Report No. 23812-TA The United Republic of Tanzania Public Expenditure Review October 2001 Government of Tanzania Tanzania PER Working Group The World Bank PREM 2, Africa Region Document of the World Bank GOVERNMENT FISCAL YEAR FY02 = 2001/02 = July 1, 2001 to June 30, 2002 CURRENCY EQUIVALENTS Currency Unit = Tanzanian Shilling (T Sh) Interbank Market mid-rate: US$1.00 = T Sh 890 (July 2001) ABBREVIATIONS AND ACRONYMS AfDB African Development Bank MOF Ministry of Finance AIDS Acquired Immune Deficiency Syndrome MOH Ministry of Health BEMP Basic Education Master Plan MSTHE Ministry of Science, Technology and BOP Balance of Payment Higher Education BOT Bank of Tanzania MTEF Medium Term Expenditure Framework CBMS Cash Budget Management Systems NBC National Bank of Commerce CFAA Country Financial Accountability NFA Net Foreign Assets Assessment NGO non governmental organization CG Consultative Group O&E organizational and efficiency (reviews) CSD Civil Service Department O&M operations and maintenance CSRP Civil Service Reform Programme OC other charges DANIDA Danish International Development OECD Organization for Economic Cooperation Agency and Development DEO District Education Officer PE personnel emoluments DFID Department for International Development PER Public Expenditure Review DMO District Medical Officer PHC Primary Health Care DMT District, Municipal and Town Councils PORALG President's Office - Regional Authorities ESP Education Sector Programme and Local Government ESRF Economic and Social Research Foundation PRBS Poverty Reduction Budget Support EU European Union PSWB Public Sector Wage Bill FAO Food and Agriculture Organization PRSP Poverty Reduction Strategy Paper FY Fiscal Year RAS Regional Administration Secretariat GDP gross domestic product REPOA Research on Poverty Alleviation GNP gross national product SAC Structural Adjustment Credit GOT Government of Tanzania SASE Selective Accelerated Salary Enhancement HIPC Highly Indebted Poor Countries Debt SASP Structural Adjustment Support Program Initiative SDC Swiss Development Cooperation HIV Human Immuno-deficiency Virus SDP Sector Development Program IDA International Development Association SPA Special Program of Assistance IFEM Inter-bank Foreign Exchange Market SSA Sub-Saharan Africa IFMS Integrated Financial Management System STD Sexually Transmitted Disease IMF International Monetary Fund TANESCO Tanzania Electric Supply Company LA Local Authority TAS Tanzania Assistance Strategy LCC Local Cost Compensation TB Treasury Bill LGRP Local Government Reform Program TRA Tanzania Revenue Authority LPO Local Purchase Order UDSM University of Dar es Salaam MAC Ministry of Agriculture and Cooperatives UNDP United Nations Development Program MOEC Ministry of Education and Culture VAT Value Added Tax Vice President: Callisto Madavo Director: James W. Adams Sector Manager: Frederick Kilby Task Team Leader: Benno Ndulu TABLE OF CONTENTS Executive Summary ........ .....................................................i 1. Objectives, Methodology and Context ofthe FY01 PER .1 2. Evaluation of the Tanzania PER Process Since FY98 .6 3. Review of Recent Macro Performance .15 4. Review of Budget Performance .18 5. Budget Sustainability and Full Financing of the Priority Sectors .32 6. Revenue Issues in the Longer-Run .36 7. Recommendation .40 8. The Way Forward .51 Postscript .58 Annex 1: Proceedings of the PER FY01 Consultative Meeting 60 Annex 2: Aspects of budget sustainability ............................................................ 74 Annex 3: Fiscal Sustainability ......................................... 77 Annex 4: Data ............................................................ 86 List of Tables: Table 2.1: Studies and activities carried out under the FYO IPER ........................................... 8 Table 4.1 Government Revenue and External Grants (as % of GDP), FY96 - FY01 ... 19 Table 4.2: Foreign Inflows - Grants and Loans (as % of GDP), FY96 - FY01 .. 19 Table 4.3: Government Expenditures (as % of GDP), FY96 - FY01 . . 20 Table 4.4: Civil Service Employment, FY98-FYOI (December of each year) .. 21 Table 4.5: Civil Service Average Salaries, FY97-FYOO . ....................................................... 21 Table 4.6: Financing of the Fiscal Deficit (% of GDP), FY96 - FY01 . . 22 Table 4.7: Composition of Public Expenditures (as % of GDP), FY96-FYO1 .. 23 Table 4.8: Sectoral Recurrent Expenditures (actuals, as a %age of GDP) . . 23 Table 4.9: Social Sector Recurrent Expenditures (actuals, as a %age of GDP) .................... 24 Table 4.10: Tanzania: Priority Sector Spending - Other Charges . ........................................ 25 Table 4.11: Sectoral Development Expenditures (actuals, as a %age of GDP) ..................... 26 Table 4.12: Contingency allocation retained by Ministry of Finance .................................... 27 Table 4.13: Actual Reallocations to PE and OC .................................................................. 28 Table 4.14: Reallocations and Expenditure Outturn, FY00 ................................................... 28 Table 4.15: Development Expenditures by Sector, Actual Expenditures as a Share of Budgeted Expenditures, FY96-FYOO ................................................................ 29 Table 7.1: Structure of Consolidated Arrears- By Category July 1998 - December 2000 ... 45 Table 7.2: Consolidated Arrears- As a Proportion of Total Arrears July 1998 - December. 2000 .................................................................. 46 Table 8.1: Summary of Impact of Public Expenditure Process on Public Expenditure Management in Tanzania ...................................... 56 Annex Tables Table 1: Tanzania Microeconomic Indicators .............................................. 87 Table 2: Balance of Payment .............................................. 88 Table 3: Summary of Central Government Operations ............................................. 89 Table 4: Budget Frame for 1999/2000-2003/04 .............................................. 90 Table 5a: Budget Frame for 1995/96-2003/04 .............................................. 91 Table 5b: Budget Frame for 1995/96-2003/04 % of GDP ....................................... 92 Table 6a: Sectoral-Recurrent Expenditure 1995/1996-2000/01 .................................. 93 Table 6b: Sectoral-Development Expenditure .............................................. 94 Table 6c: Sectoral-Total Expenditure .............................................. 95 Table7a: Actual Recurrent Expenditure by Vote as the Share of Total Recurrent Expenditure FY96-FYOI ....................................... 96 Table7b: Actual Development Expenditure by Vote as the Share of Total Development Expenditure FT96-FYO0 ....................................... 97 Table 8a: Actual Recurrent Expenditure as a Percentage of Budgeted Expenditure by Vote FY95-FYO1 ...................................... 98 Table 8b: Actual Development Expenditure as a Percentage of Budgeted Expenditure by Vote FY95-FY0I .............................................. 99 Table 9a: Recurrent Budgeted Expenditure-Regions FY96-FYO0 ............................... 100 Table 9b: Budgeted Development Expenditure-Regions FY96-FYO0 .......................... 101 Table 9c: Budgeted Total Expenditure-Regions ................................................... 102 Table I Oa: Actual Recurrent Expenditure as Percentage of Budgeted Recurrent Expenditure- Regions FY96-FYO I ............................................................... 103 Table I Ob: Actual Development Expenditure as Percentage of Budgeted Development Expenditure-Region FY96-FYO I ....................................................... 104 Table 1 Oc: Actual Total Expenditure as Percentage of Budgeted Total Expenditure-Regions FY96-FYO1 ............................................................... 105 Table I 1: Sectoral-Recurrent Expenditure Before and After Reallocation ..................... 106 Table 12: Region-Recurrent Expenditure Reallocations FY00 ................................... 107 List of Figures Figure 4.1: Development Expenditure as a share of GDP, FY96-FYO I ........ ................ 21 Figure 4.2: External Project Support, Average FY99 and FYOO ................................. 26 Figure 7.1: Monthly Fiscal Deficit / Surplus ........................................................ 42 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 intends to fulfill two purposes. Firstly, it puts on record the activities and achievements of the work of the PER Working Group during FYO. Secondly, it serves as a public record of the findings of the external review of fiscal developments, 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 2001 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 2000 which was led by Benno Ndulu (mission leader, AFTP2) and consisted of Robert Utz, Sumana Dhar (AFTP2), Vedasto Rwechungura (AFTPS), Philip Mpango, Hamisi Mwinyimvua (consultants, University of Dar es Salaam), Ben Tarimo (consultant), Frans van Rijn (Netherlands Embassy), Tone Tinnes (Norwegian Embassy), Torben Lindqvist (Danish Embassy), Olivier Burki (Swiss Development Cooperation), Fiona Shera (DFID) and Amon Manyama (UNDP). A follow-up mission in April/May 2001 was launched to assist the Government in the preparation of the cross-sector MTEF and to present the findings of the main mission at the Consultative PER meeting held in Dar es Salaam. In addition to the work undertaken in the context of these two missions, the report also draws on work commissioned by the PER Working Group on fiscal sustainability carried out by David Bevan and by Emerging Markets Economics Consultants. 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. 