by the Health Management Information System (HMIS), are sent to 63 the LA for compilation and consolidation. Education reports are also prepared at the facility (school) level on a monthly basis. The reports are sent to the LA and indicate sources of funds and their utilization, various in-kind contributions, students attendance, number and qualifications of teachers, self reliance activities/projects, and pending requirements (classrooms, teacher houses, offices, toilets, stores). On a day to day basis, the facilities also keep records in the form of ledger, which indicate the utilization of materials purchased at the facility level or received in an in-kind form from higher levels of govermment. The facilities are audited by an LA internal auditor on occasional basis. 3.25 Monitoring (Physical Inspection, etc.): As part of control and monitoring measures to ensure that funds that are transferred to the service delivery units are used to provide the intended services, sector ministries do make physical inspection visits to see the outputs. In the case of the Health Basket Fund (and Road Fund) PORALG, working closely with the sector ministry and respective Regional Secretariat, visits samples of the councils to, inter alia, discuss the reports and inspect/follow-up on the implementation of agreed activities under the action plan. To emphasize on this, guidelines have been prepared and issued to all LAs regarding the use of the health basket fund and the road fund. The Inspectorate Division of MOEC, using its zonal officers, has also the mandate to make physical visits to the facilities (schools) to inspect/follow-up on implementation of agreed activities under action plan. Physical visits are also carried out by the LA management and sector management teams. There are also health management boards at all health facilities and school conmnittees that form the apex of the facility's management organizational structure and are charged with ensuring that the facilities are efficiently run and appropriate services are provided to clients. With regard to other donor funded programs implemented in the local authorities, PORALG is currently undertaking an inventory of the programs so as to know where they are, how much they cost and which donors are funding them. Quality of Reports and Weaknesses in Reporting Mechanism 3.26 Overall, government has in place a variety of monitoring and reporting mechanisms which if properly implemented and coordinated provide a good basis for comprehensive monitoring of public expenditure performance. Various reports are prepared that act as a feedback and control mechanism on the flow and use of public resources. However, several weaknesses exist: (i) The reports that are prepared vary in quality between various levels of government and facilities and across similar sources. For example reports that are prepared using the IFMS are far more advanced than reports being prepared at all other levels of the government. Across similar levels of government, reports that originate in urban centers (municipals and urban based service delivery facilities) are far more advanced than reports that are prepared by rural based district councils and service delivery facilities. Differences in the availability of working facilities (computers, etc.) and human capacity (professional staff) are the major reasons behind differences in quality of reports. In general the quality of financial and budget reports for the LAs has been unsatisfactory as indicated by the adverse opinion on 64 the accounts of many LAs by the CAG. As documented in past PER reports, the major problems are poor financial management and poor record keeping done by the weak/unqualified accounting technicians with no orientation in the accounts of LAs. While most LAs have fairly qualified budgeting and accounting personnel at the higher levels, the quality of accounts clerks is poor both in terms of basic education and orientation in LA accounting and finance. This partly accounts for the variations in the quality of record keeping and financial reports across LAs. (ii) Almost none of the LA reports find their way to the Treasury, despite the latter being the source of the bulk of resources that are transferred by the central government to the lower levels of government for public service provision. Most of the LA reports go directly to and end up at PORALG. Few of the reports are copied to the Regional Secretariat and sector ministries, while donor-related reports are sent to donors with a copy to PORALG. The Treasury has issued a letter on June 21, 2001 to all accounting officers including LA Directors requesting them to submit quarterly reports as per format provided. However, to date no performance reports have been submitted to the Treasury, although these reports are available at district level and at PORALG. (iii) Institutional Roles: There is a lack of clarity on role of the Regional Administration Secretariat (RAS) in the government hierarchy and in the tracking/reporting chain. Again as documented in the past PER reports, the RAS office's advisory role is currently very limited and ambiguous, leading some to think that the Regional Secretariat is an unnecessary overhead to the government. However, given the distance between ministries and the LAs, the RAS is an intermediate body that is most suited to monitor and ensure that the national standards in service delivery and policies are adhered to. The RAS is also well placed to receive all LA reports in a - given region - for review, compilation and consolidation before such reports are forwarded to higher levels of the government. There is also no overall body that pulls together information on financial performance, and on service delivery and linkages with outcomes. The issue is whether sector ministries and MOF currently get what they need to inform policy decisions, because of lack of consolidated reporting. (iv) Incentives for accurate and timely reporting are limited. At present, only the release of funds under the health basket fund is contingent of the submission of adequate reports on health sector performance and the use funds of the previous release. The question is whether there is scope for applying a similar approach for the release of block grants. However, since accounting and reporting at the local level is severely constrained by limited human capacities, any attempt to enforce reporting standards by linking it to the release of funds would require measures to both strengthen capacities of local authorities in the area of accounting and to simplify and streamline reporting formats to facilitate reporting and quality control of such reports. 65 3.27 While emphasis should be on the strengthening and better integration of government internal monitoring and reporting system, external validation has also an important role to play. The external evaluation as part of the PER process is an important component of this validation process and occasional external expenditure tracking studies or service delivery surveys will be beneficial to establishing and maintaining confidence in government's own systems. 3.6. THE ROLE OF THE PER IN TANZANIA'S POVERTY MONITORING SYSTEM 3.28 Government has developed a PRSP monitoring master plan, which provides a comprehensive system for monitoring the various dimensions of poverty reduction and economic growth as well as the implementation of the various elements of the poverty reduction strategy. This system comprises three interrelated tiers. The poverty monitoring system focuses on the measurement of impact, outcomes, and proxy indicators. Sector programs such as those for education and health focus on the tracking of output indicators with adequate monitoring mechanisms in place. The PER/MTEF processes focus on inputs with the emphasis on budget allocation and budget execution. Analytical work to be carried out as part of the poverty monitoring process is expected to establish the linkages between monitoring of impact, outcomes, proxy indicators, outputs, and inputs. 3.29 In addition to the overall PRSP monitoring process, a number of specific monitoring processes are also of importance. These include PRGF reviews as an important guide for the assessment of macroeconomic performance; the PER to monitor progress in strengthening financial accountability and expenditure tracking systems; the * Tanzania Assistance Strategy and the Independent Monitoring Group (IMG) monitoring progress in developing a constructive-and effective donor support mechanisms; quarterly reporting on the implementation of the anticorruption strategy; and investor climate surveys and investor roadmaps to monitor progress in improving the business environment as the basis for sustained economic growth. 3.30 Table 3.1 provides an overview of the monitoring system which covers outcome, output, process and input indicators. In the context of the PRSC and PRSP outcome and input indicators are those that provide direct information on the status of poverty in all its dimensions. Monitoring of these indicators is based on a broad range of surveys and administrative data systems identified in the poverty reduction master plan. The Poverty and Human Development Report (PHDR) is expected to be produced annually and consolidate poverty measures from various sources and provide the analysis of poverty trends and underlying factors. Sector development reviews will be the key monitoring mechanisms for outputs at the sectoral level while the PRGF reviews and the quarterly macro-financial developments report will be the key sources for the assessment of the macro-economic situation. Progress in public expenditure management covering both improvements in systems as well as allocative and operational efficiency enhancement in use of inputs will be tracked by the annual Public Expenditure Reviews. An additional dimension of the monitoring framework covers the government-donor relationship to 66 Table 3-1: Monitoring Mechanisms covering Outcomes, Outputs, Processes, and Inputs Outcomes Outputs Process| Inuts OVERALL PRSP Progress Report POVERTY REDUCTION (ALL DIMENSIONS) POVERTY AND HUMAN DEVELOPMENT REPORT Economic Macro-Fiscal Investment Performance Developments Climate Reports Assessment | PRGF Review Public Sector Client Survey Expenditure Performance Reports Tracking Reparts Performance Anti-corruption Budget Budget Reports Reports Execution Reports PUBLIC EXPENDITURE REVIEW (COVERING PUBLIC EXPENDITURE PERFORMANCE AND MANAGEMENT AT THE CENTRAL AND LOCAL LEVEL) Administrative Monitoring Systems for Other Sectors and Local Authorities SDP REVIEWS FOR EDUCATION, HEALTH, AGRICULTURE, AND ROADS Government IMG/TAS Donor Implementation Partnership Report monitor progress on objectives such as enhanced local ownership of the development process, participation by all stakeholders, and efficiency of aid delivery mechanisms. It is expected that an annual report of the findings of the independent monitoring group 67 (IMG)'3 on the implementation of the Tanzania Assistance Strategy (TAS) which lays out the agreed principles for a government donor relationship in line with the CDF principles. 3.7 RECOMMENDATIONS 3.31 The govermment has put in place a fairly good reporting system. Various reports are prepared that act as a feedback and control mechanism on the flow and use of public resources. However, there is lack of coherence of reporting systems that are in place and it is unclear who is supposed to put together these reports in a coherent and a user friendly way. Two types of reports exist: the first type relates to financial reports which are generated using the IFMS at the central government level and LA reports which are submitted to the councils, RAS and PORALG; while the second type relates to service delivery-like reports (e.g., on health basket) which are submitted to the sector ministries and to donors. GOT wishes to internalize expenditure tracking for finaricial and service delivery. It is recommended that GOT should review the existing tracking/reporting and monitoring systems to identify information gaps and recommend (i) how systems can be streamlined, refocused and integrated (ii) how information flows can be improved; and (iii) the appropriate roles and responsibilities of the key institutions in this process. This could be assessed as part of this year's PER expenditure tracking exercise. The exercise could also check whether the existing sectoral reports really reflect the assessment of service delivery/value for money; and whether harmonization of central government and local government fiscal years would help make rationalization of expenditure reporting systems more tractable and coherent. As of now reporting and tracking of expenditure has been unharmonious and complicated by the existence of the two different reporting years at local and central government level. 3.32 The government should also review how to pull together information on financial and service delivery and links to the poverty monitoring process. This could be examined under the expenditure tracking study and recommendations made on how well this should be institutionalized. 3.33 As a transitional mechanism, the LA financial reports could be consolidated at the regional level by the RAS before being forwarded to the PORALG for overall country- wide consolidation. The consolidated financial report could then be forwarded to the Treasury on a quarterly basis. The LA service delivery reports could also be consolidated by the envisaged PORALG's inspectorate unit before they are forwarded to CSD for overall consolidation with MDA service delivery reports. The two reports, financial and service delivery, could then be forwarded to VPO where they can be used as an input in poverty monitoring - linking resources and their impact on poverty reduction. 13 The independent monitoring group has been established in the mid-nineties to monitor the relationship between government and donors building on the findings of the Helleiner report. 68 3.34 In view of (i) and (ii) above, GOT should help put in place PORALG's inspectorate unit to cater for putting together/integrating/consolidating LA information from various regions. 3.35 There is a need for GOT to clarify the role of various levels of government (RAS, PORALG, sector ministries and Treasury) in as far as enforcing of standards, reporting and monitoring is concerned. In particular, GOT should clarify the role of the Regional Secretariat vis-a-vis the sector ministries in monitoring and evaluation and in enforcing national standards, and how sector ministries can avail resources and make use of RAS as an anchor for oversight function at LA and other lower levels of government. 3.36 Building capacity of financial departments of LAs is essential. Currently the shortage of skills and training constrains reporting, auditing and implementation of IFMS. 3.37 The PRSP table (Annex 3) on: "Disaggregation of Recurrent Poverty Reducing Expenditure", should act as a framework for informing PRSP, PRBS, etc. It should be updated and reviewed as part of the PRS progress report and MTEF. 69 4. FISCAL RISKS 4.1 MOTIVATION 4.1 It is generally recognized that the govemment of Tanzania has sizeable fiscal obligations that are unaccounted for, as is the case in most developing countries. Indeed, one of the key reasons for the continued use of the cash budget system is GOT's concem about fiscal risks. In consideration of such potential risks, deficit targets are typically set at a much more stringent level than would be required on the basis of the fiscal sustainability analysis that has been carried out in previous years. Similarly, the relatively large and growing special expenditure/contingency items in the annual budget confirms that the government is concerned about fiscal risks. The apparent reluctance of donors to move towards budget support also suggests that they are concemed with fiscal risks. Analysis of fiscal performance carried out under the previous PERs did not consider the contingent liabilities of government partly because relevant information is yet to be compiled and validated. It is only now that the government has formed a Task Force to compile dispersed information on domestic contingent liabilities. Building on this ongoing work, a general assessment of the sources of fiscal risks in Tanzania's public sector was undertaken, covering the central government, parastatals and local authorities. The major focus was on identifying specific risks. In addition to identifying the sources of risks, where data was readily available a preliminary quantification of these risks was undertaken, and the efficacy of the existing mechanism to contain risks assessed. The chapter also presents an agenda for further work on identifying and managing fiscal risk in Tanzania. 4.2 SOURCES OF FISCAL RISKS IN TANZANIA'S PUBLIC SECTOR 4.2 Work that has been done elsewhere suggest that there are many sources of fiscal risks (see Table 4.1). In the case of Tanzania, an examination of the provisions for special expenditures and contingencies in the government budget (Annex 4) indicates three main categories of special expenditure. One category which may be labeled as - 'pure contingency'- typically relates to provisions for unforeseen events and contingent liabilities of government, including public enterprise debts and retrenchment costs as well as negotiated compensation and programme aid to the Revolutionary Government of Zanzibar. The second category -'quasi-contingencies'- consists of provisions for expenditures where apparently there is a time lag between policy decisions and financing. Specifically, such provisions are made where the speed of making some policy decisions had outpaced that at which financing implications of such a decision are worked out. Examples include, provision for salary adjustment, the decision to set aside Tsh. 11 billion following the decision to abolish primary school fees, the payment of housing allowance for higher learning institutions, as well as provisions for HIV/AIDS and anti- 70 Table 4-1: The Fiscal Risks Matrix LIABILITIES DIRECT CONTINGENT (obligation in any event) (obligation if a particular event occurs) Explicit Foreign and domestic sovereign State guarantees for non-sovereign (Government liability borrowing (loans contracted and borrowing and obligations issues to sub- as recognized by a law securities issued by central national governments and public and or contract) government) private sector entities (development banks) Budgetary expenditures Umbrella state guarantees for various Budgetary expenditure legally types of loans (mortgage loans, student binding in the long term (civil loans, agriculture loans, small business servants' salaries and pensions) loans) Trade and exchange rate guarantees on private investments State insurance schemes (deposit insurance, income from private pension funds, crop insurance, flood insurance, war risk insurance) Implicit Future public pensions (as Defaults of sub-national government or (a moral obligation of opposed to civil service public or private entities on non government that pensions), if not required by law guaranteed debt and other obligations reflects public and interest-group Social security schemes, if not Cleanup liabilities of entities being pressures) required by law privatized Future health care financing, if Banking failure (support beyond state not required by law insurance) Future recurrent costs of public Failure of non guaranteed pension fund, investments employment fund, or social security fund (protection of small investors) Default of central bank on its obligations (foreign exchange contracts, currency defense, balance of payments stability) Bailouts following a reversal in private capital flows Environmental recovery, disaster relief, military financing Source: Polackova H. (1999), Contingent Government Liabilities: A Hidden Fiscal Risk. Finance & Development Vol. 36 No. 1, March corruption campaigns. These decisions were made prior to the determnination of how much exactly is to be allocated to which district or higher learning institution, and therefore lumped under special expenditure, but subsequently are re-allocated from Vote 71 50 to respective votes once strategic sector plans are developed and relative financing magnitudes are determined. The third category of provisions for special expenditures relate to payment of arrears partly due to weaknesses in commitment control, especially those related to procurement of goods and services and utility services, as identified under the FY01 PER. 4.3 The latter two main sources of fiscal risks, appearing under special expenditure, namely: (i) public enterprise debts and retrenchment costs associated with privatization, and (ii) extra-budgetary commitments related to procurement: 4.4 Parastatal debts: Since the initiation of the parastatal sector reform program a total of 211 companies have been divested to sale agreement or memorandum of understanding stage. The total number of parastatals that remain for divestiture as at June 30, 2001 is 283. One of the key problems observed is that most of the public enterprises that have been divested and those still in the pipeline for restructuring, have substantial liabilities which have to be taken over by the government in full or part. The parastatal liabilities that have fiscal implications for the government are estimated by the Task Force at Tsh. 280.47 billion and include suppliers' credits (23.8%), trade credits (20.8%), staff dues (19.9%), commercial loans (11.9%), unpaid utility bills (3.5%), and others (20.1%). It is important to note that this estimate by the Task Force is almost 6.4 times higher than the Tsh. 28.3 billion parastatal debt figure recorded in the domestic debt report produced by the Office of the Accountant General. Even when expected privatization proceeds (Tsh. 71.33 billion) are netted out of the estimated total parastatal debts, the balance remains significant. Furthermore, 92.6% of these liabilities relate to the public enterprises that are still in the pipeline for privatization and those that are not yet earmarked for privatization. 4.5 In addition to parastatal debt, information compiled from audited reports does reveal about Tsh. 172 billion of non-securitized liabilities which are either understated in the ACGEN monthly debt report or not recorded at all. The main non-securitized liabilities include suppliers' arrears (49.4%) and compensation claims against nationalized assets (34.0%). Others include exchange losses accumulated in the BOT revaluation account and domestic guarantees. The timing, terms and form of payment of these debts is unclear but are very likely to have fiscal implications for the government in not a too distant future. 4.6 Retrenchment costs: Government policy regarding the financing of retrenchment is that if a public enterprise is capable of paying terminal benefits to its workers, it is allowed to pay a statutory payment. Golden hand shakes (ex-gratia payment) may be offered by a parastatal with approval by the Treasury Registrar, provided the parastatal can afford to pay such payments. Where the public enterprise cannot afford to pay for retrenchment of its workforce, the government has committed to honor only the statutory payment. Unfortunately, most of the parastatals that are in line for privatization are in poor financial position to be able to meet retrenchment costs from their own sources even including privatization proceeds. This implies that the burden will have to be taken up by the government. Indicative costs estimates done by PSRC for meeting the costs of 72 retrenching all the 21,556. employees of five key parastatals (TRC, THA, ATC, DAWASA and TANESO) amount to Tsh. 19.9 billion, covering statutory payments alone. Even under an alternative scenario where only 8,021 workers from these parastatals are to be hived-off, the estimated statutory payments would still amount to Tsh. 17.3 billion. Although no similar attempt has been made to estimate the retrenchment costs for the remainder of the public enterprises earmarked for restructuring, the implied retrenchment costs are bound to be significant. 4.7 Commitment control and monitoring: Closely related to the issue of fiscal risks is the capacity of the Ministry of Finance to monitor comprehensively expenditure commitments made by spending units. The FY01 PER had identified a number of expenditure items, such as expenditures on utilities and expenditures for "automatic catering" which were not fully integrated in the commitment control and monitoring system and which subsequently were frequently the cause for the accumulation of arrears. This report revisis this issue by assessing the comprehensiveness of the commitment control and tracking system, issues arising from the move from monthly commitment limits to quarterly commitment limits, and progress in covering expenditures that cannot be easily managed by commitment control as first charge expenditures. 4.8 Government has achieved significant improvements in the area of commitment control and monitoring of expenditure through the IFMS and application of provisions in the new Public Finance and Procurement Acts. Commitment made through the IFMS cannot exceed the funds budgeted for and released monthly or quarterly and for the purpose. On a daily basis, MOF is in a position to note any deviations/ contraventions to the norm. The IFMS also enables the timely generation of reports for detailed scrutiny and analysis. As part of the government effort to improve on transparency, MOF has instituted quarterly disclosure of allocations to all spending units in the news papers and directed these units to also display the same on their notice boards for public scrutiny. 4.9 However, it is worth noting that audited data on the stock of arrears (table 4.1) confirm the finding of the PER FY01 that the value of claims cleared for payment are on average 31% smaller than the value of claims finally submitted for audit, reflecting inflated pricing of supplies and fictitious claims. Rejected claims (claims submitted and verified less claims cleared for payment) ranged from 5.2% for personal claims to 10.7% for goods and services and as much as 30.4% for utilities. 73 Table 4-2: Summary of Consolidated Arrears Cleared for Payment by Category July 1998 - December 2000 (% Share Vote Vote Holder Goods & Utilities Personal Total Services Emoluments 28 Police Force 9.3 7.8 21.9 11.8 29 Prison Services 12.8 6.9 16.1 11.6 38 Defense 24.2 50.7 24.5 33.1 39 National Service 6.8 5.1 1.9 5.0 Sub-total 53.1 70.5 64.4 61.5 46 Education & Culture 3.9 0.5 5.8 3.2 52 Health 1.1 0.0 1.9 0.9 All Other 41.9 29.0 27.9 34.4 ministries/depts./region 4.10 Table 4.2 also confirms the findings of the PER FY01 that the Police Force, Prisons Services, Defense and National Service alone accounted for about 2/3 of consolidated arrears. That is, 53.1% of debts for goods and services; 70.5% of debts for utilities and 64.4% of other debts. Debts on goods and services combined with utilities accounted for about 76.3% of the total debt cleared for payment. Thus, it is important for keep constant watch on possible occurrence of non-adherence to the centralized commitment system particularly for votes 28, 29, 38 and 39. Bold and prompt imposition of sanctions on Accounting Officers for extra-budgetary commitments is critical for inculcating fiscal prudence and containing future build-up of arrears, particularly for utilities. The measures taken so far to enforce the centralized commitment system through IFMS, including restricting the power to reallocate funds to MOF only are important elements to ensure overall fiscal discipline. To close remaining loopholes in the commitment control system, expenditure items that are difficult to contain, such as food for prisoners and defense units should be made first claim charges. 4.3 CONTAINMENT OF FISCAL RISKS AND AGENDA FOR FURTHER WORK 4.11 On account of the above, it was confirmed that indeed Tanzania has significant fiscal obligations that are largely not accounted and properly budgeted for. The first important step therefore, is the need to undertake a thorough exercise to identify the principal sources of fiscal risk in order to come up with practical suggestions to contain them. The following actions and measures are proposed: 4.12 Government has embarked on quantifying the magnitude of the main categories of contingent liabilities which are associated with parastatal restructuring (public enterprise debts and retrenchment costs) and produced a preliminary set of estimates. Given the complexity of the problem and incompleteness of data, it is recommended that government consider engaging a team of experts to do more detailed audit work to quantify and validate the true position of public enterprise debts and come up with options for their settlement. 74 4.13 Second, on quasi-contingency items and provision for settlement of arrears, government I encouraged to work further to improve planning and budgeting, and especially by minimizing the lag between decision making and financing. It is suggested that government establishes a medium term target to reduce the quasi-contingency items currently lumped under special expenditure. These should instead be reassigned to the relevant votes and sub-votes and thereby remain with modest amounts under special expenditure for the truly unforeseen events, whose expenditure cannot be assigned ex-ante. For example, electricity bills, HIV/AIDS and anti-corruption campaigns, VAT refunds as well as salary adjustments should gradually be transferred to the respective ministry budgets. In addition, it will be prudent for the government to consider relegating public enterprises confirmed to have negative net worth to liquidation and writing off all uncollectible taxes of parastatals. 4.14 Third, in terms of the agenda for further work in this area, there are other possible sources of fiscal risks in Tanzania, in addition to fiscal risks emanating from contingent government liabilities. These include macro-economic risks arising from fluctuations in economic growth and the fiscal revenue base, and risks from the activities of agencies and local authorities that are not directly under the commitment control system of the central government: For example, as noted in the PER FY01 report, the 1999 CAG report indicates the total stock of councils' creditors for 1999 was about Tsh. 5.5 billion. Over 90% of these payment arrears, which constitute part of the stock of contingent liabilities, are in form of suppliers credit while the remainder are statutory deductions not paid over. It is recommended that the proposed work to quantify and validate parastatal debts be extended to cover other sources of fiscal risks, including expected commitments and timing of privatization; Zanzibar debts; and potential contingent liabilities of local authorities including possible insolvency of the Local Authorities provident Fund (LAPF). It is further recommended that the proposed work to quantify and validate parastatal debts be extended to cover other sources of fiscal risks, including expected commitments and timing of privatization; and potential contingent liabilities of local authorities. The proposed audit should also cover commitments made on behalf of the government / agreements signed by accounting officers outside the purview of MOF; the actuarial position of LGPF, PPF and NIC; debt swaps and government guaranteed debts. Meanwhile, MOF needs to strengthen its monitoring and control function over LA liabilities by demanding and analyzing expenditure and revenue reports produced by LAs quarterly. Furthermnore, government is urged to finalize the amendments to the Loans, Guarantees and Grants Act of 1974 to check government guaranteed debts. H.E. President Benjamin William Mkapa has reiterated recently that only the Minister for Finance is mandated to contract debts on behalf of the government of Tanzania. 75 5. EXTERNAL DEBT MANAGEMENT AND COMMITMENT CONTROL 5.1 INTRODUCTION 5.1 Following the establishment of macroeconomic stability and the liberalization of prices during the first phase of its reforms, the government of Tanzania has now focused public policy reform explicitly to promote growth - led by the private sector - and poverty reduction.14 Debt relief under the enhanced HIPC initiative supports Tanzania's reform program by freeing resources for the implementation of poverty reduction programs. In order to ensure that government borrowing is consistent with poverty reduction efforts it is important to closely monitor Tanzania's external debt. In addition, the system of external debt management and commitment control is crucial to ensure the integrity of the public financial systems. Problems in external debt management can not only derail the poverty reduction targets that the government intends to achieve, it can also undermine sustainability of the stable macro-economic environment. 