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 THE FY01 PER PROCESS Tanzania has been conducting annual Public Expenditure Reviews (PERs) since FY98. The two principal objectives of the PER process are to support the budget process in Tanzania and to provide peer review on fiscal developments to the Government of Tanzania and to make the findings available to a wide range of stakeholders. The process is government led and benefits from wide participation within government, by donor agencies, research institutes, NGOs, and the private sector. The PER working group, which meets bi-weekly, is the focal point of the PER process. It develops the annual program of work and supervises the implementation of specific activities. During FY01, these activities included studies on the following issues: expenditure tracking in key poverty sectors, fiscal sustainability and local government indebtedness. The PER working group also supports the preparation of sectoral PERs and Medium Term Expenditure Frameworks for the priority sectors (education, health, roads, agriculture, justice, water, HIV/AIDS, lands). The PER macro sub-group provides inputs to the budget guidelines committee on the macro framework, overall prioritization, collects information on donor support as a key element of the government's resource envelope, and contributes to the preparation of the cross-sector medium term expenditure framework. As an integral part of the PER process, the World Bank leads the external evaluation of fiscal performance, covering macro-fiscal development, strategic allocation issues, as well as budget management issues. This report covers mainly the findings of the extemal evaluation covering developments during FY00 and the first eight months of FY01, while sectoral issues are covered in individual sector reports prepared by the sector working groups. The findings of the external evaluation have been presented at the PER consultative meeting which took place in May 2001 and served as background to government's presentation of the medium term expenditure framework for the period FY02-FY04 to a wide range of stakeholders. BUDGET PERFORMANCE FY00 AND FIRST EIGHT MONTHS OF FY01 During FY00 domestic revenue remained at the same low level as in the previous year, i.e., at around 11.5 percent of GDP. The first eight months of FY01 saw a slight improvement in the revenue situation with domestic revenue increasing on an annualized basis to 12 %. The improvement in the revenue situation is mainly due to increased revenue from VAT on petrol products and increased income tax revenue. Aid inflows in the form of grants and concessional loans also increased in FY00 to 4.8 percent of GDP. The first eight months of FY01 saw a further increase to 5.5 % of GDP on an annualized basis. On the expenditure side, expenditures increased in FY00 by one percentage point to 15.9 percent of GDP. Expenditures during the first eight months of FY01 were slightly lower in magnitude due mainly to reduced development expenditure. The increase in expenditures in FY00 was evenly divided between providing funding for the implementation of the first stage of pay reform and enhancing expenditures on other charges. The first eight months of FY01 saw a decline in the share of expenditures on wages and salaries, as no pay awards were granted. Expenditures on OC on the other hand increased quite significantly. Despite the significant increases in spending on OC in FY00 and FY01, expenditures on OC were below the budgeted amounts, pointing towards the continued disconnect between budgets and actual expenditures. Over the past few years, some progress has been made in enhancing funding for priority activities in the areas of primary education, health care, roads, and water, with most of the increased spending going to the districts. Between FY96 and FY99, recurrent expenditures on the social sectors increased from 3.5 percent of GDP to 3.7 percent of GDP. Expenditures for the social sectors declined slightly in FY00 to 3.6% of GDP. This share increased during the first quarter of FYOI to 4.1 % of GDP. However, spending on administration increased even faster, from 1.4 percent of GDP in FY97 to 2.1 percent of GDP in FY00. Expenditures on defence and security have declined from 2.5% of GDP in FY97 to 2.0% in FY00. Progress has been made on bringing a greater share of donor financed development spending into the budget. Recorded Development expenditures in the appropriation accounts increased from 0.9 percent of GDP in FY97 to 1.8 percent of GDP in FY99 and 1.5 percent of GDP in FY00. Correspondingly, actual development expenditures as a share of budgeted development expenditures have increased from 6.1 percent in FY97 to 52% in FY00. However, further work needs to be done to improve the capturing of donor flows in the budget. Deviations between the budget presented to the National Assembly and expenditure outturns remain significant, especially for non-priority sectors. The analysis in this report decomposes deviations into (a) deviations due to reallocations and (b) over or under spending by spending units. Reallocations occurred mainly to distribute funds retained at the Ministry of Finance for "special expenditures" or for contingencies. These reallocations are of concern because they considerably change the sectoral allocations approved by Parliament during the budget session. Significant underspending occurred during FYOI because resource availability from domestic and foreign sources was significantly below the budgeted amounts. Priority sectors were protected from resource shortfalls and received their full allocations. However, continued efforts to reduce such budget deviations will be necessary, including both improvements in budget projections but also consideration of financing options to absorb resource shocks. Contingency funds retained by the Ministry of Finance are substantial. Although retaining unallocated funds for contingencies is important in an environment of large and unpredictable shocks impacting on the budget, efforts should be made to enhance ex-ante transparency in determining the size and allocation of these contingency funds. The main problem of public expenditure management persists, which is the mismatch between available resources and the intended scope of Government activities. By preventing access to deficit financing, the cash budget system highlights this mismatch, resulting in permanent under funding of non-priority spending units which renders them virtually incapable of providing any services. The mismatch between revenue and intended scope of activities also becomes visible in symptoms such as the accumulation of arrears. Priority sectors are also critically under funded. If the poverty reduction targets laid out in the PRSP are to be achieved, additional resources becoming available will need to be devoted to the priority sectors. Possibilities for widening the resource envelope are limited. The structure of the economy makes it difficult to raise more revenue, although the revenue potential of the growth sectors, i.e. mining and tourism, is not yet fully exploited. With Tanzania already being one of the main recipients of foreign aid in the region, the potential for further increases in aid inflows seems limited. Given the limited prospects for expanding the resource envelope and the huge unmet needs of priority sectors for poverty reduction and growth, bridging the gap between available resources and funding requirements of the current government structure can only be achieved if the scope of government is brought in line with the medium term resource availability. ii Finally, improving access and transparency of information on fiscal developments is important to support other efforts to enhance accountability and transparency in the public sector. MANAGING STRESSES FROM CASH BUDGET SYSTEM There are three main types of budgetary stresses that have been identified in respect of under- funding of budgets and contributing to pressures for undermining prudent budget management. The first relates to the overall amount disbursed being far short of requirements for effective delivery of services. Frequently this problem has been addressed in terms of enlarging the overall resource envelope and sector as well as sub-sector prioritization. However, the budget stress is accentuated by the fact that the budget preparation stage has typically overlooked completeness in activity prioritization (sub-items). Currently statutory payments (CFS) and personal emoluments (PE) are treated as first charge. Two more first-charge type of expenditures, if under-funded, often lead to commitments being made outside the budget and build up of arrears. These are utilities (power, water, telecommunications) and "automatic catering" such as food needs of prisons, security forces, boarding schools and hospitals. This may not be an exhaustive list but together they account for about 37 percent of all arrears built up. For the former, until such time as prepaid -service systems are installed the commitment is essentially automatic. Recommendation 1. The report recommends that the Government reviews the classification of first charge expenditures for completeness and ensure that the guidelines for FY02 include instructions for fully funding these. All spending units should provide reasonable estimates of these in their MTEF and budget submissions. The hierarchy in prioritization then would be: (i) statutory expenditures and other first charge sub-items, (ii) Priority sectors and sub-sectors (iii) Other sectors and sub-items. The second dimension of the under-funding problem relates to the mismatch between the monthly cash requirements and exchequer releases - amounts and timing. The lumpiness of some expenditure categories such as road maintenance, pension funds, and examinations present major stresses in the overall functioning of the Government budget when they occur without provision for cash flow gap smoothing. The pattern of expenditures through the year varies greatly across sectors to render a general rule of releasing 1/12 of the approved budget each month disruptive to the effective provision of public services. The monthly cash stress is typically accentuated by unforeseen expenditures triggered by emergencies. The main spending units that have to deal with such crises include Health for epidemics, Defense and Home Affairs for security-related emergencies, and Works and Agriculture for flood disasters entailing emergency repairs of infrastructure and food security. During FY00 it included handling of cholera outbreaks, refugee situations, food relief and security in the game parks. While it is possible to preempt some contingencies, such as dealing with tourism protection through a program of strengthening security presence in the parks, the bulk of other unforeseen events can only be dealt with on a contingency basis. Exchequer releases are typically issued between the 15th and 20th of each month instead of the stipulated date of the 5th day of each month. This concern was strongly expressed by all spending units met. The delay adds to further uncertainty in expenditure planning. Since there is a predetermined rule to base releases on the average of three previous months revenue collection plus projected external program support, it is hard for spending units to understand the delay. One of the causes of delay is a process of conforming to the monthly monetary programming necessitating protracted exchanges with BOT for finalizing the exchequer issues. It was noted that such exchanges could be programmed in such a way to enable meet the stipulated date for exchequer releases. iii Recommendation 2: So as to have a more effective cash management system the report recommends four actions: As a first step to address cash flow gaps, the report recommends that the Budget Guidelines require spending units to present a cash flow plan along with their annual budgets, which conform to the respective aggregate ceilings but clearly show the time pattern of their cash requirements. Particular attention should be paid to the typical lumpiness in expenditure patterns in these units. Currently submissions are being made for some lumpy activities such as for education and health and the suggestion here would systematically broaden the coverage of submissions to all spending units. Once consolidated the overall budget would also have a cash flow plan to provide a more systematic projection of exchequer releases. The report recommends that the Government considers a cash flow smoothing instrument that respects the overall budget ceilings for the year. Two options or their combination were discussed with the concerned stakeholders. One is a cash reserve operated as a Government deposit that can be drawn down or built up depending on the net cash requirement position versus available cash. There were indications of possible contribution to such a reserve by donors, provided it can be protected from becoming a financing instrument. The other option is to establish a prudent credit line with the BOT as advances subject to the same rule that the net position as end year be zero. The law allows for such advances with clear limits. The BOT currently provides such credit based on the shortfall between projected extemnal program support and actual disbursement and the credit is repaid when actual flows take place - and applying agreed automatic adjusters. The recommendation here is to broaden this coverage to include projected domestic revenue with specific safeguards against abuse by turning advances into a financing instrument as it occurred in the past. The previous PERs and this current one established that typically the variance between actual and projected domestic revenue collection is smaller than that for external resource flows, and that there is a concentration of external inflows in the last quarter of the year, which could also be used as a safeguard against such abuse. The Government may wish to consider the two options or their combination based on comparative cost assessment( including the opportunity cost of maintaining a cash reserve) and select an appropriate instrument for smoothing cash flow gaps beginning in FY02. * The report recommends that, a separate line item be provided for contingencies under Vote 50 to minimize the stress from unforeseen expenditures on the overall functioning of the Government. The current provision under vote 50 predominantly provides for salary adjustments and regular shortfalls in OC. The current arrangement does not permit to address these contingencies on a timely manner. To help speed up responses to emergencies, the Government may wish to have reasonable estimates of emergency requirements based on past experience and to develop and agree on automatic triggers for release based on predetermined criteria. * The report recommends that exchequer releases be issued on the stipulated date of the 5th of each month so as to minimize uncertainty on available cash. The timing of the meetings/exchanges between Treasury and BOT for ensuring conformity to the monetary program could be set within the stipulated timing target. The third area of budget management stress relates to payment arrears. There are two dimensions to this problem - dealing with the accumulated stock of arrears and preventing the building up of new flows of arrears through appropriate incentives. There is currently an initiative funded by the European union to clear up the stock of domestic arrears of the central government. A major problem is the authenticity of the size of the stock. Initially the stock was estimated to be within the range of 60-80 billion shillings but to date under SASP IV the stock that has been cleared up is already up to 120 billion shillings. An independent audit has rejected a substantial proportion of the stock initially verified by CAG. The main problems appeared to be linked to over-pricing of supplies and collusion between some officials and suppliers to short change the government. There is no similar operation of relieving the local governments of their debts, whose stock as of end 1998 stood at 8.65 billion shillings (per CAG report). More than half of this stock was due to iv councils' creditors and slightly more than 15 percent due to non remission of statutory deductions from employees for contribution to Local Authority Provident Fund. The gravity of the problem varies across local governments. For some of the poorer authorities, the stock of arrears makes up over 50 percent of their own revenue collection, representing a major potential drag on financing their development programs. Most better-to-do and fewer poorer local authorities visited by the PER mission have began, on their own, to reduce the stock of debt and often at the expense of much needed funding for public services in the poorer ones. A major issue in such debt relief programs is the moral hazard problem and preventing the recurrence of build up of arrears. The report discusses appropriate disincentives to building up new flows and possible approaches to tightening the financial management system against recurrence of unsustainable stock. Recommendation 3: The report recommends that the Government considers a relief program to clear up the stock of debt of poor Local Authorities (Highly Indebted Poor Local Authorities - HIPLA) to give them a fresh chance under the local government reform program. A separate study under the PER Working Group has been commissioned for a more in-depth analysis of this problem and recommendations for way forward. Following the central and local government relief operations, the report recommends the following measures to prevent recurrence of the unsustainable stock of debt. New payments arrears should be integrated into the following year's respective spending unit's budgets as first charge from their allocations. This used to be the practice in the past in order to enforce discipline in budget management. It would, however, require timely compilation of information on arrears (in the Platinum system where available) and verification. To prevent abuse of extra-budgetary commitments will require enforcing the centralized system for issuing LPOs (even where the automated system is not operative); enhanced enforcement of discipline by the accounting officers (AOs) to ensure that no invoices are held outside the financial reporting system; and strengthening the internal audit functions in ministries, regions and local governments. To minimize collusion with suppliers, it is desirable to publicize/make it very clear to the private suppliers that any LPO issued outside the central payment system will not be honored by the MOF. It will raise the risk costs to corrupt suppliers from non-payment for goods and services supplied inappropriately. Rolling out the Integrated Financial Management System (IFMS) The government has introduced the Integrated Financial Management System (IFMS), which when fully operational would permit improved transparency of public financial operations through real time information, and better controls through centralized payments and procurement processing. The EPICOR (formerly PLATINUM) is the software currently used to operate the IFMS. The Exchequer Ordinance has been revised inter alia to provide a legal basis for operating this new system effectively and accepting its reports for fiduciary accountability. To date all ministries in Dar-es Salaam are using the IFMS. However, two are not on line. In practice this means that all expenditure transactions are now being executed through the system. In the case of the two remaining institutions (President's Office and the Ministry of Defense and National Service and agencies under it), although not yet on line, the IFMS and EPICOR software have now been implemented. Central government agencies located in regions and districts are served through 19 Sub- Treasuries. Manual operation of IFMS has been in place since last fiscal year. During the financial year 1999/2000 all payments were processed at the central payment office in the sub-regional treasuries. The structure of the chart of accounts in use is the same as that in use at the Central Payments Office but all payments and revenues are captured manually. Following failure of the first attempt to automate the system, on line links are being established in the 19 sub-treasuries from where payments and commitments are being executed for all central government agencies located in the 19 regions. An earlier attempt at v automating IFMS in the sub-treasuries using the PLATINUM software suffered technical hitches in hardware and software services making connectivity inoperative. Furthermore, the system used dial up technology for linking with Dar es Salaam. Consequently the links were subject to telecommunication interruptions. The government is in the process of reinstalling equipment and linking the sub-treasuries with the center via satellite technology. Sub-treasuries connected to the IFMS center can now operate on line and the task to connect all the sub-treasuries is expected to be completed by end August 2001. IFMS and the EPICOR software are also being rolled out to 28 local authorities on a pilot basis under phase I of the initiative. 