5.2 The report takes stock of issues in (i) debt contracting; (ii) debt servicing and monitoring by the government of URT; and, (iii) provides recommendations to the government for strengthening the system to avoid occurrence of problems due to weaknesses in actual operation of (i) and (ii), including the operation of multiple debt contracting authorities. The first step in this exercise was to look at the institutional arrangement as embodied in the 1974 Act and 1999 Debt Strategy, and compare it with the accepted best practice in this field, in order to trace 6ut the extent of deviation in Tanzania and the reasons thereof. The second step was to link the legal and operational strategy as codified in Tanzania to the current operation of debt management through roles of the various departments in the MOF and the BOT in (i) and (ii); and, disclosures in the fiscal accounts (through operation of the CDS and IFMS). 5.3 The consideration of external debt commitment and management was deemed timely in PER FY02 for providing feedback in three contexts: (i) Tanzania is expected to be able to reach a sustainable debt level and have a basis for a robust exit from recourse to further debt relief measures, having completed all conditions prior and obtained debt relief from multilateral institutions under the enhanced HIPC initiative, amounting to US$ 2,026 million in net present value terms.'5 The relief under this program is being delivered through debt service reduction such that Tanzania's NPV of debt to 14 The PRSP (2000) and its recently completed first annual update (October 2001) embody steps towards implementation of poverty alleviation objectives. It builds on the NPES (1997); and, ongoing processes adopted in the annual PERs, the MTEF consultations, and the CG meetings. 15 IDA/IMF decision point was reached in April 2000; and the completion point was reached in November 2001 76 exports ratio is expected to remain below the sustainability threshold of 150 percent during the period 2000-20. This debt relief also provides scope to the government to prioritize usage of its resources for targeting poverty reduction without defaulting on its external obligations. However, this development path is anchored on prudent borrowing strategies and effective debt management to protect long term debt sustainability, in addition to good macroeconomic perfornance and reasonable growth projections. Annex 5 gives the summary findings on debt management presented by the IDA/IMF team during the HIPC Completion Point discussion with their Board of Directors in November 2001, based on the questionnaire filled by GOT. The feedback from the PER FY02 is expected to carry this process a step further, in helping the government assess the alignment of the roles of its various institutions in debt management with best practices. (ii) President Mkapa has recently enjoined the Minister of Finance to take stock and review all sector and parastatal loans initiated and existing by February 2002 as the first step for the development and implementation of an appropriate debt management strategy. He has reiterated the legal standing of the Minister of Finance (or his authorized agent) as the sole debt contracting agent for the government. (iii) In adapting and adopting laws to streamline its public financial commitments Tanzania reached the stage of implementing the Public Finance Act 2001, and the Public Procurement Act 2001. In September 1997 the government of Tanzania prepared a National Debt Strategy; and, in March 1999 revised and consolidated it (NDS: Part 1 - External Debt; March 1999) as guidance for proper debt management and its ultimate reduction to a manageable level. However, the legal foundation for external debt contracting in Tanzania remains the Government Loans, Guarantees and Grants Act, No. 30, of 1974. The MOF is scheduled to propose revisions to this Act soon. 5.2 SUMMARY OF FINDINGS 5.4 The major areas of potential slippage from established procedures for debt contracting and management are: (i) Co-ordination among the various players in the government system, with very different historical roles and human capacities in debt administration; (ii) MOUs entered into by sector ministries with suppliers and financiers for contracting debt close to the threshold of concessionality; (iii) Impact of the liabilities (and significant earnings, if any) of independent companies in which sector ministries own stock; (iv) Presentation of projects, with associated loan financing arrangements, to the Ministry of Finance at an advanced stage of the project formulation. This not only undermines credible debt management, but also precludes the possibility of MOF mobilizing alternative funding on grant/highly concessional termns; and, (v) Weak capacity for analyses of long term debt sustainability in PAD. 5.5 Recommendations to improve debt management cover five broad areas: 77 1. Strengthening the National Debt Strategy 1999 in the area of risk identification and provisioning for external debt 5.6 Table 5.4 compares the best practice guidelines of public debt management (as issued by the Bretton Woods institutions; March 2001) with the National Debt Strategy (Part 1 - External Debt, March 1999). It recommends that the strategy should be supplemented with specific actions to be undertaken by the government in regular periodic assessment of the structure of the debt portfolio and risks; and, the identification of mechanisms to monitor, evaluate, and prevent build-up of liquidity exposures, especially in foreign currency. This is essential for the Tanzanian external debt to remain sustainable and for its structure not to induce or propagate economic crises. The issues for clear specification are the criteria and governance arrangements for contingent liabilities; identification of 'off-budget' claims arising through contingency/guarantee and unsound business practices; and removal of debt misreporting arising from inadequate co-ordination and unclear responsibilities. The broader strategy issues for elaboration are about provisioning for debt of state-owned or majority state shareholding in enterprises; and, consistency with development of an efficient government security market. 2. Risks of Various Types of Debt 5.7 Supplier Credit Arrangements: It is recommended that special scrutiny and oversight be exercised on the external debt obligations, especially from commercial sources, that arise from supplier credit arrangements for goods and services provided to sector ministries. The commitment control system should ensure that such credits are not secured external to established procedures. 5.8 Short term debt and non-concessional finance: It is recommended that special scrutiny and oversight be exercised on the external debt obligations that arise from short term financing for the sectors ministries. Any credit from commercial sources, such as private banks, should not be accepted for ranges of grant element less than 35 percent (as has been the GOT practice in recent years) and receive maximum scrutiny in the range of 35-50 percent grant element by the DMC. As per the NDS, it should be demonstrated by the sponsoring sector ministry, and confirmed by EFD, that grant financing is not available. 5.9 (iii) Independent Companies: There is concern that for independent companies in which the government is a beneficiary of net income, government could be enjoined in any proceedings of default. Hence, it would be useful for the GOT to seek a legal opinion on possible exposure to risk in international arbitration. As an example of the debt exposure and fiscal risks involved, the case of Meremeta Ltd is presented. It is recommended that GOT being a shareholder of any such company, .the scrutiny of its debt contracting should be on similar rigorous grounds as all other government or public guaranteed debts. The Minister of Finance should sign off on such company debt only when satisfied that all tests have been met. More importantly, as such a company may prospectively be a large source of extra budgetary funding, such funding should be brought into the overall expenditure allocation to preserve the integrity of the MTEF. In cases where similar self-financing is practiced, the retention scheme is subject to the budget ceilings consistent with the overall priority allocations. 78 5.10 Legal Provisions: It is recommended that the laws of establishment of public executive agencies, parastatal companies and sector ministries that grant these entities special exemptions in external debt contracting - both for finance and for acquisition of goods and services - be scrutinized and revised for consistency with the Loans, Guarantees and Grants Act, No. 30 of 1974 (and revisions to it being prepared by the MOF) and the National Debt Strategy 1999 (Part 1). Over the medium term, such exemptions should be minimized and abolished. The public - especially the legislature - should be educated and sensitized on the legal roles and requirements; that prescribed sanctions be transparent; and stringency in enforcement of sanctions be instituted to deter future occurrences. 3. Institutional Roles 5.11 It is recommended that the roles of various agencies as envisaged in the legal structure be adhered to more strictly than has been in the recent past. Thus, the Minister of Finance is the sole signatory in all of URT's public debt contraction. For his support, the following MOF entities have the responsibility for: * Policy Analysis Department (PAD) - strategic manager of debt for sustainability and guidance to other entities; * External Finance Department (EFD) - negotiator in specific debt contracting; and * Accountant General (ACGEN) - custodian in charge of maintaining all legal agreements and debt obligation records; issuer of payment instructions against all creditor claims. In addition, * Bank of Tanzania (BOT) - banker to the government for executing payment instructions from the ACGEN; and, externalizing the foreign component of such payments; and, * Debt Co-ordination Committee/Debt Management Committee (DCC/DMC) provides oversight on all proposed loans and government guarantees.16 5.12 In order to fulfill the designated institutional roles most efficiently, it is recommended that the following steps be considered: (i) PAD: It is recommended that steps be taken to re-institute the key role of PAD in exercising oversight over all sovereign debt, assessing the impact of specific debt obligations on overall public debt sustainability, and providing overall guidance. Analysis of the concessionality of all proposed loan financing must be accompanied by evidence that grant /highly concessional loan financing has been sought prior to the consideration of other loan terms. It is recommended that an annual public debt statement, for both external and domestic debt, be published with the budget, with responsibility of location of this located with the PAD in collaboration with the ACGEN. In the medium term it is recommended that the capacity be developed in the PAD to 16 The DCC, with members from GOT entities key in debt management, is given technical support on matching the loans with GOT priorities, the terms of loans, and the timing of repayments (bunching up) by their technical team, DMC. PAD functions as the secretariat for DCC/DMC. 79 conduct the debt sustainability analysis for Tanzania.'7 The recent debt contraction and management through the cross-departrnental DMC and the DCC, for which PAD functions as a secretariat, should not divert attention from this need to develop capacity. (ii) EFD: For sake of complete coverage of the external debt portfolio, it is recommended that EFD staff should be assigned the specific negotiating responsibility for non-concessional bilateral and for commercial loans.. This would prevent occurrence of past slippages due to lack of coverage. (iii) EFD: It is recommended that a sector/line ministry-specific oversight be developed in the EFD, especially in the priority sectors, with the use of MOF staff already assigned to collaborate with line ministries. In the medium term, it is recommended that consideration is given to EFD moving away from the current donor-specific assignment of staff in debt contraction and supervision to sector-specific assignment. This will ensure that MOF is alerted to project formulation (and associated loan financing) at an early stage. It will also ensure that awareness is drawn to the diverse range of donor procedures within a given sector; and, hence accelerate the pace of harmonization of donor practices at the sector level. (iv) ACGEN: It will be important to synchronize the databases on external loans and grants located in BOT and ACGEN's office; and to actively keep the ACGEN informed of all negotiations. In the medium term, it is recommended that the government aim to remove parallel systems of debt reporting and locate the information solely with the ACGEN - its custodian of financial records and payment authorizer - without any loss in efficiency in external transactions. A time-bound plan should be formulated to transfer the management of the database from BOT to ACGEN, allowing for an appropriate time span for the systems to run in parallel so that verification may be assured. (v) DCC/DMC: It is recommended that DCC/DMC meet at regular intervals, to execute their respective responsibilities to ensure that (i) a proposed loan adheres to GOT priorities as stated in the PRSP; (ii) it aims to attain its objectives efficiently; and, (iii) its financing terms are the best possible. (vi) MOF: To successfully implement measures for improved debt management, appropriate measures to strengthen staff capacity and skill development will be necessary. Moreover, for retention of skilled staff and good governance it is recommended that staff compensation in PAD, EFD and ACGEN's office be considered under programs already being executed by the GOT (e.g. the Selective Accelerated Salary Enhancement Scheme, SASE). (vii) BOT: In addition to the BOT's role as the government's banker, it also carries responsibility towards the overall supervision of Tanzania's Balance of Payments. To fulfill this role efficiently BOT needs accurate and timely information on all external public obligations 17 This should address issues of debt structure; sustainability; and risks in debt management - e.g. exchange rate, rollover, fiscal implication of debt; interest rate risks - in the context of the prevailing macroeconomic environment, which are inadequately highlighted in DSAs by BOT. 80 from the MOF (ACGEN). In addition, BOT is the custodian of all information on private loans that have an external obligation component. 5.3 OVERVIEW OF TANZANIA'S DEBT 5.13 In 2000 the bulk, 93.7 per cent, of Tanzania's external debt was owed by the government. The remaining 6.3 per cent of the external debt was owed by public corporations (2.5 per cent) - with no government guarantees - and the private sector (3.8 per cent). The Tanzanian debt situation compared to its neighbors and the average in Sub-Saharan Africa, Latin America and Asia in 2000 is presented in Table 5.1. Tanzania's debt to GDP ratio is slightly above the average for Sub-Saharan Africa, but significantly above that of Kenya and Uganda. However, as a result of Tanzania's larger exports of goods and non-factor services, it's debt to export ratio is smaller than that of Uganda and Kenya. Nonetheless, Tanzania has to devote a significant share of its public resources to debt service payments, as outlined in the section on the review of fiscal performance. Table 5-1: Selected Debt Indicators for Tanzania, Uganda, Kenya, Sub-Saharan Africa, Latin America, andAsia, 2000 Tanzania Uganda Kenya SSA Latin Asia America otal Debt (billion US $) 7.5 4.1 6.6 246 704 870 ebt/GDP (%) 90 63.4 61.6 87 135 3 ebt per Capita (US $) 243 185 194 382 1,408 1 Dibt Service/Exports of 19 23 26.7 15 35.5 18 Goods and NFS (%) Source: Global Development Finance, World Bank, 2001. 5.14 Based on its track record in macroeconomic stabilization, market-friendly reforms, and demonstrated commitment to poverty reduction since 1995, Tanzania was assessed by the World Bank (IDA) and IMF to be eligible for debt relief under the enhanced HIPC initiative at the decision point in April 2000. Such relief is envisaged to promote the government's growth and poverty reduction objectives, without the debt servicing to external creditors siphoning scare resources away from priority social sector expenditures of the government or defaulting on its obligations to these creditors. With the completion of the benchmarks for the interim period, its eligibility under the enhanced HIPC was affirmed in November 2001 when Tanzania reached the completion point. As noted earlier, with this relief from the multilateral donors, Tanzania's NPV of debt to exports ratio is expected to remain sustainable (below a threshold of 150 percent) over 2000-20. Many of the bilateral donors have also agreed to provide additional debt relief that goes beyond what is provided under the HIPC initiative to give Tanzania greater flexibility in targeting its resources. 5.15 Table 5.2 shows the structure of debt as of June 2001 owed to the various types of creditors - multilateral institutions; bilateral donors; commercial sources. 81 Table 5-2: Tanzanian External Debt Outstanding, end-June 2001 Share (%) Nominal debt NPV of debt Multilateral creditors 56.8 45.7 o/w World Bank 40.0 31.8 /w IMF 7.9 .5 Paris Club Bilateral 25.0 28.4 Other Official Bilateral 14.9 21.1 /W China 2.6 3.0 DW Kuwait D.8 1.1 Commercial .2 4.7 rotal (In mil US$) 6185.8 4155.40 5.4 PERCEIVED RISKS OF VARIOUS DEBT CATEGORIES 5.16 Under its PRGF/ESAF programs with the IMF, the government has agreed to adhere to borrow only on concessional terms. The international threshold for this benchmark is a minimum of 35 percent grant element; and all of Tanzania's recent borrowing has fulfilled this cut-off. The discipline and the controls around contracting debt have been standardized to a considerable extent. 5.17 Multilateral Financial Support: Loans, especially from multilateral institutions, remain a very important (56.8 percent of nominal debt as of June 30, 2001) source of foreign finance in Tanzania. Procedures for debt contracting are most standardized in this category, both due both to government oversight and donor requirements; and the DMC/DCC have the least time cost of preparation in pre-appraisal.18 In new borrowing from this donor source the government should be guided by the government's poverty objectives reflected in the PRSP. In debt management, the loans in this area have a high concessionality. Note also that in this category enhanced HIPC debt relief has been recently obtainred for the debt servicing over period 2000-20 with different donors provisioning through various mechanisms.19 5.18 Bilateral Financial Support: Most bilateral donor assistance has moved to grants - either budget or project support - in Tanzania which, though indicating external dependence, has no financing ramifications.20 Of the bilateral loans the debt contracting procedures and terms of concession for the 13 Paris Club members are standardized and match on average those of the multilateral donors. New borrowing should be guided by the government's poverty objectives 18 For IDA, IMF, AfDB, NDF, IFAD; for BADEA/OPEC terms vary depending on GOT negotiating capacity 19 IDA/IBRD, IMF, AfDB, EC/EU, IFAD, NDF, OPEC, EADB, BADEA 20 However, counter contributions may be important. 82 reflected in the PRSP; and, the lending by donors guided by those PRSP conditions adopted under the PRBS to which many bilateral donors are signatories. In the management of this debt (25 percent of nominal debt as of June 30, 2001) the Paris Club members agreed to debt relief through debt rescheduling on Cologne terms in April 2000 and a stock-of-debt operation on Cologne terms provided at the completion point in the context of the HIPC initiative. 5.19 The remaining official bilateral debt is from non-Paris Club donors (15 percent of nominal debt as of June 30, 2001). The major creditors, with more than USD 100 million nominal debt each, are Algeria, Czech Republic, China, Iran, Iraq, Libya, and Yugoslavia.2' No agreements for debt relief have been reached with this group, except China and Kuwait (totaling 3.5 percent of nominal debt in June 31 2001), for treatment at Paris Club terms. Though many of the loans are old loans from the financing available to Tanzania during the socialist era, it is also here that the procedures for new debt contracting from bilateral donors and financial management of debt are least standardized. Consequently, the terms of loans from these sources should be subject to the particular scrutiny. Also, the demand notes for repayment from the creditor should go to the Accountant General for payment from the consolidated fund, in order to avoid the risk of circumventing the process by involving the sector ministries or the BoT directly and having debt service payments outside the debt servicing plans for that year. In recent years the govermment has adopted a general stand of not servicing the loans from these sources - unless they agree on Paris Club terms - sending the appropriate market signals for incurrence of new liabilities. Table 5.3 presents the structure of these loans in the period 1990-2002. 5.20 Commercial Credit, mainly provided for Supplies/ Investment: These can originate from any of Tanzania's trade partners and include financing from private banks. These loans constituted 3.2 percent of the nominal debt in June 2001 and pose the maximum risk for the debt sustainability of the government of Tanzania despite their small overall share. Generally, the government has maintained the basic conscessionality criteria of 35 percent grant element in incurring these loans, but by narrow margins. These loans have been used for financing of investment in large capital equipment; e.g. the air traffic control system. Moreover, the due processes of debt contracting is largely by-passed in the initial negotiation and appraisal of these loans. Often the demand originates from MOUs entered into by sector ministries with suppliers and financiers of investment in their sector, with or without financing from foreign banks. Pressure is created on the central public oversight such that role of the MOF is reduced to rubber- stamping the legal agreement without adequate participation in the commitment/negotiation process, and moving away from being an anchor in external debt strategy. The impact of the liabilities on the MTEF (and significant earnings, if any) of such debt contracting may be significant; and move the government away from a sustainable debt situation and the priorities in its development program. Debt contracted by independent agencies in which sector ministries own stock poses similar risks. Officially raising the concessionality threshold of 35 percent for public borrowing might run the risk of cutting off finances from the private credit market, a source that has been unavailable historically to Tanzania because of the perceived risk of lending. Any credit from commercial sources, such as private banks, should receive special scrutiny by the DMC; and, no loans should be accepted at grant element less than the government agreed 35 21 Other large creditors are Bulgaria, India and Kuwait. 83 percent under the PRGF program (as has been the practice by GOT in recent years). As per the NDS, it should be demonstrated by the sponsoring sector ministry, and confirmed by EFD, that financing grants or credit at more favorable terms are not available. 5.21 If properly executed, acquisition of supplies and maintenance services should be under the purview of the public procurement system with international tendering. Not only is proper dissemination of information to creditors - including private banks - required, but also a measure to discourage creditors that by-pass the established systems might be the actual risk of non- payment of claims by MOF. It is recommended that special scrutiny and oversight be exercised on the external debt obligations, that arise from supplier credit arrangements for goods and service by sectors ministries. The commitment control system should ensure that such credits are not secured outside the established procedures. 5.22 Short term credit: Another key risk area in the context of acquisition of goods and services by public entities is short term financing. Ideally this should be budgeted for in the particular sector/departments plans. The National Debt Strategy does not cover public debt of periods less than 12 months, but if such short term debt obligations are not met the arrears can negatively impact the overall debt sustainability. One of the measures to discourage such behavior might be internalization of the costs in the subsequent budgetary allocations to the sector ministry that contracted the loan. It is recommended that special scrutiny and oversight be exercised on debt obligations, specially to external creditors, that arise from short term financingfor the sectors ministries. 5.23 Based on the data made available by the BOT, Table 5.3 presents the sectors and the maximum liabilities of both the bilateral non-consecessional and commercial loans. The information provided shows a wide range of grant element included in these loans, even for those categorized as 'concessional' in the BOT records. As seen, most of these loans were incurred before 1990, and none since 1996. In order to learn from experience it might be important to study the context and reasons for the specific loan acquired by the government (from Barclays' Bank for purchase of air traffic control system from BAE) and for loans by the public corporations between 1990-95. 84 Table 5-3: Status of Commercial and Non-Paris Club Loans A. Commercial Loans Number ate of Debt Contracting 2001 Major Sector of Loans Range of Size of Largest Size of Largest Grant Element Loan: DOD Principal Arrears Pre-1 990 1990-95 Post-1995 (%) (USD) (USD) Debtor: Government Hotel 0 Machinery 0 Physical Infrastruct. I 1 35.3 10,354,000 0 Transport I 1 43.4 1,957,333 5,382,667 Commodity Credit 0 0 0 Other 2 2 1.3 - 85.7 1,603,716 3,926,793 Debtor: Public Corporation Hiotel 4 1 3 0 - 2.5 3,625,000 15,950,000 achinery 3 -2 0 - 35.9 0 35,101,062 Physical Infrastruct. 6 6 7.3 - 28.5 0 44,121,727 Transport 3 3 7.6 - 13.1 0 9,211,788 Commodity Credit - 8.0 0 4,460,200 Other 10 1 1.3 - 73.4 1,603,716 78,114,363 B. Bilateral Loans - identified as "Concessional" in BOT records Debtor: Government Hotel 0 0 Machinery 4 4 25.1 - 62.9 0 49,819,636 Physical Infrastructure 12 12 7.1 - 59.1 0 14,061,249 Transport 3 3 16.5 - 58.9 0 9,700,000 Commodity Credit D 0 flier 7 7 28.3 - 62.9 0 22,596,724 btor: Public Corporation - one C. Bilateral Loans - identified as "Non-Concessional" in BOT records Debtor: Government Hotel 0 0 Machinery I l 3 o 24,000,000 Physical Infrastruct. 2 2 4.0- 18.9 0 50,000,000 Transport I l 47.5 0 10,720,809 Commodity Credit 0 0 I0 0 Other 112 112 11.1 - 42.1 0 63,827,907 Debtor: Public Corporation Hotel D 0 Machinery 2 2 15.9 - 40.6 0 6,080,427 Physical Infrastruct. I 1 7.7 0 4,200,000 Transport O 0 0 Commodity Credit 0 0 0 0 Other 6 6 4.7 - 28.0 0 20,221,840 Source: BOT External Debt Department 85 5.24 As seen above, a major fiscal risk of external debt arises from contracting of external debt outside the established systems. Below we consider cases of debt with non-transparent disclosure where the fiscal risk may arise, specially over the medium and long termn, due to unsustainability arising out of lack of disclosure in the government fiscal accounts. 5.25 Debt contracted by Zanzibar: Though the perception may be one of large risks, in actuality the Revolutionary Government of Zanzibar is an independent entity for its debt commitment and management. For transactions not governed by the Govt. Loans, Guarantees and Grants Act No. 30 of 1974; and where the RGZ is not acting as an agent of the Minister of Finance of the URT, the Rev. Govt. of Zanzibar is solely responsible for all its debt obligations. Hence, repayments are solely its responsibility and assets of the URT cannot be exposed to attachment locally or internationally. However, given the complex fiscal and financial relationship between the Revolutionary Government of Zanzibar and the government of the URT, any fiscal risks present for Zanzibar also represent a fiscal risk and potential contingent liability for the government of Tanzania. 5.26 Debt contracted by Independent Companies: Special provisions and public guarantees for debt contraction by parastatals, executive agencies and independent companies represent a major source of fiscal risk for the government budget and MTEF. During the recent annual budgets imposing a hard budgetary constraint, the government has given some degree of freedom in defined activities of various ministries, e.g. to own stock in independent companies, for funding its OC and development expenditures. The financing arrangements that these companies enter into with private banks for investment and supplies may increase the debt exposure of the government budget, depending on the liability for the companies' debts that the government.bears through the specific ministry. It also fosters non-transparency in both the actual and the perceived government liabilities. Both the earnings and the expenditure have an impact on the resources and expenditures presented in the GOT MTEF. The government needs to scrutinize the Acts establishing tltese agencies and take stock of the total liabilities that are explicitly its responsibility through these Acts of establishment, including, where necessary, by seeking an independent legal opinion regarding the likely exposure of GOT. 5.27 Debt contracted under 'special provisions' for sector Ministries: The government needs to take stock of the special exemptions granted to these institutions/ Ministries during the period of their existence for supplementing their budgetary resources. The majority of these debts arise due to financing arrangements like supplier's credit. e.g. the Ministry of Defense and National Service has been granted a mandate to directly procure military equipment of a sensitive nature for security purposes outside the now-streamlined public procurement system. But due to the existence of this exemption and past practices, much of the general procurement by the military - for uniforms, food, stationary, petrol - is done prior to commitment for these and the generation of requisite LPOs. This will be dealt with in further detail in the section of fiscal risk. If the transactions of theses agencies involve externalization, their debts are even more difficult to supervise through the public oversight systems in place currently. The case below is one such example. 86 Box 4: RISKS FROM AGENCY DEBT CONTRACTING THE CASE OF MEREMETA LIMITED The Company Meremeta is a limited company registered under the 1985 Companies Act. The company, incorporated on 19 August 1997, was established as a "non-profit making govemment agent" to implement a program under which it would purchase on behalf of GOT, unwrought gold from artisanal miners in informal gold mining in Tanzania. The govemment holds a 50 percent stake in the company, with paid share capital worth $80. The other half of the share capital is allotted to a South African private company, Triennex (PTY) Ltd. (itself a conglomerate of South African gold exporters - TFM, Grinel, GMT etc). The interests of the GOT are represented by the Ministry of Defence. R.J. Schwartz is the beneficial shareholder of this company representing the interests of Triennex. Net income from Meremeta's operations (after costs and debt service) will be govemment revenue, with no dividend rights for the GOT partner, TRIENNEX. TRIENNEX, however, receives procurement and sales commissions from the operations. "Agreement and Memorandum; and, Articles of Association" govem the operations and the substantive relations between the government and Meremeta. Objectives and The company was formed primarily as a means to supplement budget resources for Principal Activity defense operations in light of the tightness of cash budgeting, under which it is claimed only 45 percent of the budgetary needs of the armed forces were met. Many of the un-funded expenditures are deemed essential for the armed forces; and, the creation of this company by the MoDNS was envisaged a means of lifting the "unduly" hard budget constraint. Originally its principal activity was to implement a program for marketing gold from artisanal miners of the major gold production areas in Tanzania; and, to procure the beneficiation and sale of such gold. Later, technical assistance and training of small scale miners and leasing of modern equipment to them; and recently, mechanized mining was added to the list of activities undertaken by the company. The operations are largely carried out through sub-contracting arrangements - e.g. assaying services by Impro Analytics; financial services by NBC and BOE Natwest; air charter services by Aimet Aviation (PTY). Company The Board, chaired by the Permanent Secretary MoDNS, is the supreme organ of the Govemance company. The Board is made up of seven Directors, six of whom are from the MoDNS and one member representing TRIENNEX. The Directors hold no direct shares or interest in the company. The directors are responsible for the keeping proper accounts to ensure compliance with the requirements of the Companies Act 1985 and for safeguarding the assets of the company. 