42 more have been targeted for phase II which begun in January, 2001. There are only a few amongst the 28 that operate the software effectively. All ministries, Sub-Treasuries and local authorities visited by the PER mission expressed their appreciation for improvement in efficiency and controls when the PLATINUM is operative and strongly urged for its wide use. In the meantime there are some key transitional concerns that need to be addressed urgently. Recommendation 4: * As of April 2000, a centralized LPO system became operational in Dar-es- Salaam obviating the need for continued use of printed LPOs. The presumption was that since many of the Sub- Treasuries had the PLATINUM system, the need for printed LPOs would subside. Due to the fact that the rolled out PLATINUM systems were not operative, a shortage of LPOs emerged endangering the integrity of the procurement system in areas where PLATINJM was not operative. There is therefore an immediate need for ensuring that adequate supplies of printed LPOs is availed to all those in need of them. * The Government needs to quickly review the IFMS roll out program to ensure availability of functioning equipment, effective supportive technical services, and adequate training to operating staff. LOCAL GOVERNMENT FINANCIAL MANAGEMENT AND OTHER ISSUES Budgetary stress: There is an apparent difference on budgetary stress between rich and poorer local authorities (LAs), based on the extent of arrears, own revenue as percentage of total revenue, own revenue per capita and underfunding. Differential treatment of LAs in exchequer releases andfunding: By and large under-funding is still a problem both in relation to needs and approved allocation. Some districts receive less resources than approved especially those for funding other charges. There is lack of clear consistency vis-a-vis underfunding, with marked variations in the extent of funding between regions and between districts with some districts receiving far less than their approved allocations. It is not clear what criteria is being used in deciding the allocation, timing of actual releases and disbursements since some LAs do not receive releases as expected. There is a need for a study to determine the extent of inequity of treatment and transparency in the criteria used for differential treatment. Cash flow plans: There are indications from the LAs that cash flow planning would help ease cash flow problems and reduce uncertainties in budget implementation. Some LAs already prepare cash flow plans while some better to do LAs have been able to use cash reserves to bridge cash flow gaps and smoothen cash flow over time. Generally, councils indicated their agreement to including the requirement for preparing cash flow plans in the Guidelines for the budgeting process. Financial Management Capacity: The four key positions - Treasurer, Expenditure Accountant, Revenue Accountant and District Planning Officer are filled by staff with the requisite vi professional qualifications. However, the Internal Audit departments are weak in capacity, because most of the auditors are new (72 recruited in 2000) and lack job experience to man Internal Audit department, while others are not qualified for the job but only seconded from the Finance Department to do the job. Positions below the accountants are also manned by accounts assistants, most of whom have not gone through a formal training on accounting. The finance management capacity is even more worrying and totally lacks at the ward level where, apparently most of the revenue collection in the Councils takes place. There is a need for further training of accounts clerks and orientation of audit staff to the financial 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. Internal Audit reports and their usefulness: Internal audit reports are essential in uncovering irregularities and ensuring that regulations are properly followed in the course of performing day to day financial management operations. The reports are important not only as a financial management tool for the management, but also in guiding and simplifying the work of external auditors. It would be important, therefore, that these reports are regularly produced and submitted both to the management and the Council. In particular, the Councilors should demand to read such reports on a quarterly basis and subsequently ensure that the management responds to queries contained in these reports. Role of the Regional Secretariat (RAS Office): The RAS office's advisory role is currently very limited and ambiguous. This has led some to think that the Regional Secretariat is an unnecessary overhead to the Government. However, given the distance between ministries and the LAs, there is definitely a need for an intermediate body, which will monitor and ensure that the national standards in service delivery and policies are adhered to. The Regional Secretariat is well placed to perform such a function. The report recommends that there is need to clarify the role of the Regional Secretariat vis a vis the sector ministries in monitoring and evaluation, and in enforcing national standards. The national budget provision for these functions should be shared between the sector ministries and the Regional Secretariats. Arrears: Most LAs and in particular the poorer LAs have accumulated huge payment arrears. Usually in the preparation of the budget, LAs include repayment of arrears as one of the expenditure items. In practice, however, it is only the better to do LAs that have been able to spare enough resources for this item. The inclusion of arrears in the expenditure side of the LA budgets has instilled a sense of responsibility to the LAs making them wary to incur more debts. It is recommended that such practice (integrating debts in their budgets) be formnally adopted in the guidelines for budgeting at all levels of government. The report once again recommends that the Government considers a relief program to clear up the stock of debt of poor Local Authorities (Highly Indebted Poor Local Authorities - HIPLA) to give them a fresh chance under the local government reform program. Frequency and quality of quarterly and annual financial reporting and feed-back from supervisors: As noted above monthly, quarterly, and annual financial reports are prepared by the LA management and submitted to the Finance Committee and the full council. Quarterly and annual reports are copied to the RAS and MORALG. However, none of the reports go to the sub-council level directly. The assumption is that the contents of the reports would reach the sub-council levels through their representatives. The report recommends that both financial and audit reports (translated into Kiswahili) be displayed /made public at ward and village levels. The practice existing up to the early 1980s of sending copies of the quarterly reports to Treasury was very useful for accountability / feedback. The report recommends that this practice be re- instituted and a desk officer designated in MOF to deal with such reports. vii MAJOR EMERGING ISSUES FROM THE PER FY01 CONSULTATIVE MEETING The theme of the PER FY01 Consultative Meeting focused on realigning the budget and the MTEF with the PRSP which is now the overarching national framework for implementing poverty reduction measures in Tanzania and which has become the benchmark document for prioritizing expenditures in Tanzania. A key feature of the 2001 PER consultative meeting was the broader participation both in terms of the substantial increase in the number (around 250) and composition of participants, compared to the last two PERs. Major Emerging Issues 1. PER Process Mileage Gained under the PER Process: There is a consensus that the PER process has enriched the Government budgetary process. Remarkable progress has been made on progressively increasing funding to priority areas. In addition, the PER has become an increasingly GoT owned and led process as well as highly participatory and transparent. All this needs to be maintained and strengthened. Adaptation of the PER Process to Poverty Reduction: The PER process has in addition to providing inputs into the budget process and development of the cross-sector and sector MTEFs adapted to focus more on poverty reduction. The subsequent PER therefore will need to pay greater attention to the poverty focus. Extending the PER/MTEF Process to the Sub-national Level: Flow of funds to LAs and service delivery units is of concern and GoT needs to put appropriate monitoring mechanisms in place with adequate follow-up. This will require, among others, (i) rolling-out the PERIMTEF process to the lower level (Local Government Authorities) as these levels do deliver the primary services. This will improve planning, execution and control of budget at that level. In addition this will strengthen local ownership and local participation. (ii) spending more resources on building capacity at all levels of government but more so at the sub-national levels. 2. Maintain Macroeconomic Stability: There is a broad consensus that the current success of the GOT in maintaining macroeconomic stability should be maintained. 3. Further Prioritization to Foster Strategic Expenditure Allocation: There is need to continue work on prioritization to foster strategic expenditure allocation using the MTEF instrument, including revisiting the priorities, to focus on sectors, such as the transport sector (broadly defined), which can make a real impact in terms of poverty reduction. It is important that ongoing interventions targeted at realizing sustainable development in the rural sector and delivery of high quality primary education are revised and updated to reflect the requirements of Rural Development, Agriculture and Primary Education for poverty reduction. 4. Concern on the Low and Drifting Down of the Revenue Effort: Low and declining revenue effort remains an important concern. The revenue effort is seen to be driven by taxes on international trade while performance of other taxes does not show much improvement. Efforts are thus needed in improving compliance and strengthening tax administration. 5. Concern on Continued Channeling of Donor Resources Outside GOT Budget: More attention needs to be put on improving the reporting and accounting of donor resources. Administration of donor support outside the Government budget creates real problems in budget management and accountability considering that over 30% of the budget is donor funded (over 80% for development budget). More donor resources need to go through the budget so as to support the poverty reduction strategy. Addressing this problem requires promoting/enhancing trust and collaboration between the viii Government and Donors. It also requires the GOT to continue demonstrating resolve to improve budget management, transparency and accountability. 6. Need to Harmonize SDPs and PRSP: There is an apparent disconnect between Sector Development Programs (SDP) and the PRSP. There is therefore a need to harmonize the SWAPs/SDPs with the PRSP for orderly implementation of the poverty reduction strategy. This implies that Sector development programs and strategies should be reviewed to ensure consistency with PRSP to ensure a strategic link between resource allocation and poverty reduction. 7. Options on Full Financing of the PRSP: Full financing of PRSP targets/requirements has implications on resource mobilization vis-a-vis attainment of the targets. Various financing options will need to be considered if the estimated resource gap of 3% of GDP is to be mobilized. The options include: reducing the PRSP targets or lengthen the period over which targets are to be achieved, or raise more domestic resources or through borrowing (external/internal) but without prejudicing macroeconomic stability. 8. Dealing With the Problems Associated with the Cash Budget System: The cash budget system has allowed Tanzania to achieve macro stability and increase discipline in budget management. However, the system denies flexibility in terns of informed planning. The cash budget creates a considerable level of stress to the budget, taking no account of lumpiness of expenditure or the ability to implement programs in an environment of uncertain/irregular/delayed funding. Government should explore possible feasible options in dealing with the problem of cash flows. The challenge facing Tanzania is how to move from the cash budget system to a flexible budget system hinged on credible government institutions. There is a need to address issues of budget management stress from the cash-budget system by: (i) revisiting prioritization by type of categories of spending and bringing to fist charge items like utilities and automatic catering and upkeep; (ii) cash flow planning, and (iii) creating a disincentive for accumulation of arrears. 9. Need to Address Weaknesses in the Financial Sector for Poverty Reduction: The financial system in Tanzania is not sufficiently deep, even by the standard of developing countries; so that the impact of financial intermediation on the growth of the economy and poverty reduction is still low. 10. Strengthening Public Financial Management: While the establishment of the IFMS has greatly enhanced the Treasury's capacity in managing Government finances, lack of ownership of the system threatens its sustainability in the long term and makes the system susceptible to potential breakdowns and/or sabotage. More training is needed together with the roll-out of IFMS and enhancing staff motivation of staff in Government generally. 11. Monitoring of Expenditure Impact: As progress is being made in sharpening poverty-focused priorities and expenditure allocations, increased emphasis will need to be paid to the monitoring of the impact of public expenditures. This should include further tracking studies and service delivery surveys as well as participatory monitoring and evaluation. 12. Enhance Transparency and Information on Expenditure Releases to Recipients and Beneficiaries: A key measure to improve accountability for the use of resources is to provide the intended beneficiaries with information about funds made available from the Central Government for specific purposes and beneficiaries. 13. Greater Attention to Local Authorities: It should be ensured that the' public expenditure management process takes reforms under the Local Government Reform Program adequately into account. Providing some form of debt relief to local authorities who are burdened with high levels of arrears and debts should be seriously considered. ix 14. Need to pay Greater Attention to the Close Link Between Poverty and the Environment: There is a realization that poverty and the environment are intricately linked, so that greater attention to this close link deserves greater attention in the poverty reduction drive. 15. Decisive and Immediate Action Needed Against HIV/AIDS: Decisive and immediate action against HIV/AIDS is necessary and should be fully funded by GOT and donors. 16. Continue Mainstreaming of Gender in the Government Budget 17. Implementation of the Pay Reform Requires More Impetus: Progress is being made regarding the Public Service Reform Program. Donors were asked to discontinue Local Cost Compensation (LCC) and support the Selective Accelerated Salaty Enhancement (SASE) scheme. CHALLENGES FOR THE FUTURE Overall the government has maintained good progress towards the use of the MTEF as a strategic allocation instrument and has this year turned its focus to adapting the MTEF formulation to the PRSP process. The key challenge is to further strengthen links with Sector Development programs, particularly where these were developed prior to the advent of the PRSP and further improving the costing of realistic requirements (including consistency with absorptive capacity) for achieving poverty reducing outcomes. The core pillar for improving public financial accountability remains the implementation of the Integrated Financial Management System. Very good progress has been made in rolling this out to all central government agencies, which together account for nearly 80 percent of all resources mobilized through the Exchequer. Rolling out the IFMS to all LAs would also go a long way in improving financial management at the sub-national level. This is a challenge that has to be faced by GOT. Results from the expenditure tracking studies and from the participatory CFAA underline the value of making service delivery surveys and expenditure tracking part of a routine annual assessment of the effectiveness of public spending. This is particularly important in view of the growing role of sub- national authorities and local communities in delivering/organizing essential services without adequate capacity and systems for financial management. A long step forward has been an agreement and initial steps to address the vexing problem of tight cash budget and its impact on limiting predictability in expenditure planning. GOT began quarterly releases and commitments for all priority activities since December 2000 and has committed to stay with this practice. GOT has also embarked on a system of cash flow planning and adopted measures for smoothing cash availability. The extension of this practice to non-priority sectors and preventing the build up of new arrears will reduce pressures on cash flow management. The PER process and dialogue seems to be paying off in terms of improving both predictability and flexibility of budgets. A larger proportion of recorded loans and grants is now being provided as budget support and a larger proportion of the external resources are integrated through the Exchequer (including the development budget). There is still a long way to go but progress has been substantial and the dialogue is at least taking place on the same platform. This raises the stakes for improving the integrity of financial management and result-orientation to provide the necessary comfort to make further strides. Finally, it is pertinent that the PER working group learns from past experience and continuously improves the PER process, to ensure that the process significantly contributes to enhancing the government's budget process. x 1. OBJECTIVES, METHODOLOGY AND CONTEXT OF THE FY01 PER 1.1 INTRODUCTION 1.1 This introductory chapter presents the objectives, methodology and context of the Tanzania public expenditure review (PER) FYOI. It highlights the sharper poverty focus adopted under the PER FYO] with the advent of the PRSP process and related sector development programs (SDPs). It also explains how the PRSP process influenced (i) strategic resource allocation; (ii) budget flexibility and accountability; and (iii) the development of a poverty focused medium term expenditure framework (MTEF) using the updated and fully costed PRSP targets. The chapter also highlights some special events that had an important impact on public expenditure including: the second multi- party general elections in October 2000, commitment to intensify the fight against HIV/AIDS, ratification of the East African Cooperation (EAC) treaty in July 2000, Tanzania reaching the HIPC decision point in April 2000, and conflict in the neighboring DRC. 1.2 In the context of the government-led annual PER process, the Bank provides support to the process in terms of strengthening the budget management system and the annual independent assessment of budget performance. Donors and other multilateral institutions provide analytic support and some of them routinely join the Bank-led PER main mission to conduct an independent assessment of budget performance and public financial management. In September 2000 the team carried out an identification mission to agree with the government on the prospectuses and implementation schedule of the Economic Sector Work (ESW) program for FYOI as well as action plans and the supervision program for the adjustment lending program. More specifically, the mission consulted with the government and the PER working group (which includes donors, local researchers and civil society) on the prospectus for FYO1 to determine analytical support from the Bank. The objectives and the approach to implementing PER FY01, presented below, were guided by the findings from this identification mission. 