87 Government Legally, the exposure of GOT to the liabilities of Meremeta Ltd. is governed by the Exposure to the "Agreement and Memorandum and Articles of Association". In a letter of August 18, Company's 2000 the government provided the following important clarifications on the Liabilities stipulations of the 1997 constitutive Letter of Instructions to the company, " there shall be no government or Central Bank guarantees or indemnities during the implementation of the Meremeta Programme. In this respect and since the Ministry of Defense is part of the Tanzania government, the Agreements (contracted by Meremeta) should state clearly the government of the URT shall not be liable for liabilities arising out of or in connection with the implementation of the Agreements". The agreements also indemnify the GOT and the other shareholder of liabilities of all operational sub-contracts. Furthermore, in the same letter the govemrnent reiterated the Promissory Note facility would be financed only from realized sales proceeds. The primary source of such exposure would be sub-contracts for operations of the company's program and debt contracted by the company in which the government holds 50 percent share. Meremeta Ltd. has used its current and projected earnings/value added as security in signing contracts with private companies to procure capital and military equipment; and, to guarantee debt in such procurement. To date the total debt exposure of the company to such debt amounts to $95 million; and, is set to increase to $130 million to enable commissioning of the Buhemnba gold mine (in addition to the existing operations in Geita). The expected value from the gold sales (gross) of the mine was assessed by a consulting company to be $350 million over its life time. The annual net income is expected to be not less than $15 million; and, will be directed to expenditure by MoDNS. Fiscal Risks and While the agreements provide protection of government from Meremeta Ltd.'s Consequences on liabilities; and, the debt is supposedly secured by current and prospective assets of the Budget company (mainly gold in the ground), it is not quite clear whether the stipulations in Management the agreement would in practice stand under international arbitration. The GOT being a beneficiary of net income of Meremeta Ltd. would be enjoined in any proceedings of default and it would be useful to seek further legal opinion on possible exposure to risk in international arbitration. In any event, the GOT being a shareholder the scrutiny of debt contracting by Meremeta Ltd. should be on similar rigorous grounds as all other government or public guaranteed debts. The Minister of Finance should sign off on such debt only when satisfied that all tests have been met. What is perhaps even more important is the fact that Meremeta Ltd. is prospectively going to be a large source of extra budgetary funding for military operations. Such financing should be brought into the overall expenditure allocation to preserve the integrity of the MTEF. In other cases where similar self-financing is practiced, the retention scheme is subject to the budget ceilings consistent with the overall priority I allocation. 5.28 There is a risk that for any such company where government is a beneficiary of net income, government could be enjoined in any proceedings of default. Hence, it would be useful for the GOT to get legal opinion on possible exposure to risk in international arbitration. 5.29 The case of Meremeta Ltd. Provides an example of the debt exposure and fiscal risks involved. It is recommended that GOT being a shareholder of any such company, the scrutiny of its debt contracting should be on similar rigorous grounds as all other government or public guaranteed debts. The Minister of Finance should sign off on such company debt only when satisfied that all tests have been met. 88 5.30 More importantly, it is recommended that as such a company may prospectively be a large source of extra budgetary funding, such funding should be brought into the overall expenditure allocation to preserve the integrity of the MTEF. In cases where similar self- financing is practiced, the retention scheme is subject to the budget ceilings consistent with the overall priority allocations. 5.31 It is recommended that laws of establishment of public executive agencies, parastatal companies, and sector ministries that grant these entities special exemptions in external debt contracting - both for finance and for acquisition of goods and services - be scrutinized and revised for consistency with the Loans, Guarantees and Grants Act, no 30 of 1974 (and revisions to it being prepared by the MOF) and the National Debt Strategy 1999 (Part 1). In the medium term, such exemptions should be minimized arid abolished. The public - especially the legislature - should be educated and sensitized on the legal roles and requirements; that prescribed sanctions be transparent; and stringency in enforcement of sanctions be instituted to deter future occurrences. 5.5 PERCEIVED RISKS OF INSTITUTIONAL ROLES AND IMPLEMENTATION 5.32 The Minister of Finance is the final authority to contract external debt in Tanzania by the Government Loans, Guarantees and Grants Act, No. 30 of 1974. He is supported by appropriate functions in the following government agencies in order to execute his debt contracting decisions: External Finance Dept., Policy Analysis Department (with its Debt Management Unit) and Accountant General's Office in the MOF; and, the Foreign Payment Dept. and External Debt Dept. in the BOT. The roles are defined in the National Debt Strategy, as originally drafted in September 1997 and revised in March 1999. 5.33 In the recent past (mid 1980s -mid 1 990s) when Tanzania was under severe macroeconomic pressure the BoT served as the key agent in debt contracting and management to achieve macroeconomic stability and compensate for the weak capacity in the Ministry of Finance to monitor its debt obligations. But based on the achievements since 1995, which included fiscal and macro-economic stabilization as well as the strengthening of capacities in the Ministry of Finance, it may be appropriate to revert to the institutioanal roles as they were originally envisaged. The roles of PAD, EFD, and ACGEN in MOF; the BOT; and the DCC/DMC in external debt management were described in the earlier section. 5.34 In addition to its role as GOT's banker, BOT is also responsible for the supervision of Tanzania's balance of payments. To fulfill this role most efficiently BOT needs accurate and timely information on all of Tanzania's external obligations, private and public. The External Debt Department of BOT is the custodian of all information about private loans that have an external obligation component. In addition, it currently maintains information on the public loans from foreign sources, without relying on the information available from the ACGEN's database (mostly collected by BOT directly from the EFD at conclusion of each negotiation). Information available from MOF on the foreign grants negotiated is accepted by BOT, with less importance attached to the accuracy of this information since its does not increase Tanzanian's external obligations. Actions have been initiated to upgrade the accuracy and timeliness of the 89 ACGEN's database in both external loans and grants; but it is far from complete. Staff skills and remuneration are crucial constraints in efficient transition and then maintenance and use of the database in the ACGEN's office by both the MOF and BOT. Accuracy of this information is extremely important for the MOF to have full coverage of all foreign funding available for public use, directly through the budget or indirectly through projects; and extremely important for the BOT to execute its responsibilities, beyond external debt management, towards the overall supervision of Tanzania's Balance of Payments. 5.35 Table 5.4 compares the institutional agreements for debt contracting and supervision in Tanzanian with the best practice and the actual practice. It also offers recommendations for strengthening debt management. 90 Table 5-4: Assessment of National Debt Strategy vis-a-vis Best Practice and Observed Practice Guidelines for Public Debt Management Tanzania National Debt Strategy Observed Practice in Tanzania Recommendations (ref. World Bank& lMF,March2001) (Part 1: External Debt, March 1999) I Statement of Objectives a Objective should be to 'raise the Section 5.1.5 Commitment to seek Is sufficient effort always made to locate PAD and DMC/DCC should ensure that required amount of funding [and] grants-first, then financing on highly grant funding? Presentation of a the best possible terms have been achieve its risk and cost objectives' concessional terms (e.g. IDA). Loans project/donor package by the sector to secured. EFD role as resource mobilizer (para 1). only for priority areas. Commercial the MOF at a late stage undermines the to be enhanced. borrowing only in emergencies (5.1.3) ability of MOF to secure the most 'Clear debt management objectives are favourable terms. MOF should consider a sector focus in essential in order to reduce uncertainty' EFD, or liaison with MOF-based sector (para 22) specialists, to ameliorate late presentation of loan-financed projects to MOF and to enhance efficiency of resource mobilization. PRGF program places a ceiling of zero Maintain ceiling of zero. Increase on non-concessional borrowing. scrutiny on loans of 35-50% grant element. Debt Strategy needs updating post-HIPC to focus on prevalent issues, areas of 1974 Act and Debt Strategy to be risk, capacity building, coordination. reviewed in context of identified risks, institutional reforms and capacity needs *b 'The objectives for debt management Sections 4.3 and 4.4 acknowledge the There may be a need to increasingly Annual public debt statement prepared should be clearly defined and publicly important role of civil society. promote the National Debt Strategy in by PAD at time of budget. Should disclosed ... .experience suggests that . the public domain and to hold reiterate main focus of debt strategy and such disclosure enhances the credibility discussions with key civil society recount major developments of past year of the debt management program' (para members, e.g. Tanzania Coalition on (i.e. new loans and purpose, DSA 21) Debt and Development (at the FY02 outlook). PER Meeting?) c 'Seek to ensure that both the level and Sections 5.1.7 (Appraisal Criteria) and Debt Strategy needs updating to reflect DSA and portfolio analysis capacity in growth of public debt is sustainable [by 5.1.8 (M & E) outline measures to post-HIPC situation and increased PAD to be strengthened. Time-bound designing] a credible debt strategy' improve scrutiny of loan-funanced attention to organizational issues and program for strengthening institutional (para 2) projects. capacity building. arrangements to be identified. Capacity for DSA is acknowledged as Also a need for focus on areas identified MOF to consider Missions weak (5.2.2) in the PER relating to debt management, recommendations. I risk and legal safeguards. 91 Guidelines for Public Debt Management Tanzania National Debt Strategy Observed Practice in Tanzania Recommendations (ref. WorldBank&IMF, March2001) (Part 1: Extemal Debt, March 1999) T 2 OrganizationaUlhistitutional Structure a 'The legal framework should clarify the Section 4.2 recounts the 'Government Government is committed to adhering to Simultaneous reflection on Act and Debt authority to borrow and to issue new Loans, Guarantees and Grants Act' of Act (despite past slippages) and is also Strategy required to address legal, debt' (para 29) 1974 - Minister of Finance only considering amendments. institutional, procedural and authorized signatory. administrative issues. Minister of Finance to retain sole contracting It is acknowledged that the 1974 Act has authority. Mechanisms for improved not been strictly adhered to. adherence to be identified. b 'The organizational framework should Chapter Four. Roles of MOF, Planning, In practice the arrangements are Clarify and make explicit the roles of be well specified and ensure that Attorney General, BOT, somewhat different to those documented each department/institution in the Debt mandates are well articulated"(para 31) sectors/implementing agencies are in the Strategy, especially with regard to Strategy (as per Section 111.2 of documented. BOT and ACGEN. Recommendations). The division of responsibilities within Planning seems to be completely Projects must be evaluated for MOF is not always made clear and bypassed (acknowledged in 4.1.3) - consistency with national guidelines and requires elaboration. The lack of does this mean that projects are not assurance of efficiency (value for coordination is acknowledged (4.1.5 iii) given proper scrutiny for consistency money). but solutions are not proposed. with national objectives; likelihood of attaining objectives; efficiency/value for Section 5.2.2 outlines the need for money? clearer information channels and linkages. In a few cases, it appears that creditors Role of ACGEN to be recognised as per do not advise ACGEN/MOF of Section 111.2 of Recommendations. payments falling due, rather sending the demand to the implementing agency. Off-budget projects create difficulties in recording disbursements. BOT also continues to receive some demand notes directly. Strengthening sector-management Weaknesses in project monitoring are capacity at MOF (in EFD?) would create noted (4.1.8) which may result in greater awareness of project status. continued disbursement (and consequent indebtedness) for projects that might be better cancelled. ___ 92 Guidelines for Public Debt Management Tanzania National Debt Strategy Observed Practice in Tanzania Recommendations (ref. World Bank & IMF, March 2001) (Part 1: External Debt, March 1999) c 'Debt managers, fiscal policy advisors The Debt Strategy serves this purpose. All key institutions appear to be aware It will be important to maintain this high and central bankers should share an The Debt Coordination Committee and of the objectives. level of awareness as institutional roles understanding of the objectives of debt the Debt Management Committee are clarified and personnel changes are _ management' (para 16) maintain awareness. effected. d 'Debt management, fiscal and monetary This is the role of the Debt Coordination In the context of the budget and the Increased coverage of contingent authorities should share information on Committee and the Debt Management MTEF, this appears to be satisfactory liabilities to be attained over time. the government's current and future Committee with respect to medium- with regard to all recorded debt. Setting of database at ACGEN must liquidity needs' (para 18) long-term repayment profiles of total maintain ability to project amortization debt. and interest payments. e 'There should be a policy in place to Section 4.1. 10. All payable debt service Annual meeting on payable debt and Procedures in place should continue enable the authorities to meet with a is budgeted at the start of the FY. MOF identification in the budget. Clear with the role of the ACGEN being high degree of certainty their financial transfers funds to BOT which then procedures and responsibilities for strengthened as it assumes responsibility obligations as they fall due' (para 48) externalizes. Procedures appear to be timely externalization. Appears to for data and processing payments. clear. function adequately for most recorded Capacity at ACGEN will remain a debt, although ACGEN does not receive concern and must be addressed in light Zanzibar debt is identified as a problem all demand notes directly, resulting in of these increased responsibilities. area occasional penalty interest charges. f Debt management activities should be Section 5.2.2 commits GOT to building Need to resolve the respective Time-bound program for building supported by an accurate and improved systems responsibilities of BOT -v- ACGEN ACGEN capacity and transferring comprehensive management information (loan - BOT, grant - ACGEN?; on-line responsibility for maintaining database system' (para 36) system?) from BOT. Parallel systems to be maintained for [2] years to ensure The BoT system appears to be superior, integrity of database. Continued sharing although Acc Gen is the designated of data with BOT to be facilitated. keeper of records. g Build appropriate technical Section 5.2.2 commits GOT to building Need to resolve BOT /ACGEN roles As above infrastructure - such as a central registry improved systems and devise time-bound program for and payments and settlement system' realizing any change of roles para 1) . 93 3 Transparency and Accountability a 'effectiveness can be strengthened if the Section 4.3 defines the role of civil Particularly in the irnmediate post-HIPC PAD to produce annual statement on goals and instruments of policy are society and the media in terms of, inter era, assurances should be given at the external debt at the time of the budget. known to the public' (para 19) alia, 'forming a commnon front' on debt highest level of continued adherence to issues and ensuring that loan financed the debt strategy. H.E. The President and Cabinet to projects achieve their intended goals. endorse new Debt Strategy. 4.3.2 outlines a broad partnership Efforts should be made to sensitize civil Parliament and civil society to be approach between GOT and CSOs, society (e.g. through Tanzania Coalition sensitized on debt issues. Perhaps at including information sharing and on Debt and Development). Is it fair to May 2002 PER, for example? networking. say that 4.3.2 has not been attained? b 'The allocation of Section 4.4 promotes awareness of all Debt strategy to be a public document?; Annual statement prepared by PAD to responsibilities .., should be publicly aspects of debt. greater Parliamnentary oversight (at least facilitate this. disclosed' (para 20) at Committee level)? c 'The public should be provided with Section 4.4 promotes awareness of all Releases and quarterly performance are The PER FY02 'Easy Reference' guide information on the past, current and aspects of debt and undertakes to published. PER increasingly inclusive should include a section on debt together projected budgetary activity' (para 25) provide this information. and efforts in place to enhance with other economic dissemination. management/performnance issues. d 'The government should regularly Section 4.4 promotes awareness of all Existing publications such as 'Bank of Technical reporting (BOT joumals) to publish information on the stock and aspects of debt and undertakes to Tanzania Monthly Economic continue. Public annual report, prepared composition of its debt' (para 26) provide this inforrnation. Review/Quarterly Economic Bulletin' by PAD, to be released at time of address this need. budget. e 'Debt management activities should be No external audit. MDF was subject to BOT Quarterly Economic Bulletin The main areas of risk need, in the first audited annually by external auditors' independent audit, however. reports externat debt developments. instance, to brought under closer central (para 28) Regular govemment audits (with more Governnent control and scrutiny. teeth?) also address this in part. ACGEN to be subject to standard accountability oversight. 94 4 Identification of Risks a 'Poorly structured debt ... has been an Maturity structure, interest rate Informally since 1999 and formally DCC/DMC to be supported by PAD important factor in inducing or considerations are observed. since 2000, this is done at the loan (which serves as secretariat) to ensure propagating economic crises' (para 3) Consideration is also given to consideration stage by DCC/DMC - this function is fulfilled. Portfolio amortization profile's impact on GOT with consideration of the GOT's priority analysis to consider profile of liquidity. sectors; concessionality terms of the amortization schedule with respect to loans; and, bunching up of liquidity constraints disbursements/ repayments. However, 4.1.5 (i) reports that counterpart funding is sometimes not Counterpart funding issues should be To be addressed in context of cash flow available. This liquidity problem results investigated. management. __________________________________ in delays in project implementation. b 'There is a need for governments to limit As above The exposures of the multilateral and Need for increased scrutiny by DCC on the build up of liquidity exposures' (para bilateral Paris Club debts are well loans with terms near threshold of 4) managed. GOT is not servicing its non- concessionality. Sector driven Paris Club and commercial loans; hence negotiations where the PAD/EFD and sending the appropriate market signal. ACGEN are brought in at late stages of appraisal should receive increased _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ ___ scru tin y c 'risks can be readily addressed ... by Section 5.1.4 Guarantees to be Need to identify contingent liabilities, to Commercial/Supplier credits and short- reviewing criteria and govemance scrutinized by the Debt Coordination assess and mitigate risk and to improve term credits identified as major source arrangements in respect of contingent Conunittee (DCC) disclosure/monitoring mechanisms and of risk to Debt Strategy. Identifying liabilities' (para 5) incentives. contingent liabilities, verifying GOT liability and enforcing commitment Supplier credits for sectors are also controls/existing procurement legislation problematic. will minimize exposure d 'Debt managers should consider the As above As above This should include an independent legal impact that contingent liabilities have on opinion, where necessary, on GOT the govemment's financial position' exposure with regard to independent companies part owned by GOT or a public institution. e 'Need to ensure oversight of 'off- As above As above As above balance sheet claims, including contingent liabilities such as state I guarantees' (para 15) _ 95 f 'monitor and review potential Section 5.1.4 Borrowers are to be Is securing collateral a credible strategy? The feasibility of this policy may need exposures ... .from guaranteeing debts of required to pledge assets as security (i.e. would GOT move to cover its to be reviewed if a credible strategy is to sub-central governments and state- exposure by actually taking control of be identified. _ owned enterprises' (para 15) assets?) g 'Risks of govermnent losses from Not directly addressed More consideration to: Stricter adherence to established inadequate controls should be managed Commitment controls procedures has been shown to be according to sound business practices, Clearer guidelines on supplier credits effective. Continued strengthening of including well articulated Continued incursion of penalty interest this capacity will continue to be _ responsibilities' (para 33) for late payments beneficial. h 'The risks inherent in the government's Not directly addressed More consideration to issue of supplier As above debt strategy should be carefully credit, short-term borrowing. monitored and evaluated' (para 39) Identification of strategies to identify and to ensure adherence to debt strategy. i 'Misreporting of contingent or Measures to mitigate are not addressed This is a relic of the past and it is hoped Continued vigilance is necessary to guaranteed liabilities [may be caused by] that measures will preclude this in the monitor contingent liabilities and to inadequate coordination or procedures' future. GOT has prudently budgeted ensure GOT exposure is made clear in (Box 2, No. 3) Tsh. II billion for 'unpleasant surprises' the case of debt contracted by _ _. ____________________________ independent companies (Section IV.A) j 'To assess risk, debt managers should Section 5.2.2 acknowledges that MOF DSA capacity needs to be reinforced PAD identified as department where regularly conduct stress tests of the debt requires additional training in these (especially in MOF). Monitoring of DSA skills must be concentrated. portfolio' (para 53) functions. debt/macro variables, export trends etc Training program to be presented for and consideration of new instumments to implementation. ameliorate commodity price swings. k 'In order to minimise cost and risk, debt No discussion on linkage to domestic Perhaps not of immediate relevance for managers should ensure that their debt issues or the development of the extemal debt issues but important for policies are consistent with the domestic financial sector. longer term development of the funancial development of an efficient govemment sector and domestic/extemal debt securities market' (para 64) linkages (e.g. should domestic borrowing once more be used to finance _ ~~~~~~~~~~~~~~~~~~~~~~expenditure). I 'Debt managers should carefully assess Not addressed (implicitly assume that and manage the risks associated with GOT canot raise such capital on such _foreig curncy debt' (para 45) concessional terms locally) 96 5 Broader Considerations a 'Sound debt management policies are no Chapter I (& 3.2 - 3.3). Debt strategy Commitment to stronger economic, Continued prudent economic panacea or substitute for sound fiscal is part of a comprehensive economic budgetary and debt management is management and pursuit of reforms will and monetary management' (para 6) management strategy which places due acknowledged by central ministries. complement the Debt Strategy. emphasis on monetary, fiscal and other (e.g. trade, investment) considerations Need for making realistic projections of PAD and EFD to consider scenarios on the impact of new debt liabilities from new borrowing, particularly in light of all government agencies and institutions the PRSP financing gap. with public guarantees over the medium term in the MTEF, to be transferred into the budget at the appropriate time. Need to manage fiscal risks arising due The PER addresses fiscal risks to behavior of government agencies separately but contingent liabilities and outside the standard rules that have been other risks are considered above. put in place. b 'authorities should pay greater attention Debt Strategy and macro fundamentals Debt Strategy overseen by PAD who Need to stress the legal role of MOF to the benefits of having a prudent debt will dictate terms for filling financing work with Budget staff to ensure (Debt Department in the PAD; and management strategy that is coordinated gap. accurate projections on disbursements/ DMC) as key in debt strategy, with a sound macro policy framework' repayments for MTEF. DSA work weak particularly with regard to DSA work _ (para I I) with respect to macro indicators. linking debt and macro variables. c 'Capacity building and TA must be Sections 5.1 .8(ii) makes reference to No integrated medium term capacity The post HIPC Debt Strategy must view carefully tailored to meet stated policy training. 5.2.1 states that 'a strong building programn for each institution debt contracting and management as one goals' (para 9) capacity in MOF will be built' but more appears to be in place. Historically, integrated process. The specific details would be welcome. capacity building has been located in capacity needs of all key entities should BOT, followed by current capacity be considered in a coherent plan under enhancement in ACGEN. the CFAA/PFMRP follow-up. 97 ANNEXES 98 ANNEX 1: REALLOCATIONS FROM AND TO THE MINISTRY OF FINANCE CONTINGENCY POSITION MOF Contingencies Reallocation From Ministry of Finance to the Sectors 2000/01 Reallocation From the sectors to the Ministry of Finance, 2000/01 Sectors Amount Sectors Amount Public Debt Service 1,126.275,000 Public Debt Service 5,125,127,030 Presidenfs Office & Cabinet 2,021,185,752 Ministry of Home Affairs 667,566,536 Civil Service Department 876,400,000 Ethics Secretariat 5,902,200 Permanent Commission of Enquiry 83,878,000 Pemmanent Commission of Enquiry 54,808.000 Prime Minister's Office 547,690,000 Civil Service Commission 18,600,000 Ministry of Justice & Constitutional Affairs 79,850,000 Office of the Speaker 220,000,000 Ministry of Lands and Human Settlement Ministry of Agriculture and Food Security 2,257,123,000 Development 2,261,143,000 Ministry of Industry & Trade 28,980,000 Ministry of Home Affairs 796.055,640 Ministry of Education 1,041,135,000 Ministry of Health 401,765,100 Ministry of Community Development, Women Ministry of Home Affairs 121,000 Affairs and Children 62,892,200 Ministry of Health 92,110,600 Law Reform Commission 27,900,419 Ministry of Defence and National Services 1,574,218,000 Planning Commission 1,806.000,000 Ministry of Energy and Minerals 640,110,837 Teachers Service Commission 23,918,600 Ministry of Science, Technology and Higher Accountant General 6,300,000.000 Education 206,807,700 Ministry of Cooperatives & Marketing 723.582.300 Ministry of Tourism and Natural Resources 382,415,771 Vice President 198,833,020 Regions 378,440,455 Ministry of Home Afrairs- Police Force 3,693,183,478 TOTAL 10,500,355,434 Ministry of Foreign Affairs and International Cooperation 6,568,141,583 Ministry of Defence 3,050,528,100 The National Service 628,212,000 Judiciary 804,245,000 Office of the Speaker 360,000,000 Exchequer and Audit Department 120,000,000 Ministry ofWorks 14,466,554,831 Ministry of Water and Livestock Development 600,000,000 Ministry of Community Development, Women and Children 33,100,000 Industrial Court of Tanzania 1,744,000 National Electoral Commission 1,722,000,000 Ministry of Communication and Transport 481,695,980 Local Govemment Service Commission 2,420,000 Ministry of Labour Youth and Development 50,000,000 Ministry of Science, Technology and Higher Education 350,460,600 Ministry of Natural Resources and Tourism 175,471,130 Regions 9,135,090,647 Total 59,834,339,858 99 ANNEX 2: DEBT RELIEF AND FISCAL DEFICITS Tanzania reached the completion point under the HIPC initiative in November 2001. This implied irrevocable debt relief in the amount of US$2,026 million in net present value terms and a reduction in the nominal value of debt service by 58 percent. The expectation of the donor community is that Tanzania uses the resources freed through the debt relief to augment poverty reducing expenditures. In order to be able to apply these resources to poverty reducing measures, it is important that the government's fiscal strategy, and in particular fiscal deficit targets, are appropriately adjusted to take into account the debt relief. A complication arises from the fact that the various participating creditors use widely differing modalities to provide debt relief. The modalities range from the reduction in interest rates to the provision of grants to cover debt service to an overall reduction in debt service, which in cases of highly concessional lending implies essentially a reduction in loan repayments. As we show in this annex, the type of modality used to provide debt relief can directly affect the magnitude of the fiscal deficit shown in Central Government Operations tables, even if everything else remains unchanged. To illustrate this point, table Al shows the fiscal accounts of an hypothetical economy. Column 2 shows the fiscal situation of the country prior to debt relief where 50 local currency units (LCUs) are spent on interest payments and 100 LCUs are spent on repayments of debt. Table Al: Impact of HIPC debt relief on measures of the fiscal deficit Debt Relief through Before Debt Relief Reduction in Reduction in Principal Interest Payments Grant Repayment (In Local Currency Units) Revenue 300 300 300 300 Interest payments 50 10 50 50 Other Expenditures 270 310 310 310 Primary Deficit 30 -10 -10 -10 Deficit before Grants -20 -20 -60 -60 Grants 50 50 90 50 DeficitafterGrants 30 30 30 . -10 Repayment 100 100 100 60 Other Financing and Capital 70 70 70 70 GDP 1000 1000 1000 1000 (As a Percentage of GDP) Primary Deficit 3.0% -1.0% -1.0% -1.0% Deficit before Grants -2.0% -2.0% -6.0% -6.0% Deficit after Grants 3.0% 3.0% 3.0% -1.0% The country receives debt relief which results in an annual reduction in debt service by 40 LCUs. In line with expectations of the international community, this debt relief is used to augment expenditures which increase by the same amount from 270 LCUs to 310 LCUs. 100 Columns 3-5 sow the impact of the debt relief on various deficit measures, depending on which modality to deliver debt relief is being used and how it is being accounted for in the fiscal accounts. It is important to keep in mind that there is absolutely no difference between the three scenarios in terms of their economic impact since revenue, non-wage expenditures, and the capital and financial balance net of repayments are always the same. Column three shows the case were all the debt relief is in the form of reducing interest payments. Expenditures on interest payments drop from 50 CUs to 10 LCUs, reflecting fully the debt relief of 40 LCUs. In this case, the only deficit measure that differs from the situation prior to HIPC is the primary deficit (i.e., the deficit before grants excluding interest payments) from a surplus of 1% of GDP to a deficit of 3 %. This is the result of non-interest expenditure having been raised by the amount of HIPC debt relief while revenues remain the same. All other deficit measures remain unaffected, as the reduction in expenditures is exactly offset by the increase in non-interest expenditures. Column 4 shows the second case where debt relief is provided as a grant to be used to service debt. In this case, the deficit before grants declines from 2 percent prior to HIPC to 6 percent after HIPC. The deficit after grants remains unchanged. Column 5 shows the last case where all debt relief is directly reflected in the fiscal accounts as a reduction in repayments, which decline from 100 LCUs to 60 LCUs. In this scenario, all three deficit measures, i.e. the primary deficit, the deficit before grants and the deficit after grants deteriorate by the full amount of HIPC debt relief. As this example illustrates quite clearly, HIPC debt relief with an equivalent increase in non- wage expenditures will generally lead to a deterioration in various measures of the fiscal deficit. This is important to take into account in the design of fiscal policy and in the setting of fiscal deficit targets. Leaving targets for fiscal deficits at the level prior to HIPC debt relief would in fact imply that HIPC debt relief cannot fully be used to increase poverty reducing expenditures. A second point which this example illustrates is the limited value of the fiscal deficit as an indicator of fiscal and debt sustainabity and that the analysis is more appropriately done on the basis of the changes in the net present value of a country's debt. The way HIPC debt relief is treated at present in the fiscal accounts in Tanzania is a combination of the three modalities discussed. Debt relief on IDA and other multilateral debt is shown as a grant while the figures for interest payments and repayments are shown as gross figures. For bilateral debt, the Ministry of Finance shows the net amounts for interest payments and repayments after the application of HIPC debt relief. With the annual reduction in debt service from HIPC amounting to about I percent of GDP, deficit targets will need to be adjusted to take into account the impact of HIPC and to allow for the use of HIPC funds to increase poverty reducing expenditures as intended. Part of the deterioration in the fiscal accounts of the previous year is also attributable to the fact that interim debt relief had already been provided by some multilateral and bilateral donors to the government of Tanzania. 