1.2 OBJECTIVES As in the previous years, the PER FYO had the interrelated twin objectives of (a) providing support to the budget process in Tanzania and (b) carrying out an external review of budget performance. (a) Support for strengthening the budget process and public financial management 1.3 The overall objective was to support the government's effort to improve the quality of its budget preparation and reflect explicitly the strategic public expenditure issues identified in the PRSP. The specific targets were to (i) engender strategic allocation of public resources through prioritization, and ensure that the overall spending program is consistent with maintaining macroeconomic stability; (ii) improve the predictability and integrity of budgets through the consolidation of the MTEF process; and (iii) adapt the PER process to support the implementation of the PRSP and institute pro-poor tracking of expenditure. (b) Assessment of budget performance and arising issues 1.4 As in the previous years, a traditional task of the PER FY01 was to carry out an independent assessment of budget performance for FYOO and the first quarter of FYOI. The specific aims were (i) to assess the consistency of budget allocations with the objectives of poverty reduction and economic growth, (ii) to review the extent to which actual spending corresponds to government plans and budget in terms of both levels and composition of spending; (iii) assess the efficacy of government I spending as evaluated by the Controller and Auditor General and other stakeholders; and (iv) identify the major concerns and issues arising from the reviews for future action. The findings would form the basis for recommendations for improving public financial management. For the FYOI PER, the external review intended to focus on three specific budget management issues: Downward drift of domestic revenue 1.5 A frequently mentioned concern is the downward drift of the revenue effort in the past three years. Previous analysis has focused on factors related to the efficacy of tax administration. The mission planned to investigate more comprehensively the main causes of this drift and specifically focus on the consequences of two phenomena: (i) the impact of past and prospective policy actions ( e.g. tariff compressionAiberalization, investment related tax exemptions and privatization); and (ii) the fact that GDP may be expanding faster than the tax base as mining and tourism (enjoying major tax exemptions) dominate the growth process. Dealing with the negative consequences of the cash budget system 1.6 Previous PERs (based on reports from spending units) and an IMF mission report iin early 2001 expressed concerns regarding the negative consequences of the rigidity of the cash budget system under which monthly expenditure releases are based on average revenue collection in the preceding three months. These consequences were primarily the unpredictability in expenditure planning and pressures for building up arrears through budget commitments made outside the official system. The FY01 PER set out to (i) explore the desirability and feasibility of integrating cash flow plans into the MTEF process to provide a better basis for overall cash flow projections and management; (ii) explore with the Ministry of Finance, donors and the Bank of Tanzania the feasible options for bridging intra-year cash flow gaps; and (iii) investigate the most effective and 'durable way for dealing with the problem of accumulating arrears focusing on sources of such commitments and ways for plugging avenues for making commitments outside the system, whereby emphasis would be placed on minimizing the moral hazard problem of bail outs. Financial management at local government level 1.7 The PRSP and a variety of other reports prepared under the Local Government Reform Program emphasize the need for improving financial management at the local government level, as Local Authorities increasingly assume the primary responsibility for the delivery of essential public services. A World Bank study on decentralization in Tanzania further underscored this concern. Building on these reports, the FY01 PER aimed to make an on the ground assessment of the status of the financial management system and review progress in strengthening the system. More specifically, it set out to assess (i) the effectiveness and capacity for the use of a bottom-up approach in expenditure planning by reviewing the expenditure prioritization process and the extent and reasons for variance between requirements and approvals of expenditure; (ii) the frequency and quality of quarterly and annual financial reporting and feedback from supervisors; (iii) the efficacy of the supervisory functions in budgeting and financial reporting/accountability systems of the local authorities. 1.3 METHODOLOGY 1.8 Bank support to the Tanzania PER is provided in the context of the annual PER process, which now has been in place for three years. This process has two overlapping main phases. The first phase focuses on technical work for strengthening budget management and feeds into the preparation of budget frame/MTEF. The second phase combines evaluative work and open consultation on strategic resource allocation, prioritization and budget management issues. 1.9 During the first phase the Bank's contribution was mainly in the form of providing inputs to the Budget Guidelines Committee in setting the budget framework, expenditure ceilings, and broad 2 cross-sector prioritization for the Tanzania FY02 - FY04 MTEF and the budget for FY02. In this respect the Bank team working with others in the PER Working Group, supported work on resource and expenditure projections consistent with the macroeconomic frame agreed with the IMF and with indicative external finance. In this task, emphasis was placed on reviewing progress and providing support to the costing of programs for meeting poverty reduction targets set out in the PRSP. It is important to note that at this analytic stage the PER process leverages substantial professional and financial resources from donors and other stakeholders. 1.10 In phase two of the PER process for FYOI, the Bank's main contribution was to lead the independent external evaluation (PER main mission) to (a) assess the consistency of budgetary allocations with growth and poverty reduction objectives, (b) review the extent to which actual spending corresponds to government plans and budgets in terms of both levels and the composition of spending, (c) assess the efficacy of central and local government budget management, (d) review the poverty-focus of the MTEF in line with the PRSP targets, (e) evaluate the openness of the budget process to consultation with all stakeholders, and (f) identify major concerns and issues (both cross- sectoral and sector-specific) in the area of public expenditure management. The findings from this review and other analytic evaluative studies, e.g. expenditure tracking and budget sustainability assessment, are coalesced into papers for discussion during the annual PER consultative meeting during which, the government conducts an open review of past performance and its plans for the future. The open review involves a wide range of participants including the press, parliamentary committees, civil society, donors, government at all levels, the private sector and researchers. 1.4 THE CONTEXT Initial poverty reduction efforts 1.11 Poverty reduction has been a key policy objective in Tanzania since Independence. During the 1970s and 1980s, under the socialist regime, poverty eradication was embraced as the main goal of development. The strategy adopted comprised income distribution and high aid intensity to support public spending programs on social services. This strategy yielded significant improvements in human development indicators. However, the strategy did not give due emphasis to economic growth. As a result these achievements could not be sustained as requirements for maintenance and operation of the largely-foreign-funded capacity expansion programs by far exceeded growth in the domestic resource base. Pre-PRSP poverty reduction initiatives 1.12 More recent efforts to tackle poverty predate the PRSP. The following initiatives were most prominent in preparing the ground for the PRSP: (i) The National Development Vision 2025, set the national aspirations and the way to attain a middle-income status by 2025. (ii) The National Poverty Eradication Strategy (NPES) set out objectives, indicators and targets for eradicating abject poverty by 2010. (iii) Sector-wide programs (SWAPs) beginning with one for the health sector, aimed to facilitate prioritization, improved resource allocation, improved aid coordination and national ownership of the development programs. (iv) The Tanzania Assistance Strategy (TAS) is a medium- term strategy encompassing joint efforts of the government and the international community, a partnership approach with enhanced Tanzanian ownership of her development agenda, and enhanced transparency in the conduct and application of aid, with a view to increasing the effectiveness of aid for poverty reduction. (v) The PER process focused on evaluating and supporting the development of strategic public expenditure allocation and monitoring through the medium-term expenditure framework (MTEF). The process emphasized the impact of public expenditures on poverty reduction, bringing the review of the integrity and effectiveness of public spending into the public domain, and enhancing the efficacy of public service delivery. (vi) The Local Government Reform Program (LGRP) devolves the responsibility and the means for the delivery of basic social and 3 infrastructure services to the local level in order to achieve greater efficiency and accountability in service delivery and to empower local communities. Advent of the PRSP and sharper poverty focus 1.13 While the PRSP is complementary to the earlier initiatives, it attempts to address the tension between what the sector programs consider to be ideal and what can realistically be achieved. The PRSP stresses that the policy menu should be limited and priorities should be highly focused on poverty reduction. While the PRSP is based on a long-term perspective, reflected in the National Vision 2025 and the National Poverty Eradication Strategy, it has translated long-term goals into annual monitorable targets covering a three-year horizon in the framework of the MTEF. There are several important aspects to the development of this framework in Tanzania: * First, the MTEF embodies a strategic approach to public expenditure allocation according to defined priorities and ensures consistency with macroeconomic stability necessary for sustained growth. In Tanzania, the MTEF, which predates the PRSP, is the basis for applying the strategic resource allocation approach in the annual budgeting process. * Second, as limited public resources (both domestic and external) are expected to play a pivotal role in financing the poverty reduction strategy, the planning framework, notably the sector-wide programs, are being brought under the discipline of the overall hard budget constraint. Furthermore, the preparation of sector programs takes into account human and institutional capacity limitations. * Third, the evaluation of the implementation of the PRSP will incorporate an assessment of the impact of actions taken. This builds on the experience from the past two years when, under the Multilateral Debt Fund facility (MDF), prioritization and protection of social sectors in public spending was monitored on a quarterly basis. The MTEF and the budget provided the benchmarks for evaluation. The MDF has now been transformed into Poverty Reduction Budget Support (PRBS), as Tanzania has qualified for debt relief under the enhanced HIPC initiative. However, monitoring mechanisms developed under the MDF will continue to play a central role under new budget support mechanisms such as the PRBS or the Bank's planned Poverty Reduction Support Credit. e* Fourth, the existing sector-wide programs will be recast and new ones developed guided by priorities defined by the national poverty reduction strategy. The PRSP targets serve as a guide in the medium term and the NESP for longer term targets and objectives. In this approach, the PRSP sets the framework for other development programs, while it is also a product of those plans. Given the hard budget constraint, the most appropriate framework for ensuring consistency between SWAPs and the PRS targets is the MTEF/PER process. * While the implementation of the PRSP envisages a shift in focus from input to results/impact orientation, the ability to carry out the needed analytical work is uneven across sectors and across levels of government. In this context, strengthening capacity is critical. Judging from the work that has been carried out on the various sector strategies, the desired financing of the programs far exceeds the levels consistent with the available and projected resource envelope. This underpins the necessity to make a further distinction between "priority" poverty reducing expenditures, to be covered under the government budget, and "other" expenditures to be funded as additional resources become available, resources from the private sector and self-generated funds. In this regard, the government of Tanzania has continued to urge development partners to channel most of their financial assistance through the budget to facilitate the planning and prioritization of the poverty reduction program, and simplify aid disbursement procedures and thereby foster government budget flexibility and accountability. 4 1.5 SPECIAL EVENTS THAT INFLUENCED PUBLIC EXPENDITURE DURING FY01 2000 General elections 1.14 The FY01 PER and budget took into account the government's commitments and expectations. One major commitment was that of financing the second multi-party general elections which combined Presidential, Parliamentary and civic elections in October 2000. The government allocated Tshs.45.8 billion to cater for activities like registration of voters, preparation of polling stations and voting materials, and allowances for election officers. Interim HIPC relief 1.15 The preparation of the budget for FY01 and the related MTEF took into account the availability of interim HIPC debt relief following the Tanzania reaching the decision point in April 2000. This enabled the government to enhance outlays (in real terms) to the priority sectors so as to improve the provision of social services which have a direct bearing on the living standard and capabilities of the poor. In this regard, the exercise of firming up the macro frame under the PER FY01 process involved the development of a high case scenario to take into account additional resource projections from the enhanced HIPC debt relief. Full PRSP 1.16 The government prepared a PRSP to address poverty issues in the context of the enhanced HIPC initiative. The process of preparing the PRSP had important expenditure implications during FY01, considering that the PRSP was to be generated through a participatory process which necessitated among others, conducting several zonal and national workshops to canvass views from all stakeholders including central and local governments, the private sector, NGOs, cooperative societies, and faith groups so as to arrive at a national consensus. The PRSP also set targets for poverty reduction in the medium term, necessitating a re-benchmarking of the MTEF after fully costing the means for achieving the targets. Commitment to intensify the fight against HI VIAIDS 1.17 One of the major thrusts of the FY01 budget was to reduce the spread of the HIV/AIDS pandemic. Consequently, in order to give HIV/AIDS the necessary attention in government expenditure programs, each ministry, department and region was required to allocate funds for this activity under the MTEF. In addition, the PER working group commissioned a study whose main objective was to develop an MTEF for a multi-sectoral HIV/AIDS response. Other events 1.18 Other events that influenced public expenditure during FY01 include the ratification of the East African Cooperation (EAC) treaty in July 2000, the preparation of the first medium term plan for the implementation of the National Development Vision 2025 and continued civil conflict in the neighboring Democratic Republic of Congo (DRC), Burundi and Rwanda. 5 2. EVALUATION OF THE TANZANIA PER PROCESS SINCE FY98 2.1 THE PER PROCESS IN RETROSPECT Introduction 2.1 Since FY98, the government of Tanzania working in partnership with the World Bank, other development partners/donors and local stakeholders, has undertaken PERs, which have greatly influenced pertinent public policy formulation and budget management. This chapter outlines the evolution of a participatory PER process in Tanzania adopted since FY98 to assess the usefulness of the PER and the associated MTEF process in terms of improving the quality of budget management and policy making. The assessment begins with an evaluation of the PER process, in the light of the context and objectives discussed in Chapter 1, and taking a retrospective look at the previous PER in terms of process as well as results. The focus then turns to the FY01 PER, covering the adoption and deepening of the PER and MTEF process; supportive technical & sector studies (including a synopsis of content); policy pressure from the PER working group (WG) on foreign resources disclosure and inclusion into the GOT budget; poverty focus of PER FY01 taking into account the context and key events; and analytic support provided. The evaluation is also cast in terms of issues raised, progress made to date regarding flexibility, predictability and budget accountability; and with respect to translating the MTEF into the budget. Finally the chapter draws up some important lessons to inform the future PER process. 