101 ANNEX 3: DEFINITION OF POVERTY-REDUCING EXPENDITURES IN THE BUDGET VOTE SUBVOTE ITEM DESCRIPTION P.E. O.C. TOTAL BUDGET TOTAI .RFGIIRRFNT FXPFNnITIIRF. (e xclidinv CFS1 1. RASIC FnilrATlON- 70-89 1001" all O.C.Subvention to Local Authorities 70-89 2004 260207 Examination exDenses in the RAS budget 70-89 2004 all Other O.C. Items in the RAS budget 46 3001 all MoEC - Basic Educ.Subvote O.C. 46 5001 all MoEC - Teacher Educ.Subvote O.C. 46 2002 all MoEC - 75% of InsDectorate Subvote O.C. 46 2001 280558 Subvention to Institute of Adult Education 46 1001 280510 Subvention to Nat. Examination Council 67 all all 75% of TSC O.C. Total Basic Education 2. PRIMARY HEALTH: 70-89 1001"* all O.C.Subvention to Local Authorities 52 2001 260402 Local Authorities drugs allocation (kits) budgeted under MoH 52 3001 all MoH Preventive Subvote 0 C. 70-89 3002 all Regions Preventive Subvote O.C. Total Primary Health 3.RURAL WATER 70-89 1001** all O.C.Subvention to Local Authorities 70-89 2004 270101 Regions O.C. items 49 4001 all Rural Water SuDplY O.C. Total Water 4. RURAL ROADS 70-89 1001"* all O.C.Subvention to Local Authorities 56 2002 280605 Road Fund budgeted for Districts 47 7001"* all MoW-Rural Roads SubvoteO.C. Total Rural Roads S. LAW AND ORDER 40 all all All O.C. under Judiciary DeDartment 41 all all All O.C. under Min. of Justice & Const. 59 all all All O.C under Law Reform Commision 60 all all All O.C. under Industrial Court of 64 all all All O.C. under Commercial Courts Total Law and Order 6. AGRICULTURE RESEARCH AND EXTENSION 70-89 1001" all O.C.Subvention to Local Authorities 43 3001 all MoA - Research DeveloDment Subvote 43 2001 all MoA - CroDs Develovment Subvote O.C. 43 4001 all MoA - Cooperative Development Subvote 43 6001 all MoA - Livestock DeveloDment Subvote 49 1004 all Livestock Research and Training 49 7001 all Veterinary Services 49 8001 all Animal Production Total Agriculture Research and Extension 7. HIV/AIDS* Total PrioritV Sectors % Share of Total Recurrent Budget fexcl. CFS) B Breakdown for suhvention to I ocal Authorities is found in Annendices to Volume III of Fstimate Book Source: Ministry of Finance. Budget Department. 102 ANNEX 4: COMPOSITION OF SPECIAL EXPENDITURE ITEM UNDER VOTE 50 (FY00 - FY02) Composition of Special 1999/00 2000/01 2001/02 Expenditure Budgeted (Tsh. %/. of (3) Budgeted % of (3) Budgeted % of (3) Bn) (Tsb. Bn) _ _ (Tsh. Bu) (1) Contingencies 66.6 61.0 11.7 26.3 137.9 59.3 Salary adjustments 41.5 38.0 0.0 42.7 18.4 (ministries & parastatals) Parastatal debts, 10.0 9.2 0.0 44.0 18.9 retrenchment _ Unforeseen emergencies / 9.2 8.4 11.7 26.3 9.3 4.0 contingency Payment of afears 4.0 3.7 0.0 41.9 18.0 Pension 2.9 2.7 0.0 0.0 0.0 0.0 0.0 (2) Other special 42.5 389 32.8 73.7 94.5 40.7 expenditures TRA 20.9 19.1 3.9 8.8 0.0 Electricity bills 17.0 15.6 0.0 0.0 Contractual obligations 4.0 3.7 4.0 9.0 0.0 CMSA 0.6 0.5 0.0 0.0 New employment (primary 0.0 2.5 5.6 0.0 school teachers) Health insurance 0.0 9.4 21.1 10.9 4.7 HIV/AIDS campaign 0.0 4.8 10.8 4.4 1.9 5.5% of grants to Zanzibar 0.0 2.0 4.5 6.3 2.7 Govvernment Parliamentarians pension 0.0 5.5 _ _ 12.4 0.0 Anti-conuption campaign 0.0 0.6 1.3 0.7 0.3 Airport service charge _ 0.0 0.0 1.6 0.7 Housing allovwance (higher 0.0 0.0 3.5 1.5 learning institutions) Abolition of primalry school 0.0 0.0 11.0 4.7 fees VAT refunds 0.0 II_ 0.0 56.0 24.1 0.0 __ 0.0 0.0 (3) Total Special 109.3 100.0 44.5 100.0 232.4 100.0 Expendiure 103 ANNEX 5: TANZANIA DEBT MANAGEMENT AS ASSESSED BY THE JOINT IMF/IDA TEAM FOR HIPC COMPLETION POINT22 Institutional framework Debt management is the joint responsibility of the Ministry of Finance and the Bank of Tanzania (Central Bank). Public or publicly guaranteed debt can only be contracted with the approval of the Ministry, however the technical work related to the granting of state guarantees is handled by both the Ministry and the Bank of Tanzania. The government prepares and publishes a comprehensive annual debt strategy report in addition to monthly and quarterly reports. Debt recording and reporting Monitoring of public and publicly-guaranteed debt has been good. The loan data is maintained using the Commonwealth Secretariat database system, CS-DRMS. The database is generally up-to-date and is regularly maintained by both the Ministry and the Bank of Tanzania. Information on contracts, disbursements, debt service and reports are exchanged weekly between the Ministry and the Bank of Tanzania. However, the current system is rot able to produce data directly in an analytical form. Debt management and borrowing strategy A committee is in place, composed of the Ministry of Finance, the Bank of Tanzania, the Planning Commission, the Attorney General Chamber and Sectoral Ministries, to oversee debt management and coordination. Procedures are also in place to track the public sector's debt servicing needs and the impact on the budget and Balance of Payments. Borrowing policy is developed by a committee comprised of the Ministry of Finance, the Bank of Tanzania, the Planning Commission and Attorney General Chambers. Analytical capacity The technical staff at both the Ministry of Finance and the Bank of Tanzania are experienced and knowledgeable about debt issues and have carried out numerous debt restructuring renegotiations. There is more limited capacity within the debt teams to integrate debt and macroeconomic simulations and have tended to be produced in conjunction with the Bank and the Fund. 22 This is Appendix I from Memorandum and Recommendation of the President of IDA to the Executive Directors on Assistance to the URT under the Enhanced HIPC Initiative; Report no. P7492-TA, Nov. 8, 2001' This was based on the attached Questionnaire filled up by the GOT. 104 ANNEX 6: QUESTIONNAIRE ON EXTERNAL DEBT MANAGEMENT IN HIPCs For Directors/Key Responsible of Debt Management Preamble The HIPC Initiative is designed to reduce the level of external public and publicly guaranteed debt to 150 percent of exports (or, in some cases, to 250 percent of fiscal revenues), thereby eliminating a critical barrier to long-term debt sustainability. However, assuring debt sustainability over the longer term will also require, among other elements, prudent and effective sovereign debt management. This was highlighted in the recent joint Bank-IMF paper "The Challenge of Maintaining Long-term External Debt Sustainability", discussed by the Board of Governors of the IMF and the Bank during the 2001 Spring Meetings, and subsequently reinforced in the joint communique of the IMFC/DC which stressed that "debt management [in HIPCs] needs to be strengthened." In this perspective, Bank-Fund staff have been requested to prepare a joint paper for the Boards of both institutions that reviews the status of debt management capacity in HIPCs and highlights the areas in need of strengthening. This Questionnaire has thus been prepared to solicit from all HlPCs information about the key aspects of their external debt management. Such a self-assessment will provide the basis for identifying remaining capacity-building needs and provide a starting point for further assistance efforts aimed at establishing the required conditions for a prudent and effective sovereign debt management. Key Areas Covered by this Questionnaire With the view of acquiring a good overview of the extemal debt management capacity and areas in need of further strengthening, this Questionnaire covers the following principal aspects of debt management: Legal framework and institutional arrangements Coordination with macroeconomic policies, debt sustainability analysis Defining and implementing new borrowing policy Managing the existing debt stock i. Basic debt management functions ii. Debt renegotiations Human and technical resources i. Staffing and training ii. Computer tools Assistance from Donors and Technical Agencies I. LEGAL AND INSTITUTIONAL ARRANGEMENTS FOR DEBT MANAGEMENT 105 1. Are the responsibilities, mandate and objectives of the external debt management agency / agencies set out in a law which is (mark answer as appropriate): _.v,_is comprehensive and clear lacks somewhat in comprehensiveness but is clear lacks comprehensiveness and clarity no law exists for this area Comments 2. If a law is indeed in place, is this law observed and implemented through appropriate decrees / regulations (mark answer as appropriate)? appropriate decrees / regulations are in place which are observed and implemented _ appropriate decrees / regulations are in place but are not observed and implemented the decrees / regulations in place are not appropriate and not fully observed no decrees / regulations are in place Comments: were not observed in the past but are now enforced. 3. Does the law or any other legal provision oblige the government and/or the Debt Unit to publish comprehensive information about the status of the country's external borrowing policies, debt stock, and sustainability analysis on a regular basis? Yes ---- if yes, how often?_ _' No 4. Does the Government prepare a comprehensive and published annual debt strategy report? V yes, a comprehensive annual report is prepared and published yes, a comprehensive annual report is prepared but not published yes, a comprehensive report is prepared, but on a less frequent basis and not published ____ a report is prepared, but it's not comprehensive nor is it published no, a report is not prepared If a report is published, is it widely available to the public?: Yes. on monthly. guarterlv and annual basis. 5. Where is the Debt Unit located (mark as appropriate)? Ministry of Finance ___Central Bank _____An autonomous body if yes, please specify: -- No single debt unit, debt management functions spread over more than one agency. If yes, specify: Ministry of Finance and Central Bank 106 6. Which of the following best describes the flow of information and debt strategy coordination among government agencies? Very good ____Good _____Needs some improvement Poor 7.a What type of information is being exchanged between the different units dealing with external debt management? Contracts, disbursements, debt service and reports. 7.b. How frequently do such information exchanges occur? Weeklv 8.a Is there an established committee or working group that facilitates the flow of information and debt strategy coordination? _ Yes No 8.b If such a committee or working group exists, what are its the composition and main activities? Composition: Ministra of Finance. Central Bank. Planning Commission, Attorney General Chamber and Sectoral Ministries Activities: Debt management & Coordination. 9. Which is the highest authority responsible for approving the granting of state guarantees on external debt? Ministry of Finance 10. Which government agency handles technical work related to the granting of state guarantees on external debt? Ministry of Finance and Central Bank 11. Who bears the foreign exchange and interest risk when the government on-lends externally borrowed funds? ± Central Government: If onlent in local currency w Secondary borrower: If onlent in foreign currence What are the 34 critical aspects of the current Legal and Institutional framework most in need of improvement/strengthening (list the critical areas and specify the deficiencies and needs for improvement)? Ist area: Debt Coordination Committee Improvements needed; Institute frequent meetings 2nd area: Debt Management Committee Improvements needed: Motivation 107 3rd area: Flow of information Improvements needed; Free flow of information between debt management agencies and from Project Coordinators to debt management agencies. 4th area: Improvements needed:_ COORDINATION WITH MACROECONOMIC POLICIES, DEBT SUSTAINABILITY ANALYSIS L.a Which agencies are responsible for undertaking medium-term balance of payments projections and forecasting the fiscal budget? Please list: (i) Bank of Tanzania (ii) Ministry of Finance I .b. Which of the following best describes the working relationship of these agencies with the Debt Unit? Well-coordinated and with efficient exchange of information _______Good working relationship with some scope for improvement in efficiency Poor working relationship with scope for major improvements Virtually no working relationship nor coordination Comments: 2. How does the coordination between external debt management and macroeconomic policy take place ? (Check as appropriate) - A committee comprising Central Bank, Ministry of Finance and debt management officials is established for articulating export performance, imports requirements, economic growth prospects in relation to the debt profile in medium-term balance of payments and fiscal projections. The Central Bank or Ministry of Finance unilaterally forms perceptions on these projections. _Largely through international financial institution's assistance e.g. IMF and World Bank Other, please specify: 3. How frequently does this coordination take place? w' every month or more frequently 108 every 3 months twice a year once a year very infrequently if ever 4. Is there a mechanism in place to discuss the cash-flow constraints (foreign and local), which debt service payments may have on the budget and the balance of payments (BOP)? ' There is a regular forum or procedures in place to track public sector's debt servicing needs and impact on the budget and the BOP. There is an ad hoc forum or procedures to track public sector's debt servicing needs and impact on the budget and the BOP. _ There is no forum nor procedures to track public sector's debt servicing needs and impact on the budget and the BOP. Comments: 5.a Do the authorities regularly produce a debt portfolio analysis (currency composition, maturity profile, interest rate structure, and associated risks, etc.)? yes, a comprehensive review of debt portfolio is carried out regularly v yes, a comprehensive review of debt portfolio is prepared, but on an ad hoc basis yes, a review of debt portfolio is prepared on an ad hoc basis, but lacks comprehensiveness _no debt portfolio analysis is prepared 5.b Is this analysis published and widely distributed? VYes _No Specify: The last one was produced in 1995 6.a Do the authorities regularly produce a debt sustainability analysis? _Yes, a debt sustainability analysis is produced every year ..Yes, a debt sustainability analysis is produced before major debt renegotiations _No, a debt sustainability analysis is not produced regularly _No, the authorities have never produced a debt sustainability analysis 6.b Is this analysis published and widely distributed? v__Yes No Specify: 7. Does the Public Investment Plan (PIP) cover projects implemented with financing received from foreign sources (including aid and lending)? 109 ± Yes, the PIP incorporates fully aid- and loan-financed projects _Yes, the PIP incorporates some aid- and loan-financed projects _Yes, the PIP incorporates only loan-financed projects _No, the PIP does not incorporate aid- or loan-financed projects If yes, what is the duration of the current Public Investment Plan? One-year Public Investment Plan Multi-year Public Investment Plan w Multi-year Public Investment Plan, updated annually 8. Does the medium-term expenditure framework (MTEF) cover projects implemented with financing received from foreign sources (including aid and lending)? 'S Yes, the MTEF incorporates fully aid- and loan-financed projects _Yes, the MTEF incorporates some aid- and loan-financed projects _Yes, the MTEF incorporates only loan-financed projects _No, the MTEF does not incorporates aid- or loan-financed projects No MTEF is being prepared, only annual budgets If applicable, please specify obstacles in including foreign financed projects in the Public Investment Plan and Medium-term Expenditure Framework: It is difficult to forecast foreign financing Foreign financed vroiects are sometimes selective/biased on location and nature of activity. 9. What are the 3-4 critical areas in the coordination of macroeconomic policy and debt management most in need of improvement/strengthening (list the critical areas and specify the deficiencies and needs for improvement)? I't area: Matching of Debt burden and resources Improvements needed to observe debt sustainabilitv levels. 2nd area: Debt Portfolio Analysis Improvements needed; There should be a rezular debt portfolio analysis (i.e.. annual) 3rd area: Improvements needed:_ 4th area: Improvements needed: III. DEFINNG AND IMPLEMENTING NEW BORROWING POLICY 1. Describe the current official external borrowing policies, in terms of existence of an overall debt management strategy, the categories of debt covered and the limits to new external borrowing: Authority to raise foreign loans by the Government is vested with the Minister for Finance or any person designated by him/her. 110 Limits: Aggregate debt service costs on all outstanding foreign loans during the financial year in which the loan is contracted and the four succeeding financial years should not exceed 15% of the average annual foreign exchange earnings of the preceding three financial years. 2. Please indicate which of the following is the highest government authority whose approval is required for all new external borrowings: _____Parliament approves through legislation - An inter-ministerial committee or equivalent (High Level Debt Policy Body) _____The Minister of Finance, Central Bank Governor or equivalent Other, please specify:_ 3. Is this authority responsible for establishing borrowing policy and limits for debt indicators to guide the level of indebtedness? v_Yes _No Comment: 4. Specify the composition of the body/committee which develops borrowing policy: Ministry of Finance, Bank of Tanzania, Planning Commission and Attornev General Chambers. 5. How could the government's decision-making process for contracting new public foreign currency debt be described? (Please mark as appropriate) < (i) The central govemment raises all external funding required by the public sector and passes /on-lends the funds to public sector agencies. (ii) The central government raises central government funding needs only and public sector agencies raise their own funding with central government guarantee. (iii) The central government raises central govemment funding needs only and public sector agencies raise their own funding without central government guarantee (iv) Other arrangements. Please explain: 6. If the answer to question 5 was (ii), (iii) or (iv), is central government approval required for public agencies (e.g., public enterprises) when they raise external funding? N/A 111 Yes, explicit approval is required for each loan agreement Yes, government approval is granted on no-objections basis No, government approval is not required 7. Is there a Ministry/agency or a committee responsible for the appraisal and monitoring of all projects which receive external funding? _____Yes, there is a designated office solely responsible for project appraisal and monitoring _____Yes, there is a designated office whose functions include project appraisal and monitoring _! No, appraisal and monitoring is done by each project implementation unit If yes, please specify the Ministry/agency or committee responsible for appraising and monitoring the externally-financed projects? 8. Are technical evaluations carried out for new borrowing proposals to analyze their impact on the currency composition, interest rate structure and maturity profile of the overall loan portfolio? _____Yes, each new loan proposal is evaluated against the overall loan portfolio ____Yes, periodic evaluation of new loan proposals is made taking into account the overall loan portfolio _____Yes, but only when there is a perceived need for major changes in the new borrowing policy _! No, evaluation of new borrowing proposals is made independently of existing loan portfolio ____No, there is no evaluation of new borrowing proposals at all Comment: 9. Are there guidelines and limits for non-concessional borrowing? ____Yes, guidelines and limits on non-concessional borrowing by government and public enterprises are strictly enforced ____Yes, guidelines and limits on non-concessional borrowing by government and public enterprises are enforced, but exceptions may be granted on a case-by-case basis ____Yes, guidelines and limits on non-concessional borrowing by government are strictly enforced, but public enterprises are not covered by these ____No, there are only ad hoc limits placed on non-concessional borrowing when needed No, there are no guidelines nor limits on non-concessional borrowing If yes, please indicate how non-concessional borrowing is defined: Concessionality is defined on Loans with Grant Element not less than 35%. 10. What are the 3-4 critical areas regarding contracting of new external borrowing most in need of improvement/strengthening (list the critical areas and specify the deficiencies and needs for improvement)? 112 I" area: Negotiation skills Improvements needed; training 2d area: Monitoring of loan disbursements Improvements needed: Flow of information from donors, Implementing Agencies. and Debt Management Agencies. 3'd area: Effectiveness of signed loans (e.g. ADB/F). Improvements needed; Regular consultations between Project Imnlementation Units (PIU) and Creditors. 4th area: Private debt Improvements needed: To ensure commercial banks submit all copies of loans contracted by private sector. IV. Managing the existing debt stock A. BASIC DEBT MANAGEMENT FUNCTIONS (RECORDING, DISBURSEMENT, PAYMENT) I . Which categories of debts are monitored by the Debt Unit? (mark as appropriate) ' Government direct debt _ Government guaranteed debt v _ .Public enterprise direct Debt " Private non-guaranteed debt # Contingent liabilities 2. Is detailed information on all categories of debt centralized in the Debt Unit, or do different agencies handle details of different categories of debt? (specify agencies accordingly) ±' Government direct debt; Ministry of Finance & Central Bank _ Government guaranteed debt ; Ministry of Finance & Central Bank ± Public Enterprise Direct Debt; Ministry of Finance & Central Bank ± Private non-guaranteed debt; Central Bank _ Contingent liabilities; Ministrv of Finance 3. Does the Debt Unit keep inventory of all loan agreements on govemment and govemment guaranteed borrowings? _ Yes, comprehensive documentation on all government and public sector debt is kept Yes, fairly good documentation on govemment and public sector debt is kept ______Yes, fairly good documentation is available only on government borfowings and loans with government guarantee ______No, a substantial portion of documentation is missing 4. How are loan agreements made available to the Debt Unit? (Mark as appropriate) " the Debt Unit evaluates loan proposals, keeps track of loan negotiations and requests loan agreements after signature from the responsible entity 113 ____No prior evaluations are undertaken, but the responsible entity readily provides copies of loan agreements to Debt Unit after signature ____No prior evaluations are undertaken. Debt Unit becomes aware of debt only after signature and has to go in search of signed loan agreements after negotiations 5. Does the Debt Unit get statements for all transactions on disbursements directly from creditors or locally from project implementing agencies?' Not always. require follow-ups 6. How frequent are delays encountered with the commencement of disbursements after signing loan commitments? _____Very often Quite often ..SOccasionally _ Never 7. Which agencies of government have responsibility for processing disbursement applications to creditors for government and government guaranteed debt? (Mark as appropriate) -.-Project implementing agencies v Ministry of Finance v Government General Account Department Other (specify): 8.a How accurate are forecasts of annual debt servicing obligations provided by the Debt Unit for annual budgetary and foreign reserve management? - Very accurate, requiring only occasional revisions. Especially for public and publiclv guaranteed debts. _____Quite accurate most of the time, requiring some revisions _____Not very accurate, requiring major revisions each year 8.b If debt servicing forecasts are not entirely reliable, what are the main reasons for inaccuracy? Exchange rate fluctuations Information flow problems especially on disbursements. Coverage of Private sector debt is not comnlete 9. How often does the Debt Unit undertake to reconcile loan data with creditors? - Every billing cycle _____At the end of each fiscal or calendar year _____Only when discrepancies are noted in billing statements w' Prior to debt renegotiations 114 _ Almost never 10. Which agencies are involved in effecting debt service payments on Govemment direct debt and Government guaranteed debt (Central Bank, Treasury, Debt Unit, and/or Other) - Direct debt: Ministry of Finance and Central Bank - Govemment Guaranteed Debt: Ministry of Finance, Central Bank and Others 11. Are difficulties encountered with regard to cooperation among agencies, which cause delays in effecting debt service payments? Quite frequently v Sometimes _ Almost never Comment: Ministry of Finance sometimes delays submission of cash cover. 12. Which stage of the debt service process is typical the source of delay? _The Debt Unit does not provide notices on payments due in time @_The Accountant General's Department causes delays in posting payment instructions to the Central Bank/Extemal Finance Department _ Central Bank/Extemal Finance causes delays in remitting funds to creditors 13. How frequent are penalty charges for late payments? _ Very frequent _ Quite frequent " Occasionally _ Never 14. How large are penalty charges on average as a percent of monthly debt service payments? Negligible. 15. What are the 3-4 critical areas in basic debt management most in need of improvement/strengthening (list the critical areas and specify the deficiencies and needs for improvement)? 15' area: Mismatch of forecast and actual debt service Improvements needed: Information flow 2nd area: Capturing of disbursements Improvements; needed: Information flow 3^ area: Improvements needed: 115 4th area: Improvements needed:_ B. DEBT RENEGOTIATIONS 1. Which agencies/groups are responsible for debt renegotiations? The Debt Unit is responsible for most debt renegotiations %. A designated government team is responsible for most debt renegotiations _ A specific government ministry is responsible for most debt renegotiations An ad hoc team is assembled for each debt renegotiation (Please specify) The team comprises members from Ministry of Finance. Central Bank. Planning Commission. The Attorney General's Chambers and the Line Ministry requiring foreign financing. 2. During your recent renegotiations of the following categories of debt, did you receive technical assistance? (yes or no by category) i) Paris club? No Data No Documentation No Strategy ii) Bilateral non-Paris Club? No Data No Documentation No Strategy iii) Commercial debt? No Data No Documentation No Strategy iv) Multilateral ? No Data No Documentation No Strategy 3. Were the outcomes (new financial terms, concessional treatment) of your recent renegotiations of the following categories of debt as anticipated and satisfactory? i) Paris club? Yes No Not applicable Comment:_ ii) Bilateral non-Paris Club? _Yes __ -No Not applicable Comment:_ iii) Commercial debt? %. Yes _No _ Not applicable Comment: Concluded Debt buy back in June 2001 116 iv) Multilateral ? _________Yes No Not applicable Comment: m 4. Are you sufficiently well informed on the latest developments in debt reduction mechanisms at the international level? ± Yes No Specify: (i) HIPC' (ii) Paris Club negotiations. Debt Buy-back Debt swaps and conversion. 5. What are the 34 critical areas in the government's debt renegotiation capacity most in need of improvement/strengthening (list the critical areas and specify the deficiencies and needs for improvement)? I ' area: Non PC renegotiation Improvements needed: International assistance in persuading Non PC creditors is required. 2nd area: Commercial debt renegotiation Improvements needed: International assistance in persuading Commercial creditors to renegotiate. 3rd area: Improvements needed: 4th area: Improvements needed:_ V. HUMAN AND TECHNICAL RESOURCES A. STAFFING AND TRAINING 1. Which of the following best describes the staffing situation in the Debt Unit: _ Adequate number of staff with sufficient training to carry out basic debt management functions and produce quality debt analysis ± Adequate number of staff with sufficient training to carry out basic debt management functions, but with only rudimentary capacity to produce debt analysis _ Inadequate number of staff with sufficient training to carry out even basic debt management functions Please comment on the key areas in which the number of staff or training is insufficient: Risk analysis Scenario analyses 2. How many professional staff in the Debt Unit, and what are their responsibilities? 117 i) Number of professional staff? 14 (e.g. accountants, economists, lawyers, managers, etc.) ii) Their responsibilities (for all professional staff. Add to the list if necessary)? Staff 1: Manages the debt department Staff 2: Supervise Bilateral debt Staff 3: Supervise Multilateral debt Staff 4: Supervise Debt policy and analysis Staff 5: 3-manaze bilateral debt Staff 6: 3-manage multilateral debt Staff 7: 4-deal with policy and analysis Staff 8: Staff 9: Staff 10: Staff I 1: Staff 12: 3. How many staff have been in the unit for: more than one year? 11 more than 3 years? 