2.2 Objectives and Organizational Framework: Prior to 1998 the PER was largely external to the Tanzanian government and most stakeholders. Besides not being effectively coordinated with government processes, local ownership was also weak and as a result, the whole exercise had minimal impact. The new PER approach adopted since 1998 in Tanzania has had four main strategic and operational objectives. First, over the past three years the PER process has focused on strengthening public accountability systems and on entrenching the participatory culture in assessing management and operational efficiency of budgets through primarily opening up the PER process to the public under the leadership of the government, and retaining the component of "external (to government) evaluation of budget performance" as a key feedback/peer review mechanism to the government. The second main purpose has been to encourage the adoption of a strategic approach to allocation of public resources through prioritization, and ensuring that the overall spending program is consistent with maintaining macroeconomic stability. The third objective has been to improve the predictability and integrity of budgets through the adoption of a medium term horizon in expenditure allocation, and the integration of sector programs as well as external finance into the MTEF. The fourth major objective has been to strengthen the institutional, legal and personnel capacity for improved budget management and effectiveness of public spending. More recently (beginning FY 01), the PER has evolved a step further to take on board poverty concerns as the overriding objective. 2.3 The Process: In Tanzania the PER has now been adopted as the main routine instrument for annual open review of budget performance and identifying critical strategic issues for improving efficacy of public spending programs. The medium term approach to strategic allocation, priority setting, and integration of sector programs into the overall budget frame have become increasingly routine. The two-phase approach to the process that was introduced in 1998 has been retained. The first phase focuses on technical work for strengthening budget management and feeding into the 6 preparation of the budget frame/MTEF. The second phase combines evaluative work and open consultation on strategic resource allocation, prioritization and budget management issues. The PER Working Group (WG), chaired by the GoT and involving sector representatives, donors, researchers, civil society and private sector, is the focal organizational point for the process. It meets once every fortnight and is responsible for managing the whole process. 2.4 The annual process begins with the approval of the prospectus (by the WG) for the year's PER process (August/September) taking into account lessons from the previous year. The WG then commissions and organizes finance as well as peer review for the technical studies (September - February/March); provides inputs to the Budget Guidelines Committee based on studies, analysis by its own Macro Group and external review of performance (conducted in a World Bank-led donor mission) (December/January); facilitates the preparation of cross-sector and sector MTEFs using outputs from technical studies (February-April); and organizes an open review (PER consultative meeting) in May to review strategic prioritization and budget management issues. The open review involves a wide range of participants including the press, parliamentary committees, civil society, donors, government at all levels, the private sector and researchers. The meeting is co-chaired by the Permanent Secretary Ministry of Finance and the World Bank Country Director. The WG then sees to the preparation of the PER report incorporating suggestions/views from the open review meeting. It is important to note that the PER leverages substantial professional and financial resources from donors and other stakeholders. The two-phase PER process followed in Tanzania has had the advantage of providing constant feedback from one phase to another as well as informing decision making throughout the budget cycle. 2.2 FY01 PER 2.5 The PER process for FY01 was initiated in early September with the first meeting of the PER working group (WG). The WG continued to be chaired by the Deputy Permanent Secretary (DPS), Ministry of Finance (MOF). The composition of the WG expanded compared to the previous PER cycle and now draws members from a broad range of stakeholders including the central and local government, bilateral and multilateral donors, research and academic institutions, private sector, civil society and NGOs. The key tasks of the WG included (i) drawing-up the prospectus of PER FY01 and MTEF work for the year; (ii) reviewing the terms of reference for the sector PER FY00 update work and supportive technical studies; (iii) peer-reviewing of all the interim or inception reports; (iv) soliciting and consolidating projected donor disbursements for purposes of firming-up the budget frame numbers; (v) wider dissemination of the major outputs of the PER process by members of the WG to their constituencies through regular briefings, sector working groups, seminars and workshops; and (vi) performing the oversight function over the PER process. 2.6 The WG has a macro sub-group and several sector working groups. During FY01, the macro sub-group was chaired by the Deputy Permanent Secretary - Ministry of Finance, while during the previous PERs it was chaired by a representative from the Netherlands Embassy. Other members of the macro sub-group from the government side included the Commissioner for Budget, Commissioner for Policy Analysis, Commissioner for External Finance, a representative from the President's Office - Planning and Privatization, and the Bank of Tanzania. Non-government members are drawn from the World Bank, IMF, European Commission, UNDP, UNICEF, DFID, SDC, CIDA, Norway, Sweden, Denmark, The Netherlands and, JICA/Japan. Key tasks performed by the macro sub-group during FY01 include preparation of the terms of reference and peer-review of two studies on overall budget sustainability; commenting on the draft Budget Guidelines regarding omissions and concerns before Cabinet approval; soliciting indications of external support over the MTEF period (by donor, sector, on/off exchequer, and project/program); developing guidelines for the full costing of the requirements to meet PRSP targets and checking for consistency of the numbers with respect to commitments made in the PRSP; and developing the structure and presentational formats of MTEFs for the PER consultative meeting. 7 2.7 The sector working groups are largely made up of officials of planning departments in the priority sector ministries. Besides their routine functions and participation in the broader WG, the sector groups are also responsible for working (alone or in collaboration with consultants) on the PER update for their respective sectors and the preparation of the sector MTEFs using the sector PER as the main input. The sector working groups also play the crucial role of disseminating the major findings and recommendations of the PER to senior sector ministry officials and stakeholders. 2.8 Financial support for the activities implemented during FY01 was provided by donors who are members of the WG as shown in the table below: Table 2.1: Studies and activities carried out under the FY01 PER STUDY/ACTIVITY FINANCIER CONSULTANT 01 Budget sustainability DFID EME (UK) & Oxford University 02 Tracking Expenditure UNDP ESRF/REPOA 03 Local Government Indebtedness EU REPOA 04 Background Study - Lands Denmark DOE /UCLAS - UDSM 05 PER Update - Education Sweden OPM 06 PER Update - Health SDC IHSD - UK 07 PER Update - Trunk & Regional Roads Norway COWI Consult / PWC 08 PER Update - District Roads SDC DOE/UDSM 09 PER Update - Agriculture Denmark ESRF 1 0 PER Update - Justice Denmark ERB/UDSM 11 PER Update - Water JICA ESRF 12 PER Study on HIV/AIDS Norway PSI/MUCHS 13 Cross-sector MTEF MOF & Macro Group 14 Preparation of PER FY00 Report. World Bank MOF/PER Secretariat 15 PER Main Mission WB, SDC, DFID, Denmark, UNDP & Norway 16 PER Secretariat World Bank 2.9 To ensure sustainability, most of the studies were commissioned by the WG to local consultants who then worked closely with the sector groups in all the priority sectors (partly as an instrument to build local capacity). Review of the supportive technical and sector studies: (i) Fiscal aggregates study: The main objectives of this study were to: (a) review fiscal sustainability in the light of recent decisions on debt relief and projected new flows as an extension of the fiscal policy and the macroeconomic context study done during the FY00 PER; (b) clarify the constraints arising from the cash budget system and the possibility of accommodating a small deficit to allow for flexibility in expenditure (i.e., what level of deficit could be pursued as a matter of policy?); (c) take the estimates of financing requirements for priority sectors and analyze the macro consequences of fully funding these levels of expenditures from various funding options; and (d) undertake a study on Tanzania's revenue potential paying attention to policy handles which link directly to revenue performance. This study has been very instrumental in suggesting fiscal policies that would pay greater attention to the funding needs of priority sectors while ensuring consistency of the target fiscal deficit with the key objective of macro-economic stability. In previous years, significant tensions have developed as government was running budget surpluses concurrent with widespread under-funding of basic social services. The fiscal aggregates work was also a key input into the full costing of PRSP targets as well as in the preparation of the FY02 budget paper for-Cabinet approval. (ii) Support to improve donor coordination and integration of aid into the budget: The PER working group solicited disclosure of and compiled external resource projections (using a questionnaire) to feed into the firming-up of donor support for the medium term FY02 - FY04. The numbers were then re-confirmed independently by the External Finance Department in the Ministry of Finance prior to incorporation into the budget frame. This exercise, which was an update of the work began during the PER FY00, aimed at addressing the problem of a large share of external resources (70%) passing outside the Exchequer system, which in turn leads to a lack of ownership and limited coherence between public expenditure and donor assistance. In particular, the undertaking was geared to get a 8 better feel of the development resource envelope, in order to integrate the development budget into