10 4. Which of the following describes best the office space and equipment used by the Debt Unit? " Both office space and equipment are sufficient to operate efficiently Office space is sufficient, but equipment is outdated or often in shortage Office equipment is sufficient, but office space is inadequate Neither office space nor equipment are sufficient to operate efficiently 5. Do all staff have clear job descriptions and terms of reference? _ Yes, majority of staff have clearly defined job description and terms of reference and assigned to the responsibilities listed therein -Yes, majority of staff have clearly defined job description and terms of reference, but actual assignments and work program is given on an ad hoc basis _No, majority of staff do not have a pre-defined job description or terms of reference Comment: 6. Does the debt unit have a procedures manual to guide staff in their daily work? v Yes, an up-to-date procedures manual exists and is used routinely _ Yes, a procedures manual exists, but is outdated and unused _ No, there is no procedures manual at all 118 7.a How many staff in the debt unit have received training in the following areas (and who provided this training)? a. Loan interpretation: 10 Nr. Staff Provider: MEFMI/COMSEC/DRI/USAID b. Debt recording software: 12 Nr. Staff Provider: MEFMI/COMSEC/DRI/USAID c. Legal aspects of debt mgt: 5 Nr. Staff Provider: MEFMI/COMSEC/DRI/USAID d. Debt renegotiation techniques: 10 Nr. Staff Provider: MEFMI/COMSEC/DRI/USAID e. Portfolio Reviewing: 8 Nr. Staff Provider: MEFMI/COMSEC/DRIWUSAID f. Evaluation of financial proposals: 0 Nr. Staff Provider:_ g. Debt Sustainability Analysis: 5 Nr. Staff Provider: MEFMI/COMSEC/DRIWUSAID h. Debt Strategy Design: 0 Nr. Staff Provider: 7.b On average, what was the impact of the training identified in 7 above? Fairly effective, and improved Debt Unit's operation substantially v Somewhat effective, and improved Debt Unit's operation to some extent Marginally effective, but did not improve much Debt Unit's operation _Not effective at all Comments: more training required 8.a Is there a need for further training (advanced or repeat)? -! Yes advanced _No 8.b Please specify areas where further training is needed in order of priority: a 1st priority e 2nd priority g 3rd priority h 4th priority Nb. The letter given correspond to those areas mentioned in 7 (a) 9. Is the salary scale in the debt unit the same as in the overall civil service? _____Yes Higher _ Lower Please explain: N/A B. COMPUTER TooLs 119 1. Which of the following describes best the computers and software tools used by the Debt Unit? Adequate number of computers and software tools exist to support debt management functions and debt analysis ___Adequate number of computers exist, but software tools are insufficient to support debt management functions and debt analysis There is a shortage of computers, but adequate software tools are available to support debt management functions and debt analysis There is both a shortage of computers and adequate software tools Comment: 2.a Which of the following debt management systems is currently used by your government for maintaining a detailed loan-by-loan database? (Mark as appropriate and specify the version of the software where applicable) @ CS-DRMS Version: 7.2 UNCTAD-DMFAS Version:_ Country's own system " Spreadsheet files (e.g. " Excel, Lotus 1-2-3, Quattro Pro) Other (specify):_ 2.b Which of the following describes the functioning of Debt Unit's debt management system? _ The system functions adequately and fulfills the operational and analytical needs of the Debt Unit v" The system has minor deficiencies, but fiulfills most operational and analytical needs of the Debt Unit _The system has major deficiencies, and fulfills only basic operational and analytical needs of the Debt Unit _The system does not function properly, and basic debt management functions are carried out manually or using temporary computer solutions Comment: 3. Does the debt management system readily produce reports such as: S Debt Service due on monthly, quarterly and yearly basis 120 v Actual Debt Service on monthly, quarterly and yearly basis __ Disbursements (actual and forecast) ' Others (please specify report and frequency) debt stock, currency composition. debt indicators etc 4. Is the debt management system capable of handling information on all categories of debt (public, publicly guaranteed, domestic public, private external, paast al debt)? If not, please specify which categories are excluded and explain why. Contingent liabilities are not recorded due to absence of well-defined terms. 5. Is the system installed in a network environment and thus accessible to all debt management staff and other agencies involved in debt management? _Yes No Comment: accessible to all debt management department staff but not to other agencies involved in debt manaaement. 6. Is the debt software effective in assisting with: -debt rescheduling exercises? "__Yes _ No -evaluating effects of new borrowings? ±__Yes _ No -Debt Sustainability Analysis? ±' Yes No 7. Does the country use Debt Pro, DSM+ or other software for conducting debt sustainability analysis? Specify software: Debt Pro & DSM + Comment: There is inadeguate training in the Debt Pro and absence of direct link between the software (DSM+ & Debt Pro) and the CS-DRMS 9. What are the 3-4 critical elements in human and technical resources for debt management most in need of improvement/strengthening (list the critical areas and specify the deficiencies and needs for improvement)? 15' area: Retaining of staff in Improvements needed: Improve remunerations the Ministry of Finance 2nd area: Insufficient debt Improvements needed: training Management skills 3rd area: Absence of a link between Improvements needed; Improve the link CS-DRMS and Debt Pro 4th area: Improvements needed: 121 VI. ASSISTANCE FROM DONORS AND TECHNICAL AGENCIES Has the Debt Unit received technical assistance or training in debt management / debt analysis from external agencies or consultants in the past? Please summarize briefly the history of external support and areas of debt management covered: Central Bank;-Debt consultant from KPMG brought by COMSEC Training in debt management funded by USAID. COMSEC & DRI Ministry of Finance: Accountant Generals department received support from SIDA. World Bank and DFID in the establishment of Public debt unit. 2. If technical assistance or training described in question I was received, how would you rate their quality and overall impact on debt management? v Very good Good Not so good Poor Specify deficiencies, if any: 3. Do the organizations providing technical assistance also provide regular training support in various debt management functions? _,Yes _ No Comment 4. Has Debt Relief International (DRI) been contacted to assist in building capacity in undertaking debt strategy analysis? '__Yes _ No Comment on assistance requested;,thev have delivered very little compared to others e.g. COMSEC & MEFMI 5. What are the 3-4 most important areas for improvement in technical assistance or training in debt management provided currently or in the past? What do you think can be done differently by external donors and technical agencies to improve the quality of technical assistance or training available to national debt offices? I area: Linkage between Debt Pro & CS-DRMS Improvements needed: Acquisition of a software to link 122 2nd area: Debt Sustainability Analysis skills Improvements needed: training 3rd area: Recruiting and retaining of staff Improvements needed: Financial assistance in the Ministry of Finance 4th area: Debt Management and analyses skills Improvements needed: advanced training in debt management and analysis. 123 ANNEX 7: PUBLIC EXPENDITURE REVIEWS AND BUDGET PROCESS IN TANZANIA AND UGANDA23 Summary Introduction In 1997, both Uganda and Tanzania experimented with a new approach to the Public Expenditure Review. In both countries, the decision was taken by Government the Bank and the donor community to develop the PER as annual exercise, closely integrated with Government's budget cycle and supporting specific reforms in budget process. The basic features of design of the PER were similar. In each case, the PER was divided into two phases, a first phase focused on direct support to government in its formulation of budget strategy, followed by a second phase focusing on external evaluation of the strategy and budgetary systems and performance, carried out in concert with domestic stakeholders. Common objectives of both exercises can be summarized as follows * Opening of the budget process within the executive. * Opening of the budget process to stakeholders beyond the executive. * Integrating donor financing within the budget. * Technical support to both phases of the process. In September 2001, a review of the current budget process in Uganda and Tanzania was carried out, focusing particularly on the innovations associated with the annual PER exercise and donor consultations on the budget framework since 1998 . The aim was to provide a brief assessment of the extent to which the broad objectives of the revised PER process have been met and make some recommendations on further development of budget process, including any specific suggestions for future design of the PER process. To get an overall view of the current process, breadth seemed more critical than depth and apologies are offered for shockingly sketchy coverage of many key areas of the process. The process reformns of 1997/8 provided a broad reference point for the review, although, in both countries, the present budget process is clearly the outcome of a wide range of initiatives by Government, donors and other stakeholders in recent years. Each country began the 1997/8 reforms with different starting conditions and have to some extent followed interestingly different paths in development of their budget process. The main conclusions offered below focus especially on some of the challenges shared in both countries, with only brief reference to major points of difference. The country notes which follow this summary explore country specific issues in more detail, in each case under the headings of the broad objectives listed above. 23 This Annex was prepared by Allister Moon (Lead Economist, ECSPE), who visited Tanzania and Uganda in Septermber 2001. Mr. Moon was one of the principle architects of the new approach to Public Expenditure Review in Uganda and Tanzania in 1997. Since 1998, Mr. Moon has been mainly working on Eastern European countries. 124 Main Conclusions * Institutionalizing a more open budget process. In both countries, the basic features of the PER process have been maintained and significantly developed over successive budget cycles since 1998. Technical support continues (especially in Tanzania) but plays a very different role from the initial 1997/8 budget cycle, largely focused on incremental further reforms while core processes within the executive are fully managed by government staff. External consultation at key points in the cycle is a routine feature in both systems * Contestablity within the executive. In both systems, significant limits exist on the degree to which the budget process allows Cabinet to make informed, contestable choices on strategic allocations. Factors include (i) timing of the budget framework exercise within the cycle (ii) capacity constraints on sector and inter-sectoral analysis (iii) ambiguity in the relation between the budget process and other integrative processes in policy choice (e.g., PRSP) (iv) effects of prior agreements with donors, outside the budget process. The incentive for greater line ministry investment in sector expenditure planning and reporting must rest in part on a belief that Cabinet makes real decisions on budget strategy and sector submissions can make a difference. * Strengthening performance orientation. A high priority for both countries is ensuring a greater focus on budget outputs, both in formulation and especially in reporting and evaluation. In both countries significant progress has been made in improving basic expenditure controls (especially commitment controls) and quarterly expenditure reporting, as well as progress towards implementation of IFMS (further in Tanzania than Uganda). Improved management of inputs should provide the platform for greater focus now on outputs. Strengthening the output focus of sector frameworks should be a high priority for phase 1 technical support where required, while the phase 2 consultations should focus strongly on comparative sector achievements in defining and reporting outputs, feeding back into design of support for the next cycle. The phase 2 consultations should also focus on progress with cross cutting reforms relevant for budget performance: this might be facilitated if the agenda currently registered in donor financing instruments (Uganda PRSC matrix, the PRBS PAF matrix in Tanzania) is included as a component of Government's annual presentation in the phase 2 consultations. * Capacity constraints. In both countries, Government capacity is severely strained by the demands of continued coordination of the process, rising expectations and technical standards on budget formulation and performance reporting and the high transaction costs of consultation. New demands on the process can probably be accommodated only by significant rationalization (see below) * Participation: A wide range of stakeholders uniformly acknowledge and welcome greater consultation in the budget process and better information on budget plans and performance. Opening up has clearly generated sufficient goodwill for continued engagement in the near term and may have irreversibly altered expectations on budget transparency, but participants are beginning to question the impact of consultation. Some stakeholders maintain unrealistic expectations of a substantive decision making role, many are bewildered by the growing array of participatory processes (PRSP, PER, CG, etc.), and face severe capacity constraints on multiple engagement ('death by participation'). For civil society participation, there is a need to (1) clarify a 125 distinction between consultation and preemptive decision making, (2) develop a clearer map of the budget process, other policy coordination processes and the entry points for stakeholder involvement, (3) ensure more effective feedback on civil society inputs, acknowledging positive influence and explaining divergent decisions and performance and (4) encourage civil society to see its primary role in holding Government to account for performance and demanding higher standards of performance reporting, rather than influencing allocations in the immediate annual budget. For the legislature, similar concerns apply but also a need for greater clarity on the respective role of legislature and executive in budget formulation (particularly critical in Uganda), and better linkage established between the PER process and formal processes of budget scrutiny by the legislature Integrating donor financing. The open process of review and consultation on the whole budget provides an important platform for the shift towards general budget support which has continued in both countries (PRSC, PRBS). Linkage and efficiency might be improved by ensuring that the PER consultations are used more broadly as a forum to establish progress on conditions of future budget support. (e.g., see comments above on strengthening performance orientation ). In both countries, significant decisions on donor financed sector programs and projects effectively take place outside the budget process, though reflected eventually in revisions in the MTEF. In Uganda, Finance staff note the increasing constraints imposed in budgetary planning by 'notional earmarking' and the claims of the PAF, even for areas within these categories where performance is known to be poor. In general, the role of donors in decision making processes within the budget is by far the murkiest area of the overall process. This potentially raises significant difficulties for the Government's management of the participation of civil society and the legislature in the budget process. * Rationalizing the budget cycle The issues discussed above point to the need for a comprehensive review of budget process and its links with other policy coordination processes such as the PRSP. The objectives should be (1) gradual rationalization of the process to minimize duplication in consultation across the budget cycle and other major policy coordination processes (2) establish a clear and public map of the annual process, clarifying core decision making processes and key points for consultation and feedback. * Clarifying objectives of annual PER phase 2. Within this broader exercise, there is a particular need to clarify objectives and long term goals of annual PER consultations and the relation to other external review activities. Efficiency in budget process requires rethinking the role of such consultations from the perspective of each major group of stakeholders, ensuring that the process serves multiple objectives as far as possible. If it is to make a sustainable contribution to public sector accountability, the phase 2 review needs to be designed as a public forum which, in the long run, makes sense for domestic stakeholders regardless of the involvement of donors. Effective demand for such accountability may be limited at present: donors have an important role in providing such demand as a temporary substitute and as a catalyst for developing effective domestic demand over the long term. Issues implied for design of the phase 2 review include (1) giving the legislature a more explicit role and making a stronger link between the review and follow up by e.g. parliamentary standing committees, especially on weaknesses in reporting and performance (2) more active 126 use of domestic research capacity in preparatory analytic work for the review (already strong in Tanzania) (3) increasing the focus of the review on assessing sector performance, rather than future allocations and output targets (4) including review and updating of cross cutting public sector reform agenda (5) more active use of the review to expose poorly coordinated donor financed expenditure (6) streamlining the use of the phase 2 review as an information base for donor financing decisions Incorporating new initiatives on PEM. The 1997/8 reforms were based on the judgement that sustainable improvements in public expenditure management are best supported by helping Government to develop its own program, provided that this is located within a process of periodic review and feedback from stakeholders, within and beyond the executive. This process of review should be integrated with Government's own policy coordination and budget processes in a manner in which domestic stakeholder involvement is maximized and external (donor) review objectives are as far as possible subsumed in accountability to domestic stakeholders. This approach was highly complementary to the aims and objectives of the subsequent PRSP initiative, although there are still issues to be resolved in integrating the PRSP (and PRSC) and budget process. Sustaining this approach in future must entail a rigorous filtering of new initiatives, ensuring that new demands on PEM capacity are realistic and that they are met as far as possible by incremental adjustment in existing processes. Further new initiatives with significant demands on PEM capacity have been plentiful in recent years, - for example, CFAA, ROSC, HIPC tracking - and there seems little prospect that the flow will abate in the near future. None of the initiatives cited are satisfactorily integrated within existing processes, although the recent CFAAs in both countries made exceptional efforts in this direction. * Technical support. Experience in both countries confirms the value of designing technical support as an input to the annual process of budget formulation and external review of budget strategy and performance. In both countries, the Bank with other donors has played a valuable role in supporting Government as it opens the budget process, while also coordinating the forum for external review and the analysis underpinning such review. The latter role has been particularly strong in Tanzania where major positive factors have been the broad based management by the PER Working Group, and the extensive use of local consultant capacity. * Future PERs. In both cases, development of the process has depended critically on (1) Bank resident mission staff and PER teams providing an invaluable secretariat function for the process, and (2) resident PER team leaders guiding the process, maintaining a sensitive and pragmatic balance between support for the Ministry of Finance and development of the function of extemal review. Technical support through the annual cycle remains very important, including the task of review required to meet the challenges identified above. The task has moved a long way from a traditional PER, for which the principal output is a Bank report. The PER task is focused primarily on the support to the process rather than report production and should be assessed primarily by the extent to which development of the process addresses the type of challenges discussed above, rather than primarily on the merits of written material produced within the process. A brief annual record of key features of development of the process may be adequate for this purpose, as in the summary report on the last Uganda PER. The points made earlier on *rationalization of the budget process, clarifying the link between policy coordination and budget process 127 and reassessing the objective of PER consultations should have important implications for definition of what are currently PER process tasks, including greater integration with e.g. the CG process, PRSC appraisal and monitoring. UGANDA The 1997/8 Agenda for Reform in Budget Process In the 1997/8 budget cycle, the Public Expenditure Review was designed jointly by the Government and donors to support several key changes in budget process. These changes can be examined under three broad headings. Opening the budget process. Prior to 1997/8, Govemment had established a three year Budget Framework as the primary instrument for defining budget strategy and establishing allocations by ministry in the annual budget. The objective in the 1997/8 budget cycle was to develop a more open process of budget formulation and review, building on the budget framework institutions already developed. This had two major dimensions. First, the objective was to enhance the process of internal consultation, especially in the stages prior to submission of the Budget Framework Paper to Cabinet. This included giving a greater role to line ministries in preparation of the paper, which had hitherto been prepared largely by Finance following bilateral discussions with line ministries. Initial sector contributions to the paper were to be prepared by sector working groups, mostly chaired and staffed from line ministries, though also including representation from the Finance ministry and other ministries with cross cutting responsibilities such as MPS and Local govemment. The introduction of cross sector workshops at key stages in the cycle introduced a multilateral approach to budget consultation which had previously been exclusively bilateral A second objective was to open the process beyond the executive. Greater openness in budget process was not pursued exclusively for its own sake. It was in part a response to well recognized weaknesses in existing budget performance, especially the evidence from budget deviations. Firstly, a more open, contestable process of budget preparation within the executive was seen as a step towards establishing a broader and more stable consensus on budget strategy and hence more likelihood of binding decisions and greater predictability in sector financing. Second, the broader base for budget formnulation and publication and review of the framework in the second phase of the process could be expected to raise the political costs of budget deviation, and improve the context for establishing greater accountability of the executive for budget perfornance. Integrating Donor Financing The aim of external review of the framework by donors and other stakeholders offered not only an instrument for binding the executive, but equally an opportunity for locking in commitments on donor financing over the medium term and encouraging greater integration of donor financed expenditure within budget strategy. A key element of the 1997/8 reforms was the introduction of an extemal review of the budget framework and budget performance as a permanent feature of the budget cycle, taking place after Cabinet resolution of the Budget framework/MTEF and prior to presentation of the detailed annual budget. In making a public commitment on medium term expenditure plans, the Government in this process invited a similar degree of forward commitment by donors. 128 As an institutional frame for scrutiny of the whole budget as a routine feature of the annual cycle, the annual consultations provided a base for donors to switch towards budget support, while also providing reference point for assessing the coherence of project financing with overall budget strategy. The consultation process - phase 2 of the PER, in the terms of the 1997/8 design - can be viewed as an ad hoc adjustment of the basic MTEF model, adjusting for a context of significant external financing. Developed country examples of MTEFs typically assume a decision making process in which Cabinet is empowered to make binding decisions on budgetary allocations. Neither theory nor practice offer clear guidance on the modifications in the MTEF model and related budget institutions required where, as in Uganda, half or more of public expenditure is financed externally. The reforns in the 1997/8 budget cycle can be viewed as a tentative experiment in design of such institutions. Technical Support Design of the 1997/8 PER emphasized a distinction between two phases, the first focused on Government's preparation of the budget framework, the second comprising external review. The PER had an explicit objective to provide technical support during the first phase, including specialist support for the SWGs. The second phase of external review was intended to address donor interest in an objective assessment of the budget framework, but also included an implicit objective to provide a catalyst for development of domestic capacity for budget analysis and feedback outside the executive. A. Opening Up The Budget Process Opening the Budget Process Within the Executive * Perfornance through four annual cycles since FY98 suggests that the basic process of internal consultation on the framework has been fully institutionalized and is fairly robust in spite of caveats below. Budget workshops have helped encourage a more open dialogue on budget strategy and performance (but could be used more to develop line ministry understanding of budget process) Sector working groups have taken on increasing formal responsibilities in sector expenditure planning. * Constraints on SWG capacity are highly evident. Typically groups report a surge of activity following the initial budget workshop, poorly sustained through the remainder of the budget cycle. The more successful groups are generally those where the work program is planned over a longer horizon, integrated with the annual work program of line ministry planning and budget staff. MFPED representation in SWGs is thinly spread, and the demands of coordinating the overall exercise fall very heavily on a handful of staff. * Some SWGs suffer from relatively loose match between the scope of the group and the underlying line ministry responsibilities. The Public Administration group is the clearest example, where the group faces the unenviable task of developing sector expenditure plans across the disparate fields of Foreign Affairs, the Office of the President, Vice President, etc. However, other SWGs have evolved pragmatically to develop a reasonably effective match between scope of the group and line ministry responsibilities: in some cases SWGs have been redefined to get a better match, in the case of the judicial sector, the existence of a SWG seems to have played a significant part in encouraging the constituent sector ministries to take a more 129 coherent sector view. Coordination/overlap between SWGs and line ministry planning and budget staff could be improved in some cases. The stakeholder forum last year pointed to cases where ministers or PSs failed to endorse or radically revised SWG submissions late in the BFP process: it might make sense to build in more formal requirements for ministries to sign off on SWG outputs at key stages. * Both heads of sector working groups and other members (including external participants, see below) raised doubts about the transparency of decisions on sector ceilings. Several participants expressed the sense that the allocations had been somehow decided by MFPED, with little opportunity for any effective argument for change over the planning horizon. The effect of this perception on SWG incentives may be more constraining than capacity limitations. * SWGs have not so far made rapid progress on increasing the output orientation and reporting on output performance in sector framework submissions. The annual PER consultations on the framework could be developed to bring a bit of competitive pressure to bear on SWGs and ministries in this area. Similarly, the valuable innovation of the public budget performance report can be developed further in the same direction. * Handling cross cutting issues : adding the PRSC sub matrix as annex to the budget framework In recent years, the submission of the Budget Framework Paper to Cabinet has fallen later in the budget cycle (planned for March currently, compared to a target date of around Dec/Jan in the mid 90s). Previously, the BFP submission preceded the call circular for preparation of the detailed annual budget, based on parameters defined in the BFP. Delays in Cabinet consideration of BFP often led to issue of the call circular based on the submission rather than the finally approved framework, in which case the relevant parameters remained subject to cabinet approval. Timing of the BFP has always been at risk of slippage and later submission is understandable in view of the greatly expanded scope of the exercise. However, one of the implications of a later date is that it is difficult to make significant changes at this stage, at least for the immediate annual budget, and there is thus a tendency for the submission to seek endorsement of single set of numbers with scope for marginal adjustments, rather than a presentation of strategic choices. The BFP becomes less of a decision making instrument, and more a presentational device. This may also contribute to concerns on the legitimacy of sector ceilings, since the MFPED carries the responsibility for revising ceilings as required from the previous cycle and coordinating much of the budget preparation cycle without the authority deriving from a full review of the framework by Cabinet. A perceived 'legitimacy deficit' in derivation of the sector ceilings could be addressed in several ways. One possibility is to submit proposed indicative ceilings to Cabinet early in the cycle, prior to the main BFP. However, an additional Cabinet submission in an already heavily loaded cycle may not be very attractive in view of constraints on the Cabinet timetable and MFPED capacity. The problem could be diminished by improving the quality and credibility of year 2 and 3 in the framework and highlighting the outer years more in Cabinet consideration of the BFP. However, in a volatile environment, some adjustments in the expected resource envelope are inevitable at the beginning of each cycle, and it is clearly important for MFPED to provide a transparent account of these adjustments and use opportunities such as the initiating budget workshop to ensure that they are clearly understood by budget participants. 130 A BFP submitted earlier in the budget cycle could take on a more explicit role as an analysis of budget options, rather than a single framework proposal inviting Cabinet endorsement. At present, options are presented principally in the form of a list of unfunded priorities, drawing Cabinet's attention to expenditure options considered in framework preparation but not included in the recommended proposal. It makes sense to continue this practice but it may also be desirable to emphasize Cabinet's decision making role by developing more explicitly strategic alternatives in the budget. Past BFPs have been explicitly used in addressing policy dilemmas. For example, in one cycle, MPS developed its own proposals on the future size of the wage bill, included without amendment in the BFP submission but contrasted with an alternative central recommendation from the Finance ministry and presented as alternative framework scenarios to be resolved through Cabinet. Following the sudden shift in policy towards rapid introduction of UPE in the 1996 elections, the BFP submission laid out alternative scenarios on the pace of increasing allocations to education and consequent reductions in other spending. Identifying strategic policy choices within the budget may not be as easy now as in earlier cycles. More areas of expenditure are now locked in to explicit sector commitments, often tied to donor financing, within the PAF , in sector programs. Identifying structured choices at the level of broad strategy is more demanding of MFPED coordination and analytic capacity than developing sector programs within fixed ceilings and simply listing expenditure options excluded within each sector. However, the effect of this may be to convey a strong sense that choices do not exist, or have been taken elsewhere: a perception which may undermine the role of Cabinet and weaken incentives for active participation in budget process from line ministries and other participants. It is to be expected that sector allocations are to some extent locked in by prior policy commitments. Past decisions on UPE, for example, lock up resources over the long term, assuming policies remain consistent over time. As an increasing proportion of expenditure is driven by medium term sector strategies, it should be expected that the area of choice shifts to the margin of the outer years or even beyond the three year frame. Absence of choice within the framework exercise could be interpreted as a positive consequence of achieving a stable consensus on national strategy and a corresponding level of predictability in inter-sectoral balance. This seems a highly optimistic interpretation of the current state of budget strategy. First, the apparent uniformity in the framework coexists with highly divergent views on budget priorities from major constituencies, including the President and legislature. Second, over recent years major reallocations have occurred in the framework: strategic choices have been made, but to a significant and perhaps increasing degree, not through the process of Cabinet review of the budget framework. The fact that strategic policy choices (with budget implications) have been made outside the budget process is not in itself cause for concern. In Uganda, national policy is made through a variety of processes and instruments, including inter-sectoral instruments such as the PEAP, PRSP and the wide and increasing array of institutions for formulating sector strategy. The budget process, narrowly defined, has no privileged role in determining policy: it would be possible to imagine circumstances in which policy is determined and its fiscal implications analyzed through other processes with sufficient discipline that very few additional decisions are required in the budget process. These exercises may contribute to the budgetary task of 131 reconciling policy with available resources (usually incompletely as shown by the 'under- funding' of PEAP objectives). However, the budget process must retain the unique responsibility for achieving a final comprehensive reconciliation between policy and available resources. Preferably, such reconciliation should be achieved in budget formulation ; if not it will be achieved de facto in budget execution. Deciding on the budget framework and annual budget must therefore remain an explicit process of Cabinet choice, regardless of the scope and 'hierarchy' of other policymaking exercises. Opening the budget process beyond the executive Civil Society Participation * There are two main stages of external participation directly in the budget process: first, the exercise of framework preparation, primarily via engagement with sector working groups and second, the annual PER consultations. Broadly, participants welcome the opening of these entry points, but express skepticism about the real impact of their inputs on budget decisions or performance. Other common views were observations that MFPED had a dominating influence on the process but staff were too thinly spread to engage in real debate on options, lack of transparency on derivation of sector ceilings, a sense that key decisions had already been taken in separate agreements with donors, and above all a view that Government needed to give more feedback on the outcome of consultation. A few observations below focus on specific participants * Private sector. PSF welcome opportunity of participation, but suspect that Government is less responsive to formal PSF participation in the open consultation process than traditional direct lobbying outside the process by UMA. The role of the private sector working group is unclear within the process: PSF felt that important budgetary implications of the competitiveness strategy had not been fully factored into the budget. * Uganda Debt Network have had an active role in both sector working groups and PER consultations. Potentially, as an organization modeled on IDASA in South Africa , they could play a valuable role in educating other interest groups in engagement in the budget process and reducing the transactions costs of broader, possibly indirect, participation * EPRC have built up a track record of work on budget issues, diversifying their clients recently with Parliamentary Budget office interest in commissioning budget analysis. This could be developed further in the building up the phase 2 review phase of the process : one of the key contrasts in the Tanzanian PER process is the major role played by domestic research institutions in analytic work underpinning the phase 2 review. * Some participants clearly harbored some unrealistic expectations of joint engagement in budget decisions through their participation. However, external participants in sector groups have been relatively few and apparently fairly modest in any lobbying role. A group of SWG chairs reported that such participation had been positive and without major management problems : it was possible to 'close the doors' where necessary in order to finalize a Government view, and this had been accepted by 132 external participants without formalizing a distinction between consultation and the rest of the SWG work program. UDN share this view (their main concern about 'closed door' discussions related to the formal CG rather than budget consultation), although they felt that there should be a clearer and more uniform policy on sharing information relevant for SWGs between Government and external participants. Over the next one or two cycles, it might be sensible to formalize the terms of external participation, clarifying the consultative role, and including commitments on information sharing. Overall, a good start has been made on encouraging external participation: engagement is still fuelled by the novelty of consultation, but expectations are rising, facing potential disappointment as budget process faces lock in, complications of new role of legislature and unresolved issues in role of donors (see further below) Role of the Legislature The recent Budget Act signals a strong intention by Parliament to play a more active role in budget process. To the extent that this implies earlier consultation on the budget framework and annual budget proposals, this could be very positive for consensus building (and would, incidentally, be consistent with the concerns discussed above on timing of the submission of the BFP). However, expectations differ widely on the appropriate role of the legislature. The Budget Act defines a process which would be consistent with the US model of relations between the executive and legislature, in which it is possible for the legislature to set aside the executive's budget proposal and fundamentally rewrite the budget. It is evident that at least some members of the legislature envisage just such a role, although this seems to be inconsistent both with the constitutional provision restricting the legislature from increasing expenditure proposed by the executive, and many other features of the budgetary system broadly derived from the Westminster model. The US model is not likely to be workable in the context of Uganda While the respective merits of the two systems can be debated, a system which mixes incompatible features from both will be unambiguously worse than either. There is a serious risk that parallel budget processes will develop, with uncertainty regarding the primary forum for budgetary choice and adverse consequences for budget predictability. The Public Finance Bill currently in preparation could play a major role in resolving some of these ambiguities, and might usefully incorporate appropriate provisions of the Budget Act within a unified budget law. Given some evident uncertainty on both sides on the practical implications of the Budget Act, the current budget cycle may be particularly important in establishing a de facto interpretation of the new legislation. MFPED has defined a revised budget timetable accommodating with some compromise the consultation requirements of the Act. Future practice may well depend on how such consultation is handled and whether the Budget Committee and other members sense that this provides an effective means of informal influence on budget choice. MFPED are clearly nervous about the process and in particular anticipate adverse reactions as members become more aware of constraints on budgetary choice arising from commitments with donors. The Act also established the parliament Budget Office. The office is headed by Moses Bekabye, a former MFPED economist, and should soon have a professional staff of about eight. The office can play a very important role in enhancing understanding of the budget process and constraints on budgetary choice. It will also be very important that the budget 133 office helps the legislature to develop its oversight role, stimulating a demand for realistic increments in reporting on budget performance, including better coverage of outputs and outcomes and effective use of the powers of parliamentary committees to hold ministries accountable. Given the considerable inexperience of many members of the legislature on budget issues, the budget office could have a significant influence on forming the practical agenda of the budgetary committees. Mr Bekabye asked about possible Bank assistance with training, and arranged to follow up with the Bank and DFID shortly. The new requirements of the legislature in budget process will be imposed on a budget timetable which already imposes a heavy burden on existing capacity in line ministries and especially MFPED. High priority should be given to minimizing overlap with other consultation demands. From this perspective, it will be important to explore how the PER consultations can be useful to the legislature. MFPED staff expressed a view that the PER consultations could extend to two days, and include more detailed sectoral discussions. Over the medium term, it will be important to develop such consultations as a public annual event designed primarily to meet the interests of civil society and perhaps particularly the legislature, and only secondarily the concerns of donors. Appropriately designed, such a forum could introduce competitive pressure across sectors to meet rising standards of reporting on performance, and would be backed by the powers of the legislature to follow up bilaterally with sessional committee discussions with individual ministries More generally, the budgetary system is likely to function more effectively if the legislature views its principal sphere of influence as focused on the oversight role, holding government accountable for performance, and a longer term view of budget strategy. Its influence over budget allocations and broad strategy should be exercised through consultation on the outer years of the framework, rather than through conflict with the executive over the immediate annual budget. The scope for influence over outcomes is arguably far greater for any given sector at the intensive margin (by pressing for better performance reporting and more effective use of a given resource envelope) rather than the extensive margin (fighting for additional resources within a tightly constrained aggregate). The likelihood of the legislature coming round to this view is probably low in any case, but will perhaps be further diminished as their closer engagement in the budget process exposes them more directly to the high degree of donor influence over budget allocations throughout the framework horizon. B. Integrating Donor Financing The development of the PRSC and a move by donors towards general budget support clearly increases the need for the opening of budget process supported by the PER. It also increases the possibility that donors can reinforce such reforms, rather than dilute them by uncoordinated support for projects or sector programs. The annual process allows Government to use the MTEF as a core reference point for coordination of donor financing. The combination of the CG and PER consultations in the last budget cycle may have helped to rationalize the process while also making a clearer connection between donor financing and the budget framework. MFPED staff suggested this could be further developed by allowing longer (2,3 days) for the sector PER consultations. 134 Several concerns were expressed about the role of donors in the budget process. * Sector programs are still developed to an advanced stage outside the budget process with inadequate reference to budget constraints, generating pressure to accommodate prior agreements with donors in subsequent development of the framework * Budget support is often provided apparently as general support but with an array of measures designed to influence allocations and demonstrate additionality. MFPED cited funding in the justice sector which operated through a holding account whereby funds would be returned to the donors if not utilized. The concept of 'notional earmarking' is frequently employed, which seems to express strong informal pressure from donors for particular allocations, without any transparent and predictable implications which can be assessed by legitimate budget decision makers. * The PAF continues to operate as a shadow earmarking arrangement. Some MFPED staff expressed concern that the PAF provides blanket protection and assured increases for expenditure with a nominal association with poverty reduction, regardless of inadequacies in execution and performance. attention is focused by donors on overall increases in the PAF. The claim has been made that the PAF has been successful in ensuring greater allocations to poverty related expenditure. If true, this raises important questions about sustainability of the reallocations in the budget and the transparency of the decision making process. The movement towards greater use of budget support, use of the budget framework as a coordinating instrument and the array of related budget reforms in recent years have effectively exposed more clearly ambiguities in the underlying model for decision making in the budgetary process. Opening the budget process to civil society and the legislature also exposes tensions in the decision making process, as discussed above. In these two cases, as discussed above, there are some steps which could be taken to clarify the respective roles in budgetary decisions, if necessary and given sufficient demand for a more transparent process. In the case of the role of donors, the ambiguities may be more intractable. To explore these ambiguities further it might be helpful to consider two polar cases as alternative models of donor financing of the whole budget, where the decision making process could be made fully transparent. First, donors might have a seat in Cabinet , with influence on budgetary decisions allocated strictly in accordance with the proportion of finance contributed. At the other extreme, in pure financier mode, individual donors might' specify precisely what financing would be available under an array of alternative assumptions on policy and performance, leaving Cabinet to make an entirely independent decision subject to the identified constraints. Neither case is at all realistic, in that the first is politically anathema, while the information requirements in the second case are unimaginable: donors have difficulty specifying what they will finance given unchanged policy and average performance, without multiplying hypothetical cases. These two polar cases are unrealistic but not perhaps quite irrelevant. The first case may be politically unthinkable, but it is not so very different in practical consequences from more familiar practices of donor financing of projects and sector programs. For example, it may happen that commitments on sector financing are decided jointly with donors through joint sector reviews or other processes outside Cabinet and the budget process, and then entered into the medium term framework as' prior obligations. Cabinet may formally endorse the 135 framework, but the net effect of this approach may be that its influence on the decision is no greater and conceivably may be less than if donors participated directly in Cabinet decisions, while the overall transparency of the decision making process is decidedly inferior. Some ambiguity is inevitable in the decision making model underlying donors engagement in the budget process. For many donors, ambiguity may seem desirable as a means of combining significant influence on budget allocations without formal decision making responsibility. This approach may not be compatible with the objective of developing clearer accountability of government to domestic stakeholders. What seems to be increasingly evident in the development of the Uganda budgetary system is the risk that lack of transparency in the role of donors in budgetary decision making may also undermine the transparency of other aspects of the system, particularly the external participation which is critical for long term sustainability of budgetary reform. C. Technical support to budget strategy formulation and external review The 1997/98 PER included substantial technical support to the SWGs. This function of direct support to SWGs through the PER has not been retained. There are good reasons for this development. The large multi-sector PER mission in 1997/8 was arguably necessary to kick start the process, particularly to support the budget workshops and to give SWGs technical guidance in taking on their new role in the process. While useful for this purpose, a large PER mission is not an ideal vehicle for support, which is better provided over a longer horizon by periodic support by specialists already working in the sectors. Such support may be provided by various donors and, given the institutionalization of the SWG role, there is no particular reason for this to be delivered under a PER umbrella; Having said this, there may be areas where technical support in phase I of the PER shculd be retained and enhanced. First, the PER might provide selective support to individual sectors where the SWG faces particular difficulty or is attempting innovations in its sector framework exercise (e.g., integration of pay reform in sector frameworks). The annual PER task seems to have a useful role as coordinator of external support to the phase I work program and meeting gaps in such support. Second, in view of MFPED concerns on capacity, it might be helpful to think about technical support to MFPED in enhancing the analysis of strategic choices in the framework exercise. This would not aim to make recommendations on inter-sectoral choice, but would assist MFPED in developing from the sector BFP submissions a framework for structured choice(perhaps beginning from a joint review of the extent to which such choice is constrained by existing donor commitments) coverage on budget management issues /cross cutting agenda (PER as direct action on budget management, but still a need for periodic review of overall system - note monitoring process through PRSC cross cutting matrix) The external review /Phase 2 PER is perhaps the least developed feature of the cycle, compared to other aspects of the reforms envisaged in 1997/8 (possibly because it was also the most vaguely conceptualized element of the reform). Combining the CG in May this year with the annual consultations on the PER/budget framework/budget performance was a positive move which should improve efficiency of the cycle for both Government and donors, and raise the profile of annual consultations. However, the linkage of the annual consultations with other budget institutions still seems to be weak. (See points above on civil society participation and the legislature, but also note that draft vol. 3 of the PEAP barely recognizes the annual consultations in what is in other respects an extensive discussion of institutions for budgetary planning and donor coordination). The annual consultations should over time be 136 developed as a feedback mechanism for civil society which would have a key role in the system regardless of the presence of donors. The consultations have an important function with respect to clarification of donor financing and donor review of the framework, but they should be designed to maximize the role of domestic stakeholders, with a view to gradual retreat from the present donors role. Long term domestic responsibility for the consultation process should be developed gradually. Joint leadership of the process by Government and donors may be inevitable in the near term, but the donor role should eventually be taken over by domestic stakeholders, perhaps formally led from the legislature or a coalition of domestic stakeholders ( cf. the Tanzania PER working group). Another relevant feature of the Tanzania process is the extensive subcontracting of domestic capacity to carry out functions of review of budget strategy and performance. UDN already has a role in this process and prepared papers for the FY01 process, something they regarded as a positive learning experience. EPRC could develop a major role (as ERSF and REPOA in Tanzania) in view of their growing experience of research in the field of budget management and performance. Tanzania 1997/8 Agenda Whereas Uganda began the 1997/8 reforms in budget process with several years experience of an annual budget framework/MTEF exercise as a core feature of its budgetary system, in Tanzania the process of developing an MTEF was one of the objectives of the reform. The previous Rolling Plan and Forward Budget had been focused on the development budget and was poorly linked with the budget process. Instead the 1997/8 focused on development of the annual Budget Guidelines , a paper normally submitted to cabinet prior to preparation of the annual budget. The objective of the PER process was to assist the Govemment in enhancing the sectoral coverage of the guidelines, extending the analysis to a three year horizon and developing the BGP as a basis for a Cabinet decision on broad budget strategy and the development of a medium term expenditure framework (an equivalent ole to the budget framework Paper in the Uganda system). During the second phase of the PER, the aim was to support a public review of sector expenditure frameworks and overall budget strategy and performance, while also assessing the coherence of donor financing within the budget. As in the Uganda exercise, the aim was to use the second phase as a means of validating the In Tanzania, significant progress had been made on stabilization by 1998, but institutions for strategic allocation were more limited than in Uganda where the medium temn budget framework had been developed over the previous several years. For better or worse, Tanzania took on the twin challenge of opening out the budget process while also attempting to build within the executive basic institutions for strategic allocation through the MTEF and related processes. It is not east to see yet whether this alternative sequence has been vindicated by experience, although the indications are that early opening of the process may have accelerated reform. However, even at this stage it is possible to see some connections between this history and present practice, as discussed further below. The different pattern of external participation between the two countries, the relative role of MOF vis a vis line 137 ministries, and a degree of blurring between responsibilities of the executive and external review in Tanzania may in part be understandable responses to this context. Opening the Budget Process Within the Executive As in Uganda, the process reforms initiated in 1997/8 have been retained as permanent features of the budget cycle. The 1997/8 cycle established a more ambitious role of the Budget Guidelines paper, engaged line ministries in pilot sectors in an initial framework exercise and established the phase 2 review as a process of consultation with external stakeholders. However, the subsequent cycles achieved further consolidation of the process, including development of a publicly available MTEF, preparation of sector expenditure frameworks with extensive support from local consultants and a considerable broadening of the stakeholder engagement in the phase 2 review. While the use of consultants (mainly local) remains high and the core team in the ministry of Finance is very heavily loaded, continuation of the basic process over several cycles has demonstrated its sustainability. While the process has entailed greater engagement of line ministries in the budget process, the early stages of the process are directed fairly tightly from the Finance ministry. Line ministries are represented in the Budget Guidelines committee which has responsibility for preparing the Budget Guidelines paper, and line ministries are responsible for preparing analysis of sector requirements which provides a background to the paper. The sector 'requirements' are noted in the subsequent BGP submission to Cabinet alongside a proposed allocation (usually much lower), but the implications of rejecting full funding of the alleged requirements are not spelled out, nor is the paper developed in way to highlight for Cabinet the implications of competing alternative allocations within a given overall resource envelope. Other information necessary for a coherent Cabinet decision on strategic allocations is also missing at the BGP stage : updating a comprehensive picture on donor financing (including projects) is only attempted at the later MTEF stage, while the link between allocations and performance targets is only developed in detail after the BGP, as the agreed allocations are linked to specific performance agreements. Line, ministries seem to be more actively engaged in the, process of developing sector expenditure frameworks later in the cycle, based on the ceilings approved by Cabinet, though the work of sector groups. However, this stage in the cycle is too late to challenge overall allocations, especially for the initial year of the framework. Developing a more contestable process of strategic allocation would probably entail advancing some elements of the current budget cycle to the BGP exercise. This might entail the following: * Replacement of the sector requirements assessment with earlier development of sector expenditure framework proposals prior to BGP submission, including comprehensive estimates of donor financing and links to output and outcomes * This would probably require earlier and more active role of sector groups. Membership of the groups would need to include both relevant line ministries and 138 central ministry representation to handle issues such as pay, performance targeting, local expenditure * Development of the BGP to provide more explicit instrument for choice on strategic allocations * Inter-sectoral budget workshops in the BGP phase of the cycle might be helpful as a means of assisting sectors/line ministries to comply with more demanding requirements for sector submissions, helping MOF to identify key inter-sectoral trade offs for Cabinet decision, and offering participants broad education on the -overall process and current resource constraints * Integrating pay reform Local government. Development of a more open budget process at district level is important for improving accountability for service delivery in key sectors. Uganda has made further advances in this area with the development of MTEFs at district level, albeit of variable quality. In Tanzania, similar developments are discussed , but in the absence of comparable political commitment to decentralization it is not obvious that a substantial investment in local level expenditure planning and consultation will be sustainable. In Uganda, improvements in expenditure planning and local level consultative processes followed slowly in the wake of firstly, an insistent and high level political commitment to decentralization, and secondly, a legal framework in the constitution and specific legislation defining revenue and expenditure assignments and the basic system of intergovernmental transfers. In the absence of equivalent clarity on the long termn framework for local level budgeting, it might make sense to develop a clearer consensus before attempting to establish district PERJMTEF processes or at least proceeding tentatively with pilot exercises in a few districts as a first step. Beyond the Executive Participation. Tanzania has developed a distinctive pattern of external participation in budget process. In Uganda, a higher degree of external participation occurs in budget formulation, through the multi-sector budget workshops and through broad participation in sector working groups. In Tanzania, direct external engagement during budget formulation is more restricted, although there is greater use of domestic consultants in this phase. However, the external review phase has been more strongly developed and has encouraged enthusiastic participation from the legislature and civil society. A key role in development of external participation and the phase 2 review has been played by the PER working group. Chaired by the Ministry of Finance, the group constitutes a representative (and large - about 40 members including government, donors, academics and other civil society) body of stakeholders, overseeing design of the process and the annual work program of analytic work supporting both the review phase and the earlier phase of framework development by the executive. The group appears to play a critical role not only in managing support to the process, but also as a forum for critical reflection on the process. (The latter function might be supplemented by- holding something along the lines of the Stakeholder Forum reviewing the budget process in Uganda last year) Several questions arise .In the longer term, it might be better to have more explicit separation between the functions of the executive and external review. The 'PER process' is in Tanzania 139 understood to refer to the whole process of both framework preparation and review. Until the current budget cycle, the document presenting the MTEF was included as part of the PER documentation, jointly produced by the Government and World Bank. Close collaboration between MOF, the Bank and other donors has clearly been highly productive in management of the process and should be maintained, but in the long term a greater segregation between executive and external review functions and documentation would be desirable for maintaining the independence and objectivity of review. In the long term, leadership of the external review function needs to shift more explicitly to domestic stakeholders - possibly headed from the legislature - along with the secretariat role currently provided by the Bank resident mission, but the present hybrid arrangement of a loose coalition of donors and domestic constituencies represents a pragmatic transitional solution within which domestic stakeholders can gradually develop their role. From discussions with civil society participants and observing their role in CG consultations it is evident that such participation is enthusiastic, generally constructive and quite broadly based. At their best, sector discussions during the CG illustrated a powerful avenue for alerting the executive to public concerns and a public platform for some incisive questions on Government performance. However, it is also clear that civil society groups are puzzled by the array of different participatory processes opening up and unclear on the relation between processes such as the CG, PRSP and PER. Some NGOs talk of overload on their capacity to participate ('death by participation'), while there is also a suspicion that consultation is a means of appeasing interest groups with an appearance of engagement, while the actual decisionmaking process takes place elsewhere. Maintaining credibility of the process is going to be an important challenge, especially as the novelty of a more open process begins to fade. First, there are some difficulties arising from the timing of consultation, very shortly before the annual budget is announced, which perhaps generates an unrealistic expectation that participants views will be reflected in immediate changes. It might be helpful to think about how the consultation cycle works, from a participant perspective. In both countries, external participants voiced concem that the forum had been established for expressing their view on past performance and forward plans, but little was heard back from Government in response. In the Uganda cycle, external participants have some role in the earlier stage of framework preparation , through the sector working groups: the later stage of consultation on the final framework approved by Cabinet in the second phase of the PER could helpfully include some feedback from the Govemment side on how earlier representations were taken into account. In the present Tanzania cycle, there is really one focal point for external consultation in the PER rather than two, so.the same two stage process would not apply exactly, but there is a similar need for more explicit recognition of inputs from participants and identification of follow up action. The legislature express positive views on the more open budget process but are keen to be involved in discussion of the budget earlier in the cycle. The ambiguity regarding the respective role of the legislature and executive is much less acute than in Uganda. Beyond any process of consultation and debate on the budget, the constitution makes it clear that failure to approve the executive's budget provides grourids for the president to dissolve Parliament. The chairman of the Finance and Economics committee standing committee responsible for the budget. 140 Integrating Donor Financing The development of an open process of consultation on the whole budget through the PER is highly complementary to the shift towards general budget support, managed through the PRBS. These developmnents are mutually supportive and the design of further steps on both the process and management of future budget support (including the planned PRSC) needs to be closely coordinated. Design of the PRSC should be greatly helped by the fact that the PRBS is already established as a multi donor facility. The PRSC should as far as possible use the existing framework established for the PRBS, with minimal additional procedures or conditions specific to the PRSC. The PRBS has made a significant contribution to reducing the transactions cost of aid and the efficiency of budget process by consolidating procedures and minimizing specific individual donor conditions for support. The PRSC should adopt a similar approach. In this spirit, a key task for appraisal of the PRSC should be review (with other PRBS donors) of the existing budget process and the parallel processes of policy coordination such as the PRSP. The aim should be further streamlining of the process, rationalizing the consultative arrangements and ensuring an appropriate cycle for appraisal and disbursement of PRBS Issues to explore might include * better integration of the PRBS with the PER process, including more focused use of the annual PER consultations in identifying and establishing compliance with conditions of medium term budget support * consolidating main elements of the PAF matrix in Govemment's progran documents (possibly as an attachment to the MTEF) annually updated by the Government, with performance and future benchmarks reviewed annually in the PER consultations. * coordination between the framework for the PRBS and the sector programs. Sector program financing appears to be far less coordinated with Government's budget process than the PRBS. Similar concerns apply as discussed above in relation to Uganda regarding lack of transparency in the decision making process on sector program financing and the risk that such decisions effectively by pass Cabinet's role in inter-sectoral allocation through the budget. Technical Support The role of technical support has been more extensive than in the Uganda process, and has continued to play a significant role in both phases of the PER. In part, this is understandable in view of the fact that in 1997/8 Tanzania had no functioning MTEF and required significant input at sector level to develop the process. However, it also reflects several other distinctive and very positive features of the Tanzania system. First, the PER working group has played a major role in identifying each year a program of support, based on a broad based perspective on system needs. Following the 1997/8 cycle, particular attention has been given to expanding the use of local research capacity (especially ESRF, REPOA), providing inputs as required through the budget cycle, in preparation of the BGP and MTEF and also in sector review work in phase 2. 141 Design of the support program may have been overly constrained by the annual PER cycle, beginning from initial programming in August or September. Occasionally, studies designed as inputs to the phase I exercise have taken longer than planned, leading to delays in BGP preparation and constriction of the remainder of the budget cycle. In practice, what is needed is a multi year rolling program of support, allowing sufficient lead time on technical support to fit the budget cycle. The proposal for pooling financing of technical support may also help in streamlining the process, though it might make sense to test this assumption empirically by moving gradually in this direction. Development of the process has depended critically on (1) Bank resident mission staff and PER teams providing an invaluable secretariat function for the process, and (2)the PER team leader guiding the process, maintaining a sensitive and pragmatic balance between support for the Ministry of Finance and development of the function of external review. Technical support through the annual cycle remains very important, including the task of review required to meet the challenges identified above. The task has moved a long way from a traditional PER, for which the principal output is a Bank report. The PER task is focused primarily on the support to the process rather than report production and should be assessed primarily by the extent to which development of the process addresses the type of challenges discussed above, rather than primarily on the merits of written material produced within the process. A brief annual record of key features of development of the process may be adequate for this purpose, as in the summary report on the last Uganda PER. 142 ANNEX 8: PER FY02 - STATUS OF IMPLEMENTATION OFTHE WORK PROGRAM (MAY 29,2002) S/N ACTIVITIES STATUS CONSULTANT(S) 01 Extending coverage of donor (I) MOF made a presentation on Extemal Finance during the PER consultative meeting focusing on: (i) The description EFD - MOF resources into the budget of the exercise and effort done this year in integrating donor finances into the budget, and (ii) What the results of this work have been (iii) The envisaged next steps. (2) EFD - MOF has been contacting donors to fimm-up numbers on project support to go into the. next budget. 02 PER Education Final report submitted Faculty of Education - UDSM 03 PER Health Final report submitted IHSD - UK. 04 PER Water Final Report Submitted ERB-UDSM 05 PER Justice System Final report submitted ERB - UDSM. 06 PER Roads Separate reports submitted for Trunk & Regional Roads and for District Roads. WG Chairman to follow up with the two MOW & PORALG PSs (MOW & PORALG) to see whether the two reports can be integrated. . 07 PER Agriculture Draft report submitted and circulated to the WG on April 25, 2002. Comments made. OPM /Agriculture Sector 08 PER Lands Final report submitted DOE - UDSM 09 HIV/AIDS TACAIDS has requested UNAIDS's help to finance work to assess the extent to which sectors addressed HIV/AIDS TACAIDS _ ~~~~~~~~~~~~issues in their PERs._ 10 PER PORALG Final Report submitted ERB - UDSM II Projection of the wage bill & Consultant submitted the wage bill study report to CSD. Report reviewed/discussed at the 7a Macro Group meeting. Mr. T. Valentine implicatiofis for the pay reform Findings of the report were taken into account in preparing the BG. The revised repon was discussed in a stakeholders workshop on April 4,2002 12 Detailed Analysis of the Revenue WG agreed that the work be done by Mokoro/REPIM as a more strategic piece of work to inform development of tax Mokoro/REPIM Effort policy and future budgets. The consultant proposed to do the work starting July 2002. The WG emphasized the need to harmonize this study with IMF's envisaged study on VAT, which will be undertaken during the mid-September mission Mokoro should thus tune their time to coincide with this mission. 13 Iracking of Resource Allocations & MOF has invited bids for the provision of consultancy services (Daily News April 9, 2002). Expressions of interest have (?) livery of Specific Inputs been received from bidders. Evaluation will follow, after which those selected will be required to submit their work _ proposals. Noted that the work has been delayed but it will provide a good starting point for the next MTEF. 14 Extemal Evaluation of Budget PER FY02 main mission held wrap-up meeting with GOT Dec.20, 2001. Full Aide memoire finalized. WB decision WB + renmark, Norway, Performance meeting done on June 4, 2002 via VC link. Production of WB/WG (this) report on External Evaluation is next. Netherlands, Ireland Aid, UNDP and DFID 15 MOF conducted training by zone (October 8-12 & 22-26, and October 29 - Nov. 2). MTEF training manual and report MOF MTEF Training for RS and 'LA circulated. MTEF training to the LAs will be undertaken during "off-season", since it has to be done by the budget Officials officers at MOF. 16 Local Govemment Indebtedness Final report submitted end-March, 2002 and circulated to the WG REPOA 17 SENy FY01 Report (Main report Draft synthesis submitted by Dr. Rutasitara and circulated for comments. WG satisfied by the report. Mr. Masoud Dr Rutasitara - DOE, UDSM / MTEF Volume, Cartoon Version) 'Kipanya' given a copy of the draft and requested to submit work plan and budget. Agreed that same consultants work on Masoud 'Kipanya'JMOF the synthesis for PER FY02 report, based on WB report and proceedings of consultative meeting. MOF yet to compile FY01 MTEF volume. I8 PER FY02 Consultative Meeting Held May 9-10, 2002. MOF/PER WG & PER Consultative eeting Secretariats 143 ANNEX 9: AGENDA OF THE PER CONSULTATIVE MEETING, MAY 9-10, 2002 PUBLIC EXPENDITURE REVIEW CONSULTATIVE MEETING KARIMJEE HALL, DAR ES SALAAM MAY 9 - 10. 2002 DAY 1: PROGRAM MORNING SESSION TIME EVENT SPEAKER / CHAIRW FACILITATOR 08.00 -9.00 Registration of Participants Secretariat 09.00- 09.15 * Welcoming note Mr. Peter Ngumbuilu - PS Ministry of Finance 09.15 - 09.45 * Official Opening Hon. Basil Mramba - Minister for Finance 09.45 - 10.00 * Vote of Thanks Mr. Frederick Kilby - Sector Manager AFIP2 WB 10.00 - 10.20 Tea n Coffee Break Management - Karimiee Hall 10.20-10.40 * Review of recent Macroeconomic Prof. J. Doriye - PS, President's OMfice Planning& Performance Privatization 10.40- 10.55 Commentary Mr. Ali Abdi - Resident Representative IMF 10.55- 11.45 General Discussion Mr. Peter Ngumbullu / Mr. Frederick Kilby 11.45- 12.15 * External Evaluation of Fiscal Mr. Robert Utz - World Bank Performance & Systemic Lssues Mr. Benno Ndulu - World Bank 12.15-12.30 Commentary Mr. Peniel Lyimo - Ministry of Finance 12.30-13.30 General Discussion Mr. Peter Ngumbullu / Mr. Frederick Kilby 13.30-14.30 LUNCH Management Karimiee Hall -___._ AFTERNOON SESSION 14.30-15.30 Economic Governance & Accountability Mr. Thomas Kiama - Controller & Auditor General Ms. Blandina Nyoni - Accountant General Dr. Kihiyo - PCB 15.30- 16.00 Commentary Piet van Heesenwich, DFID & Transparency International 16.00-16.30 Tea / Coffee Break Management - Karimjee Hall 16.30-17.30 General Discussion Mr. Peter Ngumbullu / Mr. Frederick Kilby 18.00 Reception Ministry of Finance 144 PUBLIC EXPENDITURE REVIEW CONSULTATIVE MEETING KARIMJEE HALL, DAR ES SALAAM MAY 9 -10.2002 DAY 2: PROGRAM MORNING SESSION TIME EVENT CHAIR / SPEAKER / FACILITATOR 08.00 - 09.00 Registration of Partildpants Secretariat 09.00 - 0 930 PRSP: Lessons from one and a half years of Mr. Gray Mgonja - DPS, Ministry of Finance Implementation 09.30- 10.00 Commentary Prof. Joseph Semboja - REPOA Civil Society Representative (C Mwakasege - TASOET) Gender Macro-group 10.00- 11.00 General Discussion Mr. Peter Ngumbuilu / Chair European Union 11.00 -11.20 Tea / Coffee Break Management - Karimjee Hall 11.20- 11.50 Cross-sector MTEF and Fiscal Frame Mr. Penlel Lylmo - DPS, Ministry of Finance Mr. R. Khii ah - Commissioner PAD, Ministry of Finance 11.50-12.10 Commentary Mr. All Abdi - IMF Resident Representative Ms. Jytte Laursen - Danish Embassy 12.10-13.00 General Discussion Mr. Peter Ngumbullu / Chair European Union 13.00- 14.30 LUNCH Management - Karimjee Hall AFTERNOON SESSION 14.30-15.00 External Financlng FY03-FYO5 Mr. Prosper Mbena -Commissioner EFD, MOF 15.00-15.20 Commentary Ms. Fiona Shera - DFID Rev. Dr. Fidon Mwombeki - TCDD 15.20-16.15 General Discussion Ambassadors / Donor Representatives 16.15-16.30 Tea / Coffee Break Management - Karimjee Hall 16.30-17.00 Summary of Emierging Major Issues Chair, European Union 1730 - 18.00 Closing Mr. Peter Ngumbullu - Permanent Secretry, MOF 145 DATA ANNEX 146 Table 1: Tanzania Macroeconomic Indicators Indicator Unit 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001* Population/2 Millions 24.6 25.3 26.0 26.7 27.5 28.3 29.1 30.0 30.9 31.8 32.7 Per capita Income/2 US$ 180.7 167.0 149.2 156.2 176.9 210.3 235.6 220.0 240.0 270.0 280.0 GDP Growth/2 % 2.8 1.8 0.4 1.4 3.6 4.2 3.3 4.0 4.7 4.9 5.6 GrossDomestic (asa% 14.4 19.1 20.4 19.7 9.1 10.7 7.7 5.0 5.3 11.4 10.8 Savings/2 of GDP) GrossInvestments/2 (asa% 26.3 27.2 25.1 24.6 19.8 16.6 14.9 16.2 15.5 17.6 17.0 of GDP) Inflation/2 % 28.7 21.8 24.0 33.5 27.4 21.0 16.1 12.9 7.8 6.0 5.2 Exchange Rate/2 TZS/US$ 222.6 301.9 414.5 509.6 574 8 580.0 612.1 664.7 744.8 800.4 917.4 External Sector Exports-Goods& Miu. US$ 482 555.6 747.9 937.6 1265.8 1300.9 1246.4 1141.5 1189.7 1325.8 1488.8 Services/2 Importss-Goods& Mil. US$ 1336.3 1438.5 1983.6 1812.6 2140.3 2028.5 1948.2 2360.1 2241.2 2094.4 2301.2 Services/2 I Current Account Mil. US$ -736.1 -708.1 - -711.0 -646.4 -461.2 -558.6 -998.1 -793.4 -382.0 -413.5 Balance/2 1022.0 Balance of Payments/2 Mil. US$ -260.0 -407.9 -736.7 -461.3 -382.0 -231.2 -633.4 -646.0 -372.7 -35.5 55.0 Foreign Reserves/2 Mil. USS 301.2 458.5 271.9 431.8 270.9 441.1 623.1 599.0 775.6 974.0 1152.5 External Debt/i Bil. 6.0 5.9 7.5 8.0 8.4 7.9 8.1 8.0 8.0 7.6 7.8 US$/I Foreign Direct Mil. US$ 0.0 12.0 20.0 50.0 150.0 148.5 157.8 172.2 183.4 192.8 224.8 Investment/2 Tourism Earnings/2 Mil. USS 94.7 120.0 146.8 192.1 258.1 322.0 392.4 570.0 733.3 739.1 725.0 Monetary Sector Average Deposit % 26.0 26.0 24.0 25.0 21.0 16.7 10.0 8.5 7.4 6.4 4.4 Rate/2 Average Lending % 26.0 30.0 30.0 31.5 35.5 33.5 26.5 24.4 23.0 23.1 21.2 Rate/2 Growth in Money % 26.9 42.7 39.3 35.5 32.2 8.7 13.3 10.8 18.6 14.8 17.1 Supply (M3)/2 Government Finance TotalDomestic (asa% 13.9 14.1 12.8 14.5 12.5 13.2 13.5 12.6 12.5 11.3 12.2 Revenue/I of GDP) TaxRevenue/I (asa% 12.3 12.5 11.5 13.2 11.3 11.3 11.9 11.5 11.1 10.0 10.9 of GDP) Non-Tax Revenue/l (as a % - 1.6 1.6 1.4 1.3 1.2 1.9 1.6 1.1 1.3 1.3 1.3 of GDP) TotalExpenditure/I (asa% 15.8 15.9 23.9 21.5 18.3 17.6 15.1 15.5 16.2 17.0 17.2 of GDP) Recurrent (asa% 14.1 13.2 18.8 17.0 15.1 14.0 12.5 11.5 11.9 11.7 13.4 Expenditure/] of GDP) Development (as a% 1.7 2.7 5.1 4.5 3.2 3.6 2.6 4.0 4.5 5.2 3.8 Expenditurell of GDP) Grants/i (as a% 2.4 2.7 10.5 9.9 6.4 2.5 -0.5 3.9 2.8 5.7 3.9 of GDP) Fiscal Balance/l (as a % -1.9 -1.7 -11.0 -7.0 -5.9 -4.3 -1.6 -3.0 -3.8 -5.7 -5.0 of GDP) Note /1 Fiscal year is used, and it ends in mentioned year June 30th. /2 Calendar year is used, and it ends in mentioned year December 31th. Values for the year ending December 3 Ith 1999 are likely outtums. * For calendar year the figure is provisional while for fiscal year the figure is actual. Sources: The Planning Commission, The United Republic of Tanzania, Economic Survey, various issues. Bureau of Statistics, Planning Commission, National Accounts of Tanzania, various issues. The Bank of Tanzania, Economic Bulletin, various issues. 147 Table 2: Balance df Payments (in MilL of US dollars). lnums 5993 199 1993 1994 1995 1996 1997 199 1999 2000 loot, CURtRENT ACCOUNT -736.1 -704.3 .1022.0 .711.1 -599 9 -265.1 -372.3 -823 9 .739.6 -389 4 .4t3.5 Gooxs 4865.3 .915.9 4835.6 -790.0 .6576 .448.9 -395.4 -777.5 -924.9 .672.8 -713.2 Exporu(fob) 362.3 400.7 439.3 5194 682.9 763.8 732.6 5388. 543.3 662.3 7764 lmpm(fob) 1227.6 1316.6 1274.9 1309.3 13340.5 1212.6 1148.0 1366 0 13682 3334 9 1499.7 Savices .157.5 .169.0 .3899 9 .851 .2169 .279 9 .306.4 .441.1 -226.9 .95.9 .98 9 Reseipes 142.3 367.5 3309 419 2 592 9 537.1 49398 553 0 646.4 663 7 712 4 payments 299 6 336.6 700.7 503 3 799.9 935.9 9002 994.3 873 2 7595 9 11 3 Incoeso -194.2 -223.4 -147.6 -1225 -330.3 -72.0 .122.8 .124 5 -75 5 .93.9 .44.2 Rae"opi 7.9 9.3 23.4 30.9 31.8 41.5 44.9 48.0 56.3 63.0 92.0 Peyrsa6 392.3 233.5 169.0 353 4 342.3 333.5 167.7 372.5 131.5 356 9 32631 Cinm nmters 4709 606 1 351.3 286.3 394.9 334 6 452.3 519.2 388 6 474 442.9 Infkoms 503.5 643.1 393.1 331.5 427.2 566.9 5200 546.3 333.9 55098 536. Govuommin 490.3 506.2 370.9 215.0 292.5 432.2 421.6 429.7 350.3 403 6 409.6 Privuje 23.2 134.8 30.2 96.5 134.7 134.7 98.4 316.3 161.8 147.2 307.0 Oulllows 32.6 35.0 300 23 0 32.3 32.3 67.7 26 9 123.3 76.9 73.8 CAPIrAL ACCOLUNT 353.3 2982 200.6 262.6 191.0 391.0 366.9 299.5 322.5 330.4 382.6 CapilalIansas 353.3 298.2 200.6 292.6 191 0 393.0 366.9 299.5 322.5 3304 332.6 [Rfoos 353.1 298.2 200.6 262.6 391.0 393.0 366.9 299 5 322.5 330.4 392.6 Outfkws 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 A 6smdodispaaaotmmal.pcuciod 0 0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 FIN4ANCIAL ACCOUNT 317.9 307.3 57 0 .106.9 139.5 OR4 .139.9 14 5 177.3 22.2 6 4 Direhct nUOnai 0.0 32.0 20.0 50.0 350.0 348.5 357.8 372.2 193.4 192.9 224.8 Abroad 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 00 in Tanzania 0.0 12.0 20.0 50.0 350.0 149.5 357.8 172.2 193.4 392.9 22491 Pmnfolio investmndi 0.0 0.0 0.0 0.0 0.0 0.0 0 0 0.0 0.0 0.0 0.0 Odiajrvasessw 137.9 95.3 37.0 .15698 -30.5 .146.2 .297 7 -257.8 .6.2 .170.6 -239 4 Inflow oCf lnanciarsm 272.0 333.9 633.4 434.5 437.4 338.6 400.6 468.1 503.2 503.4 270.4 Disbnrsement ofGovansnulows 260.5 267.5 372.4 274.2 244.3 224.9 269.8 275.5 335.8 166.2 348.2 Tmda aedi md dcier w0msciaflows 30.5 64.4 143.2 322.3 141.9 77.0 119.4 329.9 83.3 82.3 29.9 Dlebiscmmlotlcanbstoba & odimestmm 0.0 0.0 99.8 36.2 51.4 36.7 27.5 32.7 3.5 3.1 92.3 Ou1flow0o(3ca3rauc1 5331 236.6 5781.4 591.3 447.9 466 7 698.3 625.9 507.3 474.0 498 8 ReInymemol'oinoamscaticani 145 7 236.6 498.3 491.3 272.5 3,629 757.6 479.5 404.5 485 7 395 6 Truk eaalld oebabmsciaflllo 74 0.0 68. 75.6 162.6 43.1 .307.5 99.9 57.4 353.8 34.6 RpopcAymeorlosm;by bonb & ods sectous 0.0 0.0 13.9 24.4 32.9 63.9 60.9 47.4 53 9 53.2 5871 E Mnmud om*mn 3531 46.2 27.7 94.0 .70.0 .73.1 -54.4 33.2 9.3 .98.3 32.0 OVERALL BALANICE -232.3 -232.6 .736.7 .463.3 -329.5 .146.9 -399.7 .476.7 -229 6 -3344 .32.5 FDINAjaY4 .232.1 .252.6 736.7 46133 329.5 3469 399.7 476.7 229.6 1334.1 32.5 NetRumavesets(.loincrae) 485.3 -253.5 137.3 .77.0 60.3 -165.3 484.4 -1.13. -323.4 .137.6 -379.1 LC'AR 0.0 0.0 LaCe hate findnks 317.4 5016.3 579.3 558.3 269.4 5312.2 484.2 48789 3509 273 7 190.6 Ainun 344.7 206.7 199.0 296.4 269.4 332.2 1350.9 363 I 347.0 306.2 595 RdafecbediSl 0 0 1381.7 136.6 144.5 0.0 0.0 159.2 175.3 116.2 0.0 1333 Debtiefgvasm 0.0 25.0 23.5 23.8 0.0 0.0 194 0 351.4 87.7 1305 0 UseofFind audii .27.3 135.7 704 3.1 .14.5 -28.4 77.4 21.0 45.3 11.4 Onesallsonwing fr OP mPoA" 1 49.9 73.6 56.5 98.3 356.3 169.2 98.6 3044 0.0 ls gq . O 0. 0.0 0.0 0.0 0.0 0.0 0.0 0 0 0.O Mtemoruandm om, GD?(mp) BjULZS 3607.8 2325.3 2796.6 3452.6 428.6 5047.9 62820 7036.4 805. GDP(Wp) Milt. USD 3768.3 4042.2 3967.3 4370.4 486.7 5952.9 6994.7 7594 4 9434.5 8766.3 8770.2 CAB/GP .39.5 .17.4 -258a .17.1 .13 3 .7.7 48.0 .13.1 .9.4 -5.9 47 CABIGDP(exc3.ar O1llWIalsusefin) .32.3 .-0. .3S53 -22.2 -18.3 .11.7 .11.4 -36.5 .32.9 .90 _9.9 Gro.sOMlciaReaves 242.2 394.3 228.3 333 3 270.9 441.1 623.1 599.0 775.6 974 336.6 W..&. t1mp 8.2 12.4 6.0 9.5 6.6 1.A6 32 3. 24.2 26.3 Noee .prorbiso Soeree Bdlut of PWaynm DapuMe. Ba,k of Taissan. 148 Table 3: Summary of Central Government Operations (in Bill. Tsh.) FY91 FY92 FY93 FY94 FY95 FY96 FY97 FY9S FY99 FY00 FY01 FY02- Total Revenue 133 174 164 242 331 448 572 619 689 778 930 764 Tax revenue 118 153 146 220 300 384 505 566 616 685 828 694 Nontaxrcvenue 15 20 18 22 31 65 67 53 73 93 102 70 Total Expenditure 152 195 305 359 488 596 638 765 899 1169 1307 989 Recurrent expenditure 135 162 240 284 402 475 528 568 657 809 1021 805 Developmentexpenditureand 16 33 65 75 86 121 110 197 248 360 286 184 part of net lending Ovrall aluance (checks issued) -19 -21 -141 -116 -157 -148 -66 -146 -209 -391 -378 -225 Expenditure Float 23 34 Bank & Parataal Recapitaleation .. .. .. 10 0 Adjusnent to cash and other -9 -2 7 -37 -90 -73 -37 .82 -27 -4 -4 26 items (net) Overll balance (checks-cleared) -27 -23 -134 -153 -246 -221 -103 -229 -237 -395 -414 -233 Overall balance (aftergrants) 4 10 -76 -76 -193 -146 -21 -72 -61 -115 -121 -10 Financing: 27 23 134 165 170 86 -20 194 157 395 414 233 External gramnts 23 33 58 77 54 75 81 156 175 280 293 223 ForeignFinancing.netl/ 9 23 . 29 48 26 -45 -49 34 -19 105 87 112 Domestic Borrowing *4 -32 47 41 90 56 -53 4 -6 9 -2 -55 Privatisation Fund .. .. .. .. .. .. .. .. 7 .. 27 0 Bank & Parastal Recapital-ation .. .. .. .. .. .. .. .. .. .. 10 0 Changs in arreas .. .. ,. .. .. ., .. ., .. .. 0 -47 Contigency .. .. .. .. .. .. .. .. .. .. 0 0 As % of GDP at market prkes Total Revenue 13.9 14.1 12.8 14.5 12.5 13.2 13.5 12.6 12.5 11.3 12.2 9.1 Taxrevenue 12.3 12.5 11.5 13.2 11.3 11.3 11.9 11.5 11.1 10.0 10.9 8.3 Nontax revenue 1.6 1.6 IA 1.3 1.2 1.9 1.6 1.1 1.3 1.3 1.3 0.8 TotalExpenditureandNetLending 15.8 15.9 23.9 21.5 183 17.6 15.1 15.5 16.2 17.0 17.2 11.8 Recurrent expenditure 14.1 13.2 18.8 17.0 15.1 14.0 12.5 11.5 11.9 11.7 13.4 9.6 Development expenditure and 1.7 2.7 5.1 4.5 3.2 3.6 2.6 4.0 4.5 5.2 3.8 2.2 part of ne ledinS Oveal balance (checks isued) -1.9 -1.7 -11.0 -7.0 -5.9 -43 -1.6 -3.0 -3.8 -5.7 -5.0 -2.7 Adjustmenttocashandother -0.9 -0.2 0.5 -2.2 -3A -2.2 -0.9 -1.7 -0.5 -0.1 0.0 0.3 items (ne) Ovrall balance (checks-cleated) -2.9 -1.9 -10.5 -9.2 -9.3 -6.5 -2.4 -4.6 -43 -5.7 -5.0 -2.8 Ovaallbalance(aftergrants) 0.4 0.9 -5.9 -4.6 -72 -4.3 -0.5 -1.5 -1.1 -1.7 -12 -0.I Financing: 2.9 1.9 -5.9 -4.6 -7.2 43 -0.5 -1.5 -1.1 -1.7 5.9 2. Extenal grants 2A 2.7 10.5 9.9 6A 2.5 -0.5 3.9 2.8 5.7 3.9 2. Foreign rinancinLg nc1: 0.9 1.8 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 1.6 13 Domestic Borrowing -0.5 -2.6 4.6 4.6 2.0 2.2 1.9 3.2 3.2 4.1 0.0 -0. 1/ Includes foreign pnats pior to FY87. Note * Figures cover the period July 2001 to March 2002 Sour: The Tanzanian authorities. 149 Table 4a: Budget Frame - Analytical (in Bill. Tsh.) 1994195 1995/96 1996/97 1997/98 1998/99 1999/00 2000101 2001/02 2002/03 2003/04 2004105 Actual Actual Actual Actual Actual Actual Actual Budgct Proi. Proj. Proj. Domestic revenue 331 448 572 619 689 778 930 1,025 1.123 1,246 1381 O/W Road Toll 0 0 34 37 38 38 40 48 53 56 62 Total Expenditure 413 446 644 730 863 1,180 1,275 1,579 1,987 2.014 2,184 Recurrentexpenditure 346 416 519 544 680 820 989 1,226 1,319 1,423 1,560 Interest on extemal debt 22 39 38 48 60 47 52 47 45 48 48 lnterest on domestic debt 25 27 74 53 30 81 78 73 55 55 55 Wages/salaries 110 156 199 219 232 - 285 308 356 411 453 511 Goods/services/transfers 189 194 209 224 358 406 551 750 807 867 946 o/w RoadSector 16 16 20 17 38 38 55 48 53 56 62 Special exp. 6 35 29 44 42 0 37 201 179 187 267 CFS (Others) 19 35 24 23 23 56 62 69 74 77 79 TRA 0 0 16 17 20 22 0 0 0 0 0 Parastatal Wages 0 0 24 28 34 52 0 0 0 0 0 Retention Scheme 0 0 11 14 29 23 26 44 48 53 59 Election Costs 0 0 0 0 0 0 26 0 0 0 0 Census 0 0 0 0 0 0 0 0 0 0 0 OtherCharges 168 153 85 82 171 216 345 389 452 493 480 Development expenditure 67 30 125 187 183 360 286 353 669 591 624 Projects 67 30 84 187 183 360 286 350 664 586 619 Local 20 5 20 24 19 19 35 47 62 55 60 Foreign 47 25 64 163 164 340 251 302 601 531 559 Other Programnme Assistance 0 0 40/ 0 0 0 0 0 0 0 0 Energy Fund Songo songo 0 0 0 0 0 0 0 3 5 5 5 Overall deficit (checks issued) - -81 2 -72 -111 -173 -402 -345 -553 -865 -768 -803 before grants Grants 67 47 115 119 170 292 293 451 593 550 574 Balance of Payment SupporL 35 30 67 2 69 73 114 190 164 173 180 o/w MDF/PRBS 0 0 0 0 70 71 84 171 146 158 165 project grnts I/ 33 17 49 118 101 208 124 197 358 301 317 HIPC interim relief-Multilateral 0 0 0 0 0 11 56 64 72 76 77 2/ Expenditure float 0 0 0 0 0 0 23 0 0 0 0 Overall deficit (checks issued) - -14 49 43 8 -4 -111 -75 -102 -271 -219 -229 after grants Adjustment -20 -85 0 7 -45 -2 -4 0 0 0 0 Overall deficit (cheeks cleared) -34 -36 43 is -49 -113 -79 -102 -271 -219 -229 Financing 34 36 -43 15 49 113 79 102 271 219 229 Foreign -24 -35 9 64 27 105 87 100 253 219 229 Programtnme loans 0 8 47 80 37 55 40 129 107 112 119 project loan 14 8 15 57 63 133 128 105 243 230 242 amnortization -38 -50 -53 -72 -73 .82 -80 -134 -97 -124 -132 Local (net) 58 71 -53 -50 21 8 -2 24 0 0 0 Bank (net) 55 63 -48 -24 8 8 -19 34 0 0 0 Non-bank 3 8 -5 -26 13 0 16 -10 0 0 0 borrowing 6 -6 4 27 45 0 0 0 15 0 0 amortization -3 -4 0 -53 -38 -234 0 -10 -15 0 0 Privatisation Funds 0 18 25 0 7 0 27 20 18 0 0 Change in Arears 0 0 -34 0 0 0 -32 -42 0 0 0 Bank Recapitalisation 0 0 0 0 0 89 0 0 0 0 0 Contigency 0. 0 0 0 0 0 0 0 0 0 Financing Gap . 0 0 0 0 0 0 0 0 0 0 memo: GDPmp 2,t60 3,394 4,236 4,929 5,532 6,885 7,600 8,381 9,224 10,151 11,166 OC for distrlbuton 0 0 0 0 0 299 489 681 733 790 867 Primary Deflcit(checks issue) -34 68 40 -10 -83 -274 216 -434 -765 -665 *700 Government Savng(checks issued) -15 32 53 75 9 -42 -59 -201 -196 -177 -178 %ofGDP -0.6% 1.0% 1.2% 1.5% 0.2% -0.6% -0.8% -2.4% -2.1% -1.70% -1.6% 1/ Revised as of July 25,2001. This takes into account: i. Revenue forecast based on actual full year 2000/01 provisional collections which amnounted to Tsh. 929,624 million against revised targets of Tsh. 912,906 million H. The budgeted revenue from Gaming Tax has been revised downwards to 0 iii. Foreign interest payments have been revised upwards by Tsh. 3,450 million iv. Domestic amortisation has been revised downwards by Tsh. 5,485 million, since Steyn was paid in FY 2000/01 v. Programme loans have been adjusted upwards due to the PSAC disbursement of Tsh. 35,600 million now expected in July, 2001 consequently the bank borrowing figure has been adjusted downwards by Tsh. 35,600 million 150 Table 4b: Budget Frame - Analtica (as % of GDP) 1994/ 1995/ 1996/ 1997/ 1998/ 1999/ 2000/ 2001/ 2002/ 2003/ 2004/0 95 96 97 98 99 00 01 02 03 04 5 Actual Actual Actual Actual Actual Actual Actual Budge Proj. Proj. Proj. Domestic revenue 12.5% 13.2% 13.5% 12.6% 12.5% 11.3% 12.2% 12.2% 12.2% 12.3% 12.4% O/W Road Fund 0.0%/a 0.8%/o 0.7% 0.7% 0.5% 0.5% 0.6% 0.6% 0.6% 0.6% Total Expenditure 15.5% 13.1% 15.2% 14.8% 15.6% 17.1% 16.8% 18.8% 21.5% 19.8% 19.60% Recurrentexpenditure 13.00/o 12.3% 12.3% 11.0'/. 12.3% 11.9% 13.00/% 14.6% 14.3% 14.0% 14.0% Interest on external debt 0.8% 1.1% 0.9%/o 1.0/ 1.1% 0.7/o 0.7% 0.6% 0.5% 0.5% 0.4% Interest on domestic debt 1.0% 0.8% 1.7% 1.1% 0.5% 1.2% 1.0% 0.9% 0.6% 0.5% 0.5% Wages/salaries 4.1% 4.6% 4.7% 4.4% 4.2% 4.1% 4.1% 4.3% 4.5% 4.5% 4.6%/ Goods/services/transfers 7.1% 5.7%/ 4.90/O 4.5% 6.5% 5.9% 7.3% 9.0% 8.8% 8.5% 8.5% o/w Road Fund 0.6% 0.5% 0.5% 0.3% 0.7% 0.5% 0.7% 0.6% 0.6% 0.6% 0.6% Special exp. 0.2% 1.0% 0.7% 0.9% 0.8% 0.0% 0.5% 2.4% 1.9Yo 1.8% 2.4% CFS (Others) 0.7% 1.0% 0.6% 0.5% 0.4% 0.8% 0.8% 0.8% 0.8% 0.8% 0.7% TRA 0.0%h 0.0% 0.4% 0.3% 0.4% 0.3% 0.0% 0.0%0 0.0 0.0/O 0.00% Parastatal Wages 0.0% 0.0% 0.6% 0.6% 0.6% 0.8% 0.00/o 0.00/o 0.0%/o 0.0% 0.0%/ Retention Scheme 0.0% 0.0%/o 0.3% 0.3% 0.5% 03% 0.3% 0.5% 0.5% 0.5% 0.5% Other Charges 6.3% 4.5% 2.0% 1.7% 3.1% 3.1% 4.5% 4.6% 4.9% 4.90/% 4.3% Development expenditure 2.5% 0.9% 2.9% 3.8% 3.3% 5.2% 3.8% 4.2% 7.2% 5.8% 5.6% Projects 2.5% 0.9% 2.%o 3.8% 3.3% 5.2% 3.8% 4.2% 7.2% 5.8% 5.5% Local 0.8% 0.2% 0.5% 0.5% 0.3% 0.3% 0.5% 0.6% 0.7% 0.5% 0.5% Foreign 1.8% 0.7% 1.5% 3.3% 3.0% 4.9% 3.3% 3.6% 6.5% 5.2% 5.0%/a Other Programme 0.0% 0.0%/0 1.0%/0 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0%v Assistance Overall deficit (checks issued) - -3.1% 0.1% -1.7% -2.3% -3.1% -5.8% -4.5% -6.6% -9.4% -7.6% -7.2% before grants Grants 2.5% 1.4% 2.7% 2A% 3.1% 4.2% 3.9%/o 5.4% 6.4% 5.4% 5.1% Programmegrants 1.3% 0.9% 1.6% 0.0% 1.3% 1.1% 1.5% 2.3% 1.8% 1.7% 1.6%/ o/w PRBS/MDF 1.1% 2.0% 1.6% 1.6% 1.5% project grants 1.2% 0.5% 1.2% 2.4% 1.8% 3.0% 1.6% 2.4% 3.9% 3.00/% 2.8% HIPC interim relief 0.2% 0.7% 0.8% 0.8% 0.8% 0.7r Overall deficit (checks issued) - -0.5% 1.4% 1.0°h 0.2% -0.1% -1.6% -1.0%/o -1.2% -2.9% -2.2% -2.0%/ after grants Adjustment -0.8% -2.5% 0.0%/0 0.1% -0.8% 0.0% 0.0%/0 0.0% 0.0%/0 0.00/0 0.0%/. Overall deficit (checks cleared) -1.3% -1.1% 1.0% 0.3% -0.9% -1.6% -1.0% -1.2% -2.9% -2.2% -2.0% Financing 1.3% 1.1% -1.00/o 0.3% 0.9% 1.6% 1.0%/o 1.2% 2.9% 2.2% 2.0% Foreign -0.9% -1.0% 0.2% 1.3% 0.5% 1.5% 1.1% 1.2% 2.7% 2.2% 2.00/% importsupport loans 0.0% 0.2% 1.1% 1.6% 0.7% 0.8% 0.5% 1.5% 1.2% 1.1% 1.1% project loan 0.5% 0.2% 0.4% 1.2% 1.1% 1.9% 1.7% 1.3% 2.6% 2.3% 2.2% amortization -1.4% -1.5% -1.2% -1.5% -1.3% -1.2% -1.1% -1.6% -1.0% -1.2% -1.2% Local (net) 2.2% 2.1% -1.2% -1.0°/. 0.4% 0.1% 0.0%/o 0.3% 0.0% 0.0°/. 0.0%/ Bank (net) 2.1% 1.8% -1.1% -0.5% 0.1% 0.1% -0.2% 0.4% 0.0% 0.0% 0.0%/ Non-bank 0.1% 0.2% -0.1% -0.5% 0.2% 0.0% 0.2% -0.1% 0.0Yo 0.0% 0.00/ borroving 0.2% -0.2% 0.1% 0.6% 0.8% 0.0%/o 0.0%/a 0.0% 0.2% 0.00/% 0.0% amortization -0.1% -0.1% 0.0%h -1.1% -0.7% -3.4% 0.0W -0.1% -0.2% 0.0% 0.0% Privatisation Funds 0.0% 0.5% 0.6% 0.0% 0.1% 0.0%/o 0.4% 0.2% 0.2% 0.00/o 0.0%F Change in Anrears -0.8% 0.0% 0.0% 0.0% -0.4% -0.5% 0.0% 0.0% 0.0'h Bank Recapitalisation 1.3% 0.0%/o 0.0% 0.0% 0.0%/9 0.0% Contingency 0.0% 0.0/O -0.5% 0.0o 0.00/ 0.0% memo: GDPnp (BilL Tsh.) 2,660 3,394 4,236 4,929 5,532 6,88'5 7,600 8,381 9,224 10,15 11,166 1 Primary Deficit(checlks Lse) -1.3% 2.0%/6 0.90/o -0.2% -1.5% -4.00/. -2.8% -5.2% -8.3% -6.5% -6.3% Government Savb,g(checks -0.6% 1.0% 1.2% 1.5% 02% -0.6% -0.8% -2.4% -2.1% -1.7% -1.6% Source: Table 4a 151 Table 5a: Budget Frame - Accounting (in Bill. Tsh.) 1994/95 1995/96 1996/97 1997/98 1998/99 1999/00 2000/01 2001/02 2002/03 2003/04 2004/05 Actual Actual Actual Actual Actual Actual Actual Budget Proj. Proj. Proj. 1. TOTAL RESOURCES 453 500 731 856 974 1,262 1.411 1,765 2,084 2,138 2316 Domestic revenue 331 448 572 619 689 778 930 1,025 1,123 1,246 1,381 Programme loan and grants 35 38 114 82 106 128 .154 319 270 285 299 Projectloansandgrants 47 25 64 174 164 340 251 302 601 531 559 HIPC Interim Relief Multilateral 0 0 0 0 0 11 56 64 72 76 77 HIPC interim relief-Paris Club 0 0 0 0 0 0 0 0 0 0 0 NonBankBorrowing 6 -6 4 27 45 0 16 0 0 0 0 Bank Borrowing 55 63 -48 -24 8 8 -19 0 0 0 0 Drawdown of reserves 0 0 0 0 0 0 0 34 0 0 0 Adjustment to cash -20 -85. 0 -23 -45 -2 -4 0 0 0 0 Privatisation Funds 0 18 25 0 7 0 27 20 18 0 0 11. TOTAL EXPENDITURE 453 500 731 856 974 1,262 1,411 1,765 2,084 2,138 2.316 RECURRENT EXPENDITURE 387 470 606 670 791 902 1,125 1,412 1,415 1,547 1,692 CFS . 107 155 188 250 225 266 272 333 271 304 314 Debt service 88 120 164 227 202 211 210 264 197 227 235 interest 48 66 112 101 91 128 130 120 100 104 103 amortization 41 54 53 126 111 82 80 144 97 124 132 Others 19 35 24 23 23 56 62 69 74 77 79 Recurrent Exp.(excl. CFS) 280 315 384 420 566 636 821 1,038 1,145 1,243 1,378 o/w Salaries &wages 110 156 199 219 232 285 308 356 411 453 511 Other Charges 168 153 85 82 171 216 345 389 452 493 480 DesignatedItemsl/ 0 0 100 119 163 135 145 293 281 297 388 Payment of Arrears 0 0 34 0 0 0 32 42 0 0 0 Expenditure Float 0 0 0 0 0 0 23 0 0 0 0 CONTINGENCY2/ 0 0 0 0 0 0 0 0 0 0 0 DEVELOPMENT 67 30 125 187 183 360 286 353 669 591 624 EXPENDIrURE Projects 67 30 84 187 183 360 286 353 669 591 624 Local 20 5 20 24 19 19 35 47 62 55 60 Foreign 47 25 64 163 164 340 251 302 601 531 559 BasketFunding 3/ 0 0 40 0 0 0 0 10 0 0 0 1/ Includes contingent expenditures that will be voted at a later stage 2/ Fiscal risk provision 3/ From 2002/03 included in programme grants Note: Recurrent expenditure under the Poverty Reduction Growth Facility (PRGF) is Tshs. 933.1 billion while the budgeted figure is Tshs. 952.5 billion. This divergence is explained by Tshs. 5.0 billion increase in revenue projection which is included in the road sector, Tshs. 4.5 bilion have been re-allocated from development expenditure.to recurrent expenditure under the road sector. The remaining difference of Tshs. 10 billion is due to the different treatment of arrears - the IMF presents this amount below the line while the budget presents this anount under recurrent expenditure. Having taken the afbrementioned into consideration the budget figures are consistent with the agreed PRGF figures. Source: Ministry of Finance 152 Table Sb: Budget Frame - Accounting ( as % of GDP) 1992/93 1993/94 1994/95 1995/96 1996/97 1997/98 1998/99 1999/00 2000/01 2001/02 2002/03 2003/04 2004/05 Actual Actual Actual Actual Actual Actual Actual Actual Actual Budget Proj. Proj. Proj. .TOTAL RESOURCES 26.9% 23.9% 17.0% 14.7% 173% 17A°% 17.6% 183% 18.6% '21.1% 22.6% 21.1% 20.7°/ Domestic revenue 12.8% 14.5% 12.5% 13.2%/e 13.5% 12.6%/ 12.5% 11.3% 12.2% 12.2% 12.2% 12.3% 12.40Y Programmeloan and grants 6.2% 6.5% 1.3% 1.1% 2.7% 1.7% 1.9%/0 1.9%/ 2.0% 3.8% 2.9% 2.8% 2.70A Project oans and grants 3.2% 3.4% 1.8% 0.7%o 1.5% 3.5% 3.0%o 4.9% 3.3% 3.6% 6.5% 5.2% 5.0°A HIPC Interim Relief Multilaterai 0.2% 0.7% 0.8% 0.8% 0.8% 0.7°A HiPC interim relief-Paris Club 0.0%/0 Non Bank Borrowing 0.5% 1.8% 0.2% -0.2% 0.1% 0.6% 0.8% 0.o0 0.2% 0.0% 0.0% 0.0% 0.0°/ BankBorrowing 3A4% 0.Vh 2.1% 1.8Yo -1.1% -0.5% 0.1% 0.1% -0.2% 0.0%/ 0.0% 0.0%/o 0.0°A Drawdown of reserves 0.0% 0.0%/0 0.4% 0.0% 0.0%/0 0.0A/ Adjustment to cash 0.8% -3.1% -0.8% -2.5% 0.0%/0 -0.5% -0.8%/o 0.0% 0.0 0.0%o 0.0% 0.0%00 0.OVA Privatisation Funds O.0O 0.0%/0 0.00/h 0.5% 0.6% 0.00/h 0.1% 0.0/0 0.4% 0.2% 0.2% 0.0o/% 0.0°/ L TOTAL EXPENDI[TURE 26.9% 23.9% 17.0% 14.7% 17.3% 17.4% 17.6% 183% 18.6% 21.1% 22.6% 21;1% 20.7VA RECURRENTEXPENDITURE 21.Yo 19.4% 14.5% 13.8% 14.3% 13.6% 14.3% 13.1% 14.8% 16.90%, 15.3% 15.2% 15.2°A CFS 6.4% 5.2% 4.0% 4.6% 4.4% 5.1% 4.1% 3.9% 3.6% 4.0% 2.9% 3.0% 2.80/ Debtsenvice 6.1% 4.7% 3.3% 3.5% 3.9% 4.6% 3.6% 3.1% 2.8% 3.1% 2.1% 2.2% 2.1N! interet 3.3% 2.3% 1.8% 1.9%Y 2.6% 2.1% 1.6% 1.9% 1.7% 1.4% 1.1% 1.0% 0.9°A amortization 2.8% 2.4% 1.5% 1.6% 1.2% 2.6% 2.0% 1.2% 1.1% 1.7% 1.0%h 1.2% 1.20/ Others 0.3% 0.5% 0.7/% 1.0% 0.6% 0.5% 0.4% 0.8% 0.8% 0.8% 0.8% 0.8% 0.70/ Recufmnt Exp.(cxd. CFS) 15.5% 14.2% 10.5% 9.3% 9.1% 8.5% 10.2% 9.2% 10.8% 12.4% 12.4% 12.2% 12.3°A o/w Salaries & wages 4.5% 4.4% 4.1% 4.6% 4.7% 4.4% 4.2% 4.1% 4.1% 4.3% 4.5% 4.5% 4.6°/ Olher Charges 10.9%Y 9.8% 6.3% 4.5% 2.00/% 1.7% 3.1% 3.1% 4.5% 4.6% 4.9% 4.9% 4.3°A Designated Items 0.0%/ 0.0%0 O.0Y% 2.4% 2.4% 3.0% 2.0% 1.9% 3.5% 3.0% 2.9% 3.5°A Payment of Arrears 0.0% 0.0% 0.0%/0 0.0% 0.8% 0.0% 0.0% 0.0% 0.4% 0.5% 0.0%h 0.0% 0.00' Expemdituw Float 0 3% Contingency 0.0% 0.0% O.ONo 0.0% 0.0% 0.0°/ DEVELOPMENT EXPENDiTURE 5.1% 4.5% 2.5% 0.9Yo 2.9% 3.8% 3.3% 5.2% 3.8% 4.2% 7.2% 5.8% 5.60N Projects 5.1% 4.5% 2.5% 0.9°/. 2.0% 3.8% 3.3% 5.2% 3.8% 4.2% 7.2% 5.8% 5.6°A Local 1.8% 1.1% 0.8% 0.2% 0.5% 0.5% 0.3% 0.3% 0.5% 0.6% 0.7% 0.5% 0.5N! Foreign 3.2% 3.4% 1.8% 0.7% 1.5% 3.3% 3.0% 4.9% 3.3% 3.6% 6.5% 5.2% 5.0°A Baskd Funding 0.0%G 0.0%! 0.0/0 0.0!% 1.0% 0.0%/0 0.0%/0 0.0/!0 0.0% 0.1% 0.0% 0.0% 0.0°! Source: Table Sa Table 6a: Sectoral - Recurrent Expenditure (in Bill. Tsh.) VOTE VOTE HOLDER 199619 1996/9 1997/9 1997/9 1998/9 1998/9 199910 199910 2000/01 2000/01 2001/02 2001/02* 7 7 8 8 9 9 0 0 Budget Actual Budget Actual Budget Actual Budget Actual -Budget Actual Budget Actual ADMINISTRATION 23.0 Accountant Gencral 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 22.5 22.4 77.6 47.6 27.0 Registrar of Political Parties 0.0 0.0 3.4 3.4 2.6 2.6 2.7 1.3 2.8 2.7 3.9 2.9 30.0 Presidents Office & The 9.5 9.5 10.5 10.5 15.7 15.5 19.6 19.3 24.7 24.7 30.1 23.4 Cabinct 26 &31 Vice President 0.5 0.5 0.8 0.8 0.9 0.9 1.0 1.0 1.4 1.4 2.1 1.6 32.0 Civil Service Dept. 2.0 2.0 2.4 2.4 3.7 3.7 3.4 3.4 4.3 4.3 4.9 3.9 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 0.2) 34.0 Foreign Affairs 10.5 12.0 13.3 15.2 18.3 19.4 17.1 16.6 20.5 24.5 19.7 15.7 35.0 Penm. Comm. Enq. 0.1 0.1 0.2 0.2 0.3 0.3 0.2 0.2 0.3 0.3 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 0.5 0.5 0.6 0.3 25 &37 Prime Ministers Office 3.9 3.9 9.1 5.8 14.1 6.0 4.7 4.7 5.0 5.0 10.9 3.7 40.0 Judiciary 3.6 3.6 4.6 4.5 5.4 5.3 7.9 6.0 8.8 8.8 11.6 8.5 4 1.0 Justice &Constitutional Affairs 0.8 0.8 0.8 0.9 2.2 1.9 1.1 0.9 1.5 1.5 1.7 1.6 42.0 Office Of The Speaker 3.2 3.1 3.6 3.5 5.3 5.4 5.9 5.9 6.6 6.6 7.6 5.6 45.0 Exchequer and Audit 0.5 0.5 0.7 0.7 0.8 0.8 1.1 1.1 1.8 1.8 1.9 1.4 50.0 Finance 20.5 19.7 33.5 28.2 51.1 51.9 49.7 48.5 41.0 40.9 189.2 33.5 5 1.0 Home Affairs 1.6 1.6 2.0 1.7 4.6 4.6 5.8 4.8 5.6 5.6 6.9 3.6 54.0 Radio Tanzania 0.3 0.3 0.3 0.3 1.2 1.1 1.2 1.2 1.6 1.2 1.7 1.0 55.0 Comm of Human Rights & . . . .. . . . .. . .. 0.9 0.2 Good Gov. 56.0 MORALG . . . . 6.2 2.7 5.3 16.6 3.9 3.9 20.5 10.7 57.0 Defence &National. 1.6 1.6" 2.4" 1.'7 4.5 3.1 2.3 2.3 4.1 4.1 2.9 2.4 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 0.2 0.1 60.0 Industrial Court Of Tanzania. 0.1 0.1 0.1 0.1 0.1 0.1 0.2 0.1 0.3 0.3 0.3 0.3 61.0 Electoral Commision 0.9 0.9 1.4 1.5 3.0 3.0 16.2 5.9 30.9 30.6 0.7 0.5 63.0 Local Govt. Scrvi. Comm. 0.1 0.1 0.1 0.1 0.2 0.2 0.2 0.2 0.3 0.3 0.4 0.3 64.0 Commercial Court . . . . . .. . .. . .. 0.5 0.2 66.0 Planning Commission 0.6 0.6 0.9 0.'9 2.7 2.6 1.9 1.3 2.3 2.3 16.5 5.5 Sub Total 60.5 61.1 90.7 12.9 143.1 131A 148.1 142.0 168.5 171.6 413.5 174.6 DEFENCE AND SECURITY 28.0 Police Force 23.8 23.8 28.8 25.0 29.7 29.7 34.9 32.2 42.0 41.9 40.2 29.1 29.0 Prison Services 12.2 12.2 17.2 15.1 16.4 16.4 19.4 16.2 20.9 20.8 22.6 15.8 38.0 Defence 63.8 63.5 72.5 66.7 73.2 74.0 81.7 78.5 89.2 89.2 96.6 67.9 39.0 National Service 7.5 7.5 9.5 8.9 11.4 11.8 12.1 10.9 12.5 12.5 14.8 11.1 Sub Total 107.3 107.0 127.9 115.7 130.8 13 1.9 148.1 137.8 164.6 164.4 174.2 123.9 SOCIAL SERVICES 46.0 Education 12.1 12.0 17.1 18.7 19.6 19.5 21.0 20.7 28.7 28.6 34.4 24.81 49.0 Water, 1.5 1.5 2.8 2.5 6.4 6.4 3.4 3.4 5.9 5.4 9.2 5.7 52.0 Health 18.8 15.9 25.8 25.9 37.2 37.2 33.8 32.1 40.1 39.9 47.0 31.8 53.0 Comm. Dev.Wome. Aff. 0.9 0.9 1.4 1.4 1.8 1.8 2.3 1.5 2.0 2.0 2.7 1.7 65.0 Labour Youth Develop. 1.3 1.3 1.5 1.5 2.0 2.0 3.0 1.5 3.2 3.1 4.3 2.4 67.0 Teachers Service Comm. 0.1 0.1 0.1 0.1 0.1 0.1 0.2 0.2 1.5 1.5 1.7 1.2 68.0 Science, Tech.& H Ed 21.3 19.4 22.9 21.1 32.8 32.8 32.3 30.6 44.9 44.7 52.7 39.1 70 -89 Regions 99.8 98.9 118.3 117.3 120.9 129.0 168.6 158.2 207.5 206.8 221.6 169.8 Sub Totail 155.8 150.0 189.8 188.6 220.8 228.7 264.6 248.1 333.9 332.0 373.7 276.5 ECONOMIC SERVICES 47.0 Works. 2.3 23.3 24.4 17.7 4.9 6.3 38.0 29.0 56.0 55.9 48.2 32.4 48.0 Lands, Hous.,&Urb. Dev. 0.6 0.5 1.8 0.6 3.4 2.8 3.6 3.6 4.3 4.2 4.6 3.1 58.0 Energy and Minerals. 0.5 0.5 0.9 0.8 2.0 1.9 3.0 2.9 3.7 3.7 7.6 12.3 62.0 Comm.& Transport 2.3 2.3 5.7 3.6 13.4 13.0 10.7 11.9 10.5 10.4 11.9 7.6 Sub Total 5.7 26.6 32.8 22.8 23.7 24.0 55.3 47.4 74.4 74.2 72.3 55.4 PRODUCTIVE 24 &43 Agriculture & ivestock, Coop. 11.9 1 1.8 13.9 12.4 19.3 18.3 12.3 7.4 12.1 12.1 14.5 11.2 & Marketing 44.0 Industries and Trade 1.4 1.3 2.9 2.9 4.6 4.6 3.6. 3.4 4.1 4.1 5.5 3.6 69.0 Tour.. Nat. Res. & Env. 2.3 2.3 2.8 2.3 7.9 8.0 8.7 7.4, 13.5 11.6 14.6 8.1 Sub Total 15.5 15.4 19.6 17.7 31.9 30.9 24.7 18.2 29.7 27.8 34.6 22.9 CFS 20.0 State house 0.9 0.8 0.8 0.8 0.9 0.9 1.0 1.0 1.1 1.1 1.4 1.0 22.0 Public Debt 217.7 210.9 207.1 253.2 238.0 232.2 294.4 286.7 286.4 261.9 337.5 1752 Sub Total 218.6 211.7 207.9 254.1 238.9 233.1 295.3 287.7 287.5 263.0 338.9 176. GRAND TOTAL 5634 571.8 668.8 681.7 789.2 780.0 936.1 881.2 1058.6 1033.1 1407.1 8295 Note: * Data for FY02-all votes are from the Expenditur Flash Reports (July 2001 - March 2002) Source: MoF, Appropriations Accounts mid Expenditur Flash Report 154 d mq d d d da 0 a t . O a d d * m _ d d i o d d d a d ad d a;d dad dd dd F4 d d; .d_9 j dod d d_S ddd dd a dddd d d - d d o d o d o a O O d o d d A o o o d oqd o d d d d 3 d d 9: 3 3q a ^ s 3 d d " oi u > o j 2; t d d; d oqdO t gm"Addd oqdd ddooo:To' dga d .q ±d ddo2f | ddaddodd.dod::doqodo d 333 3 dd A dd o'd d d d ddg d'ood5oZood 3nZ '!"0-dgddd 233 dtV -dd E~ ~do.d i!a13 1 a j0031t!!|}Sl}|§§ ±i{} OOO-CO 0090 ±0000 ±4. .40 (V. *|* 0*~ O 0 ddd n dddde dd..d ..dddd - d d d - ;x-dd-d da 6 d , da 3;a dd-.dd ddgdddd dd . d 2d: 2:.2 ;::a dd 000 r3:g00 : d e n - o. 0.0 .:40 0 0 lo~~~~~~~g 00000000 0000 V.00 00~~~ S gO ta00 °-0aCa *-aO o.COO C A- 00CC+an sa°¶ 0W°:. *00f Oh WC V | d-<4_o3d Cr C- t .9 d 0t = -00 - d d_ _ Z Z _ : _ d - .d o - dd dt3Y2 d_ _ ; " _ dit d- " dd 'd q- -n '_2 :.ng;d' ~ ~ ~ ~ ~ ~ -~ CO oo ~ 00 00o0d: -_3at at- ; " n -0r: fO q :~00bfq.9t .jq 00_*4 ra - CO 9< er.C~*-~a X CO 0 tan - > 0, | I§t:XX0l 010 §1 iljlllg9 9 -l ! i go > 0 ~ 0:-~ t 4 ~ 0 Table 7a: Actual Recurrent Expenditure by Vote as as the Share of Total Actual Recurrent Expenditure VOTE VOTE HOLDER 1995/96 1996/97 1997/98 1998/99 1999/00 2000/01 2001/02* ADMINISTRATION 23.0 Accountant General 0.00 0.00 0.00 0.00 0.00 2.17 5.74 27.0 Registrar of Political Parties 0.00 0.00 0.50 0.33 0.15 0.26 0.35 30.0 Presidents Office 1.57 1.65 1.54 1.99 2.19 2.39 2.82 31.0 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 0.47 33.0 Ethics Secretariat 0.00 0.02 0.03 0.02 0.01 0.02 0.02 34.0 Foreign Affairs 3.12 2.09 2.23 2A8 1.88 2.37 1.89 35.0 Pern. 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 Ofrice 1.15 0.68 0.84 0.77 0.53 OA9 0.45 40.0 Judiciary 0.63 0.62 0.66 0.68 0.68 0.85 1.02 41.0 Justice 0.11 0.13 0.13 0.25 0.10 0.15 0.19 42.0 Office Of The Speaker 0.56 0.55 0.52 0.69 0.67 0.64 0.68 45.0 Exchequerand Audit 0.09 0.09 0.10 0.10 0.12 0.17 0.17 50.0 Finance 3.34 3.45 4.14 6.65 5.51 3.96 4.04 51.0 Home Affairs 0.29 0.27 0.25 0.60 0.55 0.54 0.43 54.0 Radio Tanzania 0.05 0.05 0.04 0.14 0.13 0.11 0.12 55.0 Comm of Human Rights & Good Gov. .. .. .. .. .. .. 0.02 56.0 Regional Adminiastraion and Local Gov. .. .. .. 0.35 1.89 0.37 1.29 57.0 Defence& National. 0.26 0.27 0.25 0.39 0.26 0.40 0.29 59.0 Law Refonn. Commi. 0.01 0.01 0.02 0.01 0.01 0.02 0.01 60.0 Industrial Court Of Tz. 0.01 0.02 0.02 0.02 0.02 0.03 0.04 61.0 Electorl Commision 4.95 0.17 0.21 0.38 0.67 2.96 0.06 63.0 Local Govt Servi. Comm. 0.03 0.02 0.02 0.02 0.02 0.02 0.04 64.0 Commercial Court .. . .. .. .. .. 0.02 66.0 Planning Commission 0.11 0.11 0.13 0.33 0.15 0.22 0.66 Sub Total 16.71 10.69 12.16 16.85 16.11 16.61 21.05 DEFENCE AND SECURITY 28.0 Police Force 4.27 4.15 3.67 3.81 3.65 4.05 3.51 29.0 Prison Services 2A9 2.14 2.21 2.11 1.84 2.02 1.90 38.0 Defence 11.14 11.11 9.79 9.48 8.91 8.63 8.19 39.0 National Service 1.27 1.31 131 1.51 1.24 1.21 1.34 Sub Total 19.17 18.71 16.98 16.91 15.64 I5.91 14.94 SOCUIL SERVICES 46.0 Education 2.47 2.10 2.74 2.50 2.34 2.77 . 2.99 49.0 Water, 0.38 0.26 037 0.82 0.38 0.52 0.69 52.0 Health 2.75 2.78 3.79 4.76 3.64 3.86 3.83 53.0 Comm. Dev.Worne. Aff. 0.16 0.16 0.21 0.23 0.16 0.19 0.20 65.0 Labour Youth Develop. 0.30 0.22 0.22 0.26 0.17 0.30 0-2' 67.0 Teachers Service Cormn. 0.01 0.02 0.02 0.01 0.03 0.15 0.11 68.0 Science, Tech,& H Ed 3.94 3.40 3.09 4.20 3.47 4.33 4.7 70 & 89 Regions 18A3 17.30 17.21 16.54 17.95 20.02 20.47 Sub Total 28AS 26.23 27.66 29.33 28.15 32.14 3333 ECONOMIC SERVICES 47.0 Works. 1.23 4.08 2.60 0.81 3.30 5.42 3.91 48.0 Lands, Hous.A Urb. Dev. 0.29 0.08 0.08 0.35 0.41 0.40 0.37 58.0 Energy and Minerals. 0.01 0.09 0.12 0.25 0.33 0.35 IA1 62.0 Comm.&Transport 0.10 0.40 0.53 1.66 1.35 1.01 0.92 Sub Total 1.63 4.65 3.34 3.08 5.38 7.19 6A6 PRODUCTIVE 43.0 Agriculture&Livestock 2.37 2.06 1.83 2.35 0.84 1.17 1.3! 44.0 Industris and Trade 0.47 0.23 0.42 0.59 0.38 0.40 OA! 69.0 Tour, Nat. Res. & Env. 1.28 0.39 0.34 1.02 0.84 1.12 0.91 Sub Total 4.11 2.69 2.59 3.96 2.07 2.69 2.71 CONSOLIDATED FUND SERVICE 20.0 State house 0.10 0.14 0.12 0.11 0.11 0.11 0.1 22.0 Public Debt 29.83 36.89 37.15 29.77 32.53 25.35 21.1: Sub Total 29.93 37.03 37.27 29.88 32.6S 2546 21.2 GRAND TOTAL 100.00 100.00 100.00 100.00 100.00 100.00 100.J Source: Table 6a, 6b, and 6c 157 Table 7b: Actual Development Expenditure by Vote as the Share of Total Actual Development Expenditure VOTE VOTE HOLDER 1995/96 1996/97 1997/98 1998/99 1999/00 2000/01 ADMINSTRATION 23.0 Accountant Gcneral 0.00 0.00 0.00 0.00 0.00 0.43 27.0 Registrar of Political Paries 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 3.30 31.0 2nd Vice President 0.00 0.00 0.33 0.74 1.51 0.30 32.0 Civil Service Dept 0.22 0.21 0.20 0.00 7.85 4.30 33.0 Ethics Secretariat 0.00 0.00 0.00 0.00 0.00 0.00 34.0 Foreign Affairs 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 36.0 Civil Service Comm. 0.00 0.00 0.00 0.00 0.00 0.00 37.0 Prime Ministers Office 4.35 8.49 . 7.32 0.09 2.83 0.25 40.0 Judiciary 0.00 #REF! 0.00 0.00 0.00 0.51 41.0 Justice 0.01 0.46 0.03 0.00 0.00 0.00 42.0 Office Of The Speaker 0.15 0.20 0.21 0.00 0.00 0.30 45.0 Exchequer and Audit 0.00 0.00 0.00 0.00 0.00 0.03 50.0 Finance 3.54 37.70 9.00 0.00 1.53 0.9 51.0 Home Affairs . 0.57 0.00 0.00 0.00 0.00 0.00 54.0 Radio Tanzania 2.75 0.00 0.95 0.00 0.25 0.19 55.0 Conmn of Human Rights & Good Gov. .. .. .. 56.0 Regional Adminisstration and Local Gov. .. .. .. 0.00 0.04 8.28 57.0 Defence & National. 6.46 0.21 4.30 0.00 0.00 0.63 59.0 Law Reform. Commi. 0.00 0.00 0.00 0.00 0.00 0.00 60.0 Industrial Court Of Tz 0.00 0.00 0.00 0.00 0.00 0.00 61.0 Electoral Commision 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 64.0 Commercial Court .. .. .. 66.0 Planning Commission 2.07 1.17 1.75 0.55 1.42 1.37 Sub Total 20.90 48.99 24.21 139 18.82 20.40 DEFENCE AND SECURITY 28.0 Police Force 1.02 0.74 0.24 0.00 0.00 0.00 29.0 Prison Services 0.28 0.15 0.00 0.00 0.00 0.00 38.0 Defence 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 Sub Total 1.30 0.89 0.24 0.00 . 0.00 0.00 SOCIAL SERVICES 46.0 Education 4.67 8.73 6.87 9.29 10.79 12.55 49.0 Water, 0.65 0.53 0.60 17.32 8.07 6.94 52.0 Health 0.65 0.92 6.29 10.50 9.59 16.1 53.0 Comm. Dev.Wome. Aff. 0.52 0.00 0.00 1.39 1.26 0.92 65.0 Labour Youth Develop. 15.67 6.52 1.36 0.32 0.61 0.16 67.0 Teachers Service ComnL 0.00 0.00 0.00 0.00 0.00 0.00 68.0 Science, Tech.& H Ed 2.49 4.16 0.89 1.85 1.47 2.66 70 & 89 Regions 13.08 5.10 6.49 8.91 0.00 #REF! Sub Total 37.74 25.96 22.50 49.59 31.78 393 ECONObMC SERVICES 47.0 Works. 23.09 8.33 0.00 17.31 19.38 10.5 48.0 Lands, Hous.,& Urb. Dcv. 0.17 0.00 0.00 0.00 0.00 0.57 58.0 Energy and Minerals. 0.02 3.17 7.70 14.56 9.27 3.35 62.0 Comm.& Tmnsport 0.01 10.18 32.38 8.33 9.41 14.1 Sub Total 23.29 21.67 40.08 40.21 38.06 28.62 PRODUCTIVE 43.0 Agriculture & Livestock 5.93 0.28 6.12 8.60 11.09 8.23 44.0 Industries and Trade 2.89 0.09 0.35 0.11 0.26 0.16 69.0 Tour., Nat. Res. & Env. 7.96 2.12 6.51 0.10 0.00 3.2 Sub Total 16.77 2.49 12.97 8.81 1135 11.6 CONSOLIDATED FUND SERVICE 20.0 State house 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. Sub Total 0.00 0.00 0.00 0.00 0.00 0.00 GRANDTOTAL 100.00 100.00 100.00 100.00 100.00 100. Soure: Table 6a. 6b, and 6c 158 Table 7c: Total Actual Expenditure by Vote as the Share of Total Actual Expenditure VOTE VOTE HOLDER 1995/96 1996/97 1997/98 1998/99 1999/00 2000/01 ADMINISTRATION 23.0 Accountant General 0.00 0.00 0.00 0.00 0.00 1.94 27.0 Registrar of Political Parties 0.00 0.00 0.45 0.29 0.13 0.23 30.0 Presidents Office 1.54 1.58 1.39 1.74 2.32 2.51 31.0 2nd VicePresident 0.04 0.09 0.13 0.19 0.27 0.1 32.0 Civil Service Dept. 0.33 0.34 0.34 0.41 1.19 0.93 33.0 Ethics Secretariat 0.00 0.02 0.02 0.02 0.01 0.02 34.0 Foreign Affairs 2.99 1.96 1.99 2.17 1.68 2.06 35.0 Penm. Comm. Enq. 0.03 0.02 0.03 0.03 0.02 0.03 36.0 Civil Service Comm. 0.01 0.02 0.02 0.02 0.02 0.0 37.0 Prime Ministers Office 1.28 1.16 1.55 0.69 0.78 0.45 40.0 Judiciary 0.61 0.59 0.59 0.59 0.61 0.8 41.0 Justice 0.11 0.15 0.12 0.22 0.09 0.13 42.0 Office Of The Speaker 0.55 0.53 0.49 0.61 0.60 0.5 45.0 Exchequer and Audit 0.09 0.08 0.09 0.09 0.11 0.15 50.0 Finance 3.35 5.58 4.67 5.83 5.08 3.56 51.0 Home Affairs 0.30 0.26 0.23 0.52 0.49 0.4 54.0 Radio Tanzania 0.16 0.05 0.14 0.12 0.14 0.12 55.0 Comm of Human Rights & Good Gov. .. .. .. 56.0 Regional Adminiastration and Local Gov. .. .. .. 0.31 1.69 142 57.0 Defence& National. 0.52 0.27 0.69 0.34 0.24 0.43 59.0 Law Refonn. Commi. - 0.01 0.01 0.01 0.01 0.01 0.02 60.0 Industrial Court Of Tz. 0.01 0.02 0.02 0.02 0.02 0.03 61.0 Electoral Commision 4.74 0.15 0.19 0.00 0.60 2.5 63.0 Local Govt. Servi. Comm. 0.03 0.02 0.02 0.02 0.02 0.02 64.0 Commercial Court .. .. .: 66.0 Plianning Commission 0.19 0.17 0.31 0.36 029 0.3 Sub Total 16.88 13.08 13A7 14.93 1640 17.11 DEFENCE AND SECURITY 28.0 Police Force 4.13 3.94 3.30 3.34 3.26 3.52 29.0 Prison Services 2.39 2.01 1.97 1.84 1.64 1.75 38.0 Defence 10.68 10.41 8.73 .. 7.95 7.4 39.0 National Service 1.22 123 1.16 1.33 1.11 1.05 Sub Total 1.I43 17.60 15.16 14.81 13.95 13.8 SOCLAL SERVICES 46.0 Education 2.56 2.51 3.19 3.35 3.25 4.0 49.0 Water, 0.39 0.28 0.40 2.87 121 2.12 52.0 Health 2.66 2.67 4.06 5.47 428 5.49 53.0 Comm. Dev.Wome. Aff. 0.17 0.15 0.19 0.38 0.28 0.2 65.0 Labour Youth Develop. 0.94 0.61 0.35 0.26 0.22 0.2 67.0 Teachers Service Comm. 0.01 0.01 0.01 0.01 0.02 0.48 68.0 Science, Tech& H Ed 3.88 3.44 2.85 3.91 3.26 3.7 70 & 89 Regions 1821 16.53 16.05 15.59 16.02 17.3 Sub Total 28.84 26.21 27.10 31.84 2.4 33.1 ECONOMIC SERVICES 47.0 Works. 2.13 4.34 2.32 2.86 5.03 6.1 48.0 Lands, Hous.,& Urb. Dev. 0.28 0.07 0.07 0.31 0.37 0.3 58.0 Energy and Minerals. 0.01 0.28 0.95 2.02 1.29 0.7 62.0 Comm.& Transport 0.09 1.01 3.99 2A9 2.21 2.7 Sub Total 2.52 5.71 732 7.68 8.90 10.0 PRODUCTIVE 43.0 Agriculture&Livestock 2.51 1.95 2.29 3.12 1.94 2.11 44.0 Industries and Trade 0.57 0.23 0.42 0.53 0.37 0.3 69.0 Tour., Nat. Rs. & Env. 1.55 0.50 1.01 0.91 0.75 1.4 Sub Total 4.63 2.68 3.72 456 3.07 3 CONSOLIDATED FUND SERVICE 20.0 Statehouse 0.09 0.14 0.11 0.10 0.10 0. 22.0 Public Debt 28.60 34.58 33.12 26.08 29.03 21. Sub Total 28.70 34.72 33.23 26.17 29.13 22 GRAND TOTAL 100.00 100.00 100.00 100.00 100.00 100 Source: Table 6a, 6b, and 6c 159 Table 8a: Actual Recurrent Expenditure as a Percentage of Budgeted Recurrent Expendituer by Vote VOTE VOTE, HOLD)ER 19906% 1996t97 1997198 1998/9 1999/00 2000/01 2001/02* ADMINISTRATION 23.0 Accouatant Genera 100 61 27.0 Regisar of Political Parties 100 100 50 99 75 30.0 Presidents Office 100 100 100 9 99 100 78 31.0 2nd VicePresident 91 100 91 100 100 100 77 32.0 Civil Service Dept. 88 101 101 100 100 100 80 33.0 Ethics Secretaiat 100 99 98 83 100 76 34.0 ForeignAffairs 112 114 115 106 97 120 80 35.0 Penn.Comm.Enq. 93 100 100 99 95 100 36.0 Civil Service Conun. 83 98 111 100 100 ISO 54 37.0 Prime Ministers Office 100 100 63 43 100 100 34 40.0 Judiciary 89 100 98 n 75 100 73 41.0 Justice 77 100 106 90 86 99 94 42.0 Office Of The Speaker 102 99 99 101 100 100 74 45.0 Exchequer and Audit 97 99 99 97 100 99 72 50.0 Finance 64 96 84 102 98 100 18 53.0 Home Affairs 86 98 85 101 83 100 52 54.0 RadioTannina 79 100 100 93 96 74 59 55.0 Comm of Human Rights & Good Gov. 18 56.0 Regional Adminissratio and Local Gov. 44 315 100 52 57.0 Defence&National. 96 100 71 68 100 100 83 59.0 LawRefomL Comi. 61 103 102 99 94 99 47 60.0 Industrial CourtOfTz. 35 100 101 100 88 100 87 61.0 Electoral Commision 242 99 106 9 36 99 71 63.0 Local Govt Servi. Comm. 59 100 100 100 99 97 84 64.0 Commercial Court 41 66.0 Planning Cotmmission 70 100 97 96 70 100 33 Sub Total 106 101 91 92 96 102 42 DEFENCE AND SECURITY 28.0 Police Force 101 100 87 100 92 100 72 29.0 PrisonServices 95 100 88 100 83 100 70 38.0 Defence 100 99 92 101 96 100 70 39.0 National Service 95 100 94 103 90 100 75 Sub Total 99 100 90 101 93 100 71 SOCIAL SERVICES 46.0 Education 100 9 110 100 98 100 72 49.0 Water, 81 100 90 100 100 91 62 52.0 Health 87 U4 100 100 95 99 68 53.0 Comm. Dev.Wome. Aff. 92 100 100 103 62 100 63 65.0 LabourYouthDevelop. 96 100 100 102 51 96 56 67.0 TeachersServiceComm. 76 100 98 100 100 100 70 68.0 Science Tech.& HEd 96 91 92 100 95 100 74 70&89 Regions 91 99 99 107 94 100 77 Sub Totd 92 96 99 104 94 99 74 ECONOMIC SERVICES 47.0 Works. 24 1004 73 128 76 100 67 48.0 Lands. Hous.,& Urb. Dev. 94 73 32 80 100 97 68 58.0 Energy andMinerals. 41 100 91 98 97 100 162 62.0 Cornm.& Tansport 62 100 63 97 I1I 100 64 Sub Total 29 462 69 101 86 100 77 PRODUCTIVE 43.0 Agriculture Livestock 85 99 90 95 60 100 77 44.0 lndutries andTrade 120 99 100 99 93 100 65 69.0 Tour., NaL Res. & Env. 98 99 82 101 85 86 55 SubTotal 92 99 90 97 74 94 66 CONSOLIDATED FUND SERVICE 20.0 Stat house 103 96 100 100 100 100 72 22.0 Public Debt 73 97 122 98 97 91 52 Sub Total 74 97 122 98 97 91 52 GRAND TOTAL 86 101 102 99 94 98 59 Source: Table 6a, 6b, nd 6c 160 Table Sb: Actual Development Expenditure as a Percentage of Budgeted Development Expenditure by Vote VOTE VOTE HOLDER 199596 1996/97 1997/98 1993199 1999/00 2000/01 ADMINISTRATION 23.0 AccoumuctGneml 62 27.0 Regisarof Politicl Parties 30.0 Presidents Office 73 64 26 100 50 31.0 2nd Vice Pesident 15 34 27 6 32.0 CivilServiceDept 23 86 12 0 211 100 33.0 Edtics Scretuiat 34.0 Foign Affairs 0 0 35.0 PemL Comm. Enq. 36.0 Civil Service Comm 37.0 Prime MinistrsOfflce 16 93 35 100 87 9 40.0 Judiciary 0 #REF! 0 too 41.0 Justice 25 95 8 2 42.0 Office Of The Speaker 100 69 37 45.0 Exchequr and Audit 9 50.0 Finsc 2 26 9 0 25 is 51.0 HomeAfirs 15 0 54.0 RadioTannui 100 100 68 99 55.0 Comm of H&n Rigbt & Good Gov. 56.0 Regionl Adminisstrtion and LOl Gav. 0 48 57. Deftwe&NaNral 43 4 355 0 0 100 59.0 Law Refoun. Cormi. 60.0 Indusrbil Cot Of T7. 61.0 Electoa Cninmdsion 0 63.0 Loal Govt ServL Comm. 64.0 Commercial Court 66.0 PlanningCommio 22 32 20 42 120 86 Sub Total 8 29 17 6 45 45 DEFENCE AND SECURJTY 28.0 Police Fome 23 33 55 29.0 PrisoServices I1 12 0 38.0 Defence 0 39.0 National Senrice Sub Total 19 25 34 SOCIAL SERVICES 46.0 Educatior 9 81 73 94 69 #REFI 49.0 Water, 1 15 7 62 28 67 52.0 Healtb 1 9 20 54 44 75 53.0 ComL Dev.Wome. Aft 9 0 0 65 51 90 65.0 lAbourYouth Develp. 121 104 27 21 30 9 67.0 Teacs Service Comm. 68.0 Science,Techb.&HEd 8 56 14 44 31 0 70&89 Regious 12 IS 25 34 0 Sub Tetd 12 35 25 54 31 66 ECONOMIC SERVICES 47.0 Wo*t. 28 38 0 37 58 53 48.0 Lands, Hous,& Urb. Dcv. 3 0 0 0 98 58.0 EneW andMinerals. 2 22 26 67 46 17 62.0 Commh& Tmasport 0 97 100 100 100 100 Sub Totl 20 44 34 53 60 53 PRODUCTIVE 43.0 Agricultureh&Lvestok 100 3 95 41 42 78 44.0 Industieaand Tmde 41 5 79 37 24 65 69.0 Tomw.,NatRe.&Egiv. 65 42 89 1 0 93 Sub Total 67 16 91 31 37 81 CONSOLIDATED PUND SERVICE 20.0 State huse 22.0 Public Debt Sub Tota GRAND TOTAL 14 32 28 46 42 57 Soue: Table 6a. 6b, and 6c 161 Table &c: Actual Total Expenditre as a Percentage of Budgeted Total Expenditure by Vote VOTE VOTE HOLDER 199896 199697 1997i98 199199 1w9900 2000/01 ADMINISTRATION 23.0 ACcountant Gncul 98 27.0 RegistM rofPolidcalParties 100 100 50 99 30.0 Prsidents Office 99 99 97 99 99 85 31.0 2nd Vice President 91 100 38 51 38 19 32.0 Civil Service DctP 82 100 69 91 160 100 33.0 EtuicsSecrciat 100 99 98 83 100 34.0 ForeigpAffirs 112 113 114 106 97 120 35.0 Perm. Comm Enq. 93 100 100 99 95 36.0 CivilServiceComm. 83 98 III 100 100 ISO 37.0 PrimcMimistcrOffice 57 97 45 43 95 59 40.0 Judiciary 89 99 98 99 75 100 41.0 Jstice 76 9 79 el 86 99 42.0 OfficeOfTheSpeaker 102 98 92 101 100 100 45.0 Excheauer mnd Audit 97 99 99 97 100 78 50.0 F;ince 26 44 31 76 89 83 51.0 HomeAffairs 62 98 74 101 83 100 54.0 RadioTanzinia 93 100 100 93 89 78 55.0 Comm of Human Rights & Good Gov. .. .. .. 56.0 Regionl Adhiimbution and Local Gov. .. .. .. 44 73 55 57.0 Defeue&Nc doNal. 59 48 155 56 70 100 59.0 LAwRcformCommL 61 103 102 99 94 9 60.0 IndustrialCourtOfTZ. 35 100 101 100 88 100 61.0 EketotalCommniion 242 99 106 35 9 63.0 LocW GovL SevLCO_OL 59 100 100 100 9 97 64.0 Commercl Court .. .. .. 66.0 Plazuning C'ision 35 52 28 77 90 92 SobTal 66 64 50 80 S4 Is DEFENCE AND SECURITY 28.0 PoliceForce 98 98 87 100 92 I0O 29.0 PrionSenvic 92 97 87 100 83 100 38.0 Defcnce 100 99 92 96 100 39.0 NalonalSeAvicc 95 100 94 103 90 100 Sb Total 98 99 S0 101 93 100 SOCIAL SERVICES 46.0 Education 58 95 98 98 85 95 49.0 Water, 14 61 29 69 35 71 52.0 Helth 53 71 60 83 74 88 53.0 Conu. Dev.Wmse.L A 43 27 30 81 56 96 65.0 LalborYoult Dewelop. 112 102 46 64 43 54 67.0 Teachen Service Comm. 76 100 98 100 100 377 68.0 Science,Teh.& HEd 75 87 78 93 86 91 70&89 Regiou 77 91 88 92 80 W0 SubTotl 68 87 78 88 75 92 ECONOMIC SERVICES 47.0 Works. 26 248 25 45 67 83 48.0 Lsads, Hou..& Ub. Dev. 57 35 20 80 88 97 58.0 Energy and MineralS. 22 28 29 70 53 26 62.0 Conimh.& Ttu,sport 7 98 93 98 106 100 Sub Total 25 143 43 63 72 7S PRODUCTIVE 43.0 Agiculture&Liveatock 86 79 91 65 47 87 44.0 Induwi ed dTrade 86 68 98 95 76 97 69.0 Tour.,NaLRer hEDv. 88 73 87 52 59 88 Sob Tol 87 77 91 64 5S 88 CONSOLIDATED FUND SERVICE 20.0 Sttehouse 103 96 100 100 100 100 22.0 PublicDebt 73 97 122 98 97 Sl Sob Total 74 97 122 98 97 91 GRAND TOTAL 71 90 79 86 83 89 Source: Table 6 b, 6b nd 6c 162 Table 9a: Recurrent Expenditure - Regions (in Bill. Tsh.) VOTE VOTE 1996/97 1996/97 1997/98 1997/98 1998199 1998/99 1999/00 1999/00 2000/01 2000/01 2001/02 2001/02* HOLPER Budget Actual Budget Actual Budget Actual Budget Actual Budget Actual Budget Actual 70.0 Arusha 6.3 6.3 7.6 7.1 7.7 8.4 12.2 11.7 15.1 15.1 16.2 13.3 71.0 Coast 3.1 3.1 3.7 3.6 3.8 4.1 5.9 5.7 7.1 7.0 9.0 6.4 72.0 Dodom 5.2 5.2 6.1 6.1 6.1 6.5 8.9 8.9 10.5 10.5 11.4 9.3 73.0 Iringa 5.6 5.3 6.6 6.5 6.8 7.3 7.3 2.3 11.4 11.3 12.0 10.2 74.0 Kigomt 3.7 3.7 4.4 4.4 4.7 4.9 5.1 6.4 7.9 7.9 8.8 6.1 75.0 Kilimanr iro 7.4 7.4 8.6 8.6 8.8 9.3 12.7 12.6 15.2 15.2 15.9 10.9 76.0 Lindi 3.0 3.0 3.5 3.5 3.6 3.9 5.1 5.1 6.4 6.3 7.1 5.7 77.0 Mara 5.0 5.0 6.1 5.9 6.1 6.7 8.6 8.6 11.0 11.0 9.1 7.8 78.0 Mbeya 6.3 6.3 8.0 7.9 8.2 8.5 Il.5 11.4 13.4 13.3 14.5 11.1 79.0 Morogorv 5.6 5.6 6.7 6.6 6.8 7.4 9.9 9.9 11.5 11.5 12.3 10.2 80.0 Mtwara 3.7 3.7 4.3 4.2 4.4 4.7 6.4 6.4 7.7 7.7 8.4 6.3 81.0 Mwanza 6.9 7.0 8.4 5.8 8.4 9.2 12.3 12.2 13.9 13.9 15.2 11.8 82.0 Ruvuma 4.4 4.4 5.2 5.1 5.2 5.6 7.6 7.6 8.8 8.8 10.1 8.4 83.0 Shinyanga 5.4 5.5 6.4 5.9 6.6 7.3 7.3 9.6 11.3 11.3 12.5 8.5 84.0 Singida 3.6 3.6 4.1 4.1 4.2 4.4 5.9 5.8 7.0 7.0 7.7 4.7 85.0 Tabora 4.4 4.3 5.1 5.0 5.3 5.5 7.1 7.1 8.4 8.4 9.8 7.5 86.0 Tanga 5.9 5.9 7.0 6.9 7.1 7.5 10.1 10.1 12.0 12.0 12.8 7.6 87.0 Kagera 5.2 5.2 6.2 6.1 6.3 6.8 9.0 8.9 10.8 10.7 12.7 7.5 88.0 DSalaam 5.6 5.6 6.4 6.4 6.7 6.8 10.0 2.7 11.5 11.4 12.1 10.3 89.0 Rukwa 3.2 3.2 3.8 3.5 4.0 4.1 5.6 5.5 6.4 6.4 7.2 6.1 Total 99.4 99.1 118.3 113.2 120.9 129.0 168.6 358.2 207.5 206.8 224.8 169.7 Note: * Data for FY02-all votes are from the Expenditure Flash Reports (July 2001 - March 2002) Source: MoF, Appropriations Accounts and Expenditure Flash Report Table 9b: Development Expenditure - Regions (in Bill. Tsh.) VOTE VOTE HOLDER 1995/96 1996/97 1996/97 1997/98 1997198 1998/99 1998/99 1999/00 1999/00 2000/01 2000/01 2001/02 2001/02* Actual Budget Actual Budget Actual Budget Actual Budget Actual Budget Actual Budget Actual 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 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 72.0 Dodoma 0.0 0.8 0.0 0.5 0.1 2.4 0.1 2.4 0.0 2.6 1.0 0.3 73.0 Iringa (.0 0.9 0.0 1.8 0.1 2.8 1.3 2.8 2.3 5.0 0.4 2.6 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 75.0 Kilimanjaro 0.0 0.2 0.0 0.2 0.1 0.3 0.0 0.3 0.0 0.4 0.4 0.4 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 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 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 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 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 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 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 83.0 Shinyanga 0.0 0.3 0.0 1.6 0.1 0.3 0.0 0.3 0.0 0.3 0.3 0.4 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 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 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 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 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 0.2 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 Total 2.4 10.9 1.9 21.7 5.4 29.2 9.8 29.2 9.7 37.1 20.9 35.2 Note: * Data for FY02-all votes are from the Expenditure Flash Reports (July 2001 - March 2002) Source: MoF, Appropriations Accounts and Expenditure Flash Report Table 9c: Total Expenditure - Regions (in Bill. Tsh.) VOTE VOTE 1996/97 1996/97 1997/98 1997/98 1998/99 1998/99 1999/00 1999/00 2000101 2000/01 2001/02 2001/02* HOLDER Budget Actual Budget Actual Budget Actual Budget Actual Budget Actual Budget Actual 70.0 Arusha 7.9 6.3 10.5 7.2 11.2 8.5 15.7 11.7 18.6 17.8 19.3 71.0 Coast 3.6 3.2 4.5 3.8 4.3 4.4 6.4 6.0 7.8 7.7 10.6 72.0 Dodoma 6.0 5.2 6.6 6.2 8.5 6.6 11.4 8.9 13.1 11.5 11.7 73.0 Iringa 6.4 5.3 8.5 6.6 9.6 8.7 10.1 4.6 16.4 11.7 14.6 74.0 Kigoma 4.2 3.8 4.7 4.5 6.1 4.9 6.5 6.4 8.1 8.1 9.2 75.0 Kilimanjaro 7.6 7.4 8.8 8.7 9.0 9.3 13.0 12.6 15.6 15.6 16.3 76.0 Lindi 3.3 3.0 4.3 3.6 4.0 4.1 5.5 5.1 7.7 6.6 8.9 77.0 Mara 5.5 5.3 6.8 7.3 7.9 9.3 10.3 10.0 14.5 12.4 13.6 78.0 Mbeya 6.7 6.3 8.4 8.0 8.7 8.6 11.9 11.4 14.1 12.0 15.1 79.0 Morogoro 5.9 5.8 8.4 6.9 7.4 7.5 10.6 9.9 14.7 8.0 15.6 80.0 Mtwara 4.2 3.7 5.1 4.3 5.0 4.8 7.0 6.4 9.0 14.2 9.7 81.0 Mwanza 7.7 7.6 9.7 - 6.6 10.2 10.0 14.0 12.9 16.1 16.1 18.1 82.0 Ruvuma 4.6 4.5 5.5 5.2 5.6 5.8 8.0 7.7 9.3 9.3 10.4 83.0 Shinyanga 5.7 5.5 8.0 6.0 6.9 7.3 7.6 9.6 11.6 11.6 12.9 84.0 Singida 4.0 3.6 4.6 4.2 4.6 4.5 6.3 5.8 7.9 7.4 8.5 85.0 Tabora 4.6 4.4 5.3 5.1 5.6 5.6 7.4 7.1 8.8 8.8 10.2 86.0 Tanga 6.7 6.0 8.4 7.0 8.5 7.6 11.6 11.0 12.9 12.9 14.4 87.0 Kagera 5.8 5.3 8.4 7.3 11.9 10.3 14.6 12.6 20.1 18.4 21.1 88.0 D'Salaam 6.3 5.6 9.5 6.4 10.7 ' 6.9 14.0 2.7 11.7 11.6 12.3 89.0 Rukwa 3.7 3.4 4.2 3.7 4.3 4.2 5.9 5.5 6.6 6.6 7.5 Total 110.3 101.0 140.0 118.6 130.1 138.8 197.8 167.8 244.7 228.4 260.0 Note: * Data for FY02-all votes are from the Expenditure Flash Reports (July 2001 - March 2002) Source: MoF, Appropriations Accounts and Expenditure Flash Report Table 1Oa: Actual Recurrent Expenditure as a Percentage of Budgeted Recurrent Expenditure - Regions VOTE VOTE HOLDER 1995/96 1996/97 1997/98 1998/99 1999/00 2000/01 2001/02* 70.0 Amsha 91% 100% 93% 108% 96% 100% 82% 71.0 Coast 80% 100% 98% 108% 97% 99% 71% 72.0 Dodoma 93% 100% 99% 106% 100% 100% 82% 73.0 Iringa 78% 95% 99% 108% 32% 100% 85% 74.0 Kigoma 87% 100% 99% 104% 124% 100% 69% 75.0 Kilimanjaro 110% 100% 99% 106% 99% 100% 69% 76.0 Lindi 46% 100% 99% 109% 99% 99% 80% 77.0 Mara 94% 100% 98% 110% 100% 100% 86% 78.0 Mbeya 97% 99% 99% 104% 99% 99% 77% 79.0 Morogoro 93% 100% 99% 110% 99% 100% 83% 80.0 Mtwara 91% 100% 99% 105% 100% 100% 75% 81.0 Mwanza 900/0 102% 70% 109% 100% 100% 78% 82.0 Ruvuma 96% 100% 98% 106% 99% 100% 83% 83.0 Shinyanga 91% 100% 92% 112% 132% 100% 68% 84.0 Singids 93% 100% 99% 106% 99% 99% 61% 85.0 Tabora 95% 99% 99% 102% 99% 100% 77% 86.0 Tanga 92% 100% 98% 106% 99% 100% 59% 87.0 Kagera 105% 100%/0 99% 109% 99% 99% 59% 88.0 D'Salaam 98% 100% 99% 101% 27% 99% 85% 89.0 Rukwa 88% 100% 92% 104% 98% 100% 85% Total 91% 100% 96% 107% 94% 100% 75%1 Source: Table 9a. 166 Table lOb: Actual Development Expenditure as a Percentage of Budgeted Development Expenditure - Regions VOTE VOTE HOLDER 1995t96 1996/97 1997/98 1998t99 1999/00 2000/01 70.0 Auisha 1% 1% 4% 3% 1% 77% 71.0 Coat 93% 18% 19% 64% 56% 100% 72.0 Dodoma 1% 1% 23% 4% 0% 39% 73.0 Iringa 1% 2% 5% 47% 83% 7% 74.0 Kigonma 21% 14% 38% #REFI 0% 100% 75.0 Kilimanjaro 1% 17% 58% #REFI 2% 100% 76.0 Lindi 1% 12% 13% 25% 2% 23% 77.0 Mara 52% 66% 192% .149% 85% 40% 78.0 Mbeya 13% 14% 25% 23% 9% 71% 79.0 Morogoro 15% 57% 18% 17% 0% 10% 80.0 Mtwara 7/o 4% 12% 18% 1% 22% 81.0 Mwanza 42% 73% 56% 46% 37% 1000% 82.0 Ruvuma 24% 67% 34% 58% 41% 100% 83.0 Shinyanga 3% 10% 7% #REFI 0% 100% 84.0 Singida 2% 5% 21% 25% 1% 500/o 85.0 Tabora 4% 8% 41% 44% 2% 100% 86.0 Tanga 1% 3% 7% 7% 62% 1000% 87.0 Kagea 8% 18% 56% 62% 67% 83% 88.0 D'Salaa 2% 3% 2% 2% 0% 100% 89.0 Rukwa 46% 40% 46% 36% 2% 100% Total 12% 18% 25% 34% 33% 56% Source: Table 9b. 167 'Table lOc: Actual Total Expenditure asa Percentage o(Budgeted Total Expenditure - Regions VOTE VOTE HOLDER 1995/96 1996/97 1997/98 1998/99 1999/00 2000/01 70.0 Arusha 78% 80% 69% 75% 75% 96% 71.0 Coast 82% 88% 84% 103% 93% 99% 72.0 Dodoma 74% 87% 93% 77% 79% 88% 73.0 Iringa 59% 83% 78% 90% 46% 71% 74.0 Kigoma 72% 90% 95% 80% 97% 100% 75.0 Kilimanjaro 100% 98% 99% 103% 97% 100% 76.0 Lindi 29% 91% 84% 100% 91% 86% 77.0 Mara 90% 97/O 108% 118% 97% 86% 78.0 Mbeya 87% 95% 96% 100% 96% 85% 79.0 Morogoro 86% 97% 83% 102% 94% 54% 80.0 Mtwara 87% 89% 85% 94% 91% 158% 81.0 Mwanza 88% 99% 68% 98% 92% 100% 82.0 Ruvuma 87% 99% 94% 103% 96% 100% 83.0 Shinyanga 83% 96% 75% 106% 126% 100% 84.0 Singida 79% 90% 91% 99% 92% 94% 85.0 Tabora 88% 94% 97% 100% 96% 100% 86.0 Tanga 72% 89% 83% 89% 95% 100% 87.0 Kagera 26% 91% 87% 86% 87% 92% 88.0 D'Salaam 73% 89% 68% 64% 19% 99% 89.0 Rukwa 74% 92% 88% 99% 93% 100% Totad 77% 92% 85% 92% 85% 93% Source: Table 9c. 168 ---------- ------ --~~- - - -# ji | IiEH!,RRO2Sat 0t %f|31 l4 !§ } i I a a_,,sg ;5f - iI .0 « ""t} 94 -58 rg a~~0 | 15nB*q8*g - l g@}gE! l 3! !M a§; I I 1~~~~~~ I S~~~~~~~~~~~~~~~~~~~ .1 O O O O, " . I- __________~~~~~~~~~~~~~ g pLiaZAR- t ' -I-- -"o$-na$wo§oII gelI A Mus I11 I g IIIs fI W s gI g I ~~~ I 1 S - ° 4°°°° °° iz °° f°°°S S i°@ °d ° f S dw. -dW d 3 144 *1 0 ji 3~~~~~~~~~~~~~~~~~~~~~~~~~~ity Table 12a: Regions - Recurrent Expenditure Before and After Reallocation, FYOO 19/00 1999/00 199/00 1999100 1999/00 1999/00 E VOTE Approved Reallocation Budget After Actual Actual Expenditure Budget After VOTE HOLDER Budget Mll Tsb) Reallocation Expenditure asa % of Budgeted Reaflocatlon as a (In Mlil. Tsb.) (In o. . n MIII. Tsh.) (in MIt. Tsh.) Expenditure After % of Approved Reallocation Budget 70.0 Arusha 9,327 2,856 12,184 11,673 96% 131% 71.0 Coast 4,948 949 5,896 5,699 97% 119% 72.0 Dodoma 6,539 2,411 8,950 8,925 1000% 137% 73.0 Iringa .. 74.0 Kigoma 5,112 1,334 6,447 6,362 99% 126% 75.0 Kilinjaro 9,704 3,027 12,732 12,584 99% 131% 76.0 Lindi 4,179 947 5,125 5,053 99% 123% 77.0 Mara 6,399 2,215 8,614 8,586 1000/% 135% 78.0 Mbeya 8,833 2,648 11,480 11,361 99% 1300/o 79.0 Morogoro 7,589 2,349 9,938 9,865 99% 131% 80.0 Mtwara 5,074 1,333 6,406 6,401 1000/% 126% 81.0 Mwanza 9,074 3,180 12,254 12,240 100% 135% 82.0 Ruvumna 6,046 1,588 7,634 7,566 99% 126% 83.0 Shinyanga 7,263 2,708 9,971 9,581 96% 137% 84.0 Singida 4,779 1,078 5,857 5,775 99% 123% 85.0 Tabora 5,545 1,582 7,128 7,066 99% 129% 86.0 Tanga 7,876 2,247 10,123 10,056 99% 129% 87.0 Kagera 6,908 2,060 8,968 8,873 99% 130% 88.0 DSalaam 7,489 2,543 10,032 2,715 27% 134% 89.0 Rukwa 4,085 1,498 5,584 5,454 98% 137% Total 126,770 38,553 165,323 ISS,833 94% 130% Source: MoF, Appropriations Accounts 171 Table 12b: Regions - Recurrent Expenditure Before and After Reallocation, FY01 2000/01 2000/01 2000/01 2000/01 2000/01 2000/01 VOTE VOTE Approved Reallocaton Budget After Actual Actual Expenditure Budget After VOTE HOLDER Budget Re Mil. Tsb.) Rtallocabon Expenditure s a % of Budgeted Reallocatlon as a (in Mil. Tsb.) n (in NMIl. Tsb.) (In MiIl. Tsb.) Expenditure After % of Approved Reallocation Budget 70.0 Arusha 14,314 834 15,147 15,134 100% 106% 71.0 Coast 6,877 221 7,097 7,024 99% 103% 72.0 Dodoma 10,083 466 10,549 10,542 100% 105% 73.0 Iringa 11,102 263 11,366 11,322 100% 102% 74.0 Kigoma 7,439 437 7,876 7,876 100% 106% 75.0 Kilirnanjaro 14,380 850 15,230 15,230 100%/0 106% 76.0 Lindi 6,232 203 6,435 6,346 99% 103% 77.0 Mara 9,877 1,138 11,015 11,009 100% 112% 78.0 Mbeya 13,256 170 13,427 13,251 99% 101% 79.0 Morogoro 11,231 312 11,543 11,500 100% 103% 80.0 Mtwara 7,479 256 7,735 7,703 100% 103% 81.0 Mwanza 13,392 541 13,933 13,946 100% 104% 82.0 Ruvuma 8,695 144 8,839 8,837 100% 102% 83.0 Shinyanga 10,911 358 11,269 11,267 100% 103% 84.0 Singida 6,943 99 7,042 6,990 99% 101% 85.0 Tabora 8,325 94 8,419 8,422 100% 101% 86.0 Tanga 11,585 387 11,972 11,972 100%/0 103% 87.0 Kagera 10,149 603 10,752 10,690 99%/0 106% 88.0 D'Salaam 10,386 1,137 11,524 11,388 99% 111% 89.0 Rukwa 6,223 147 6,370 6,365 100%/0 102% Total 198,880 8,662 207,542 206,814 100% 104% Source: MoF, Appropriations Accounts 172 I IMAGING Report No.: 24475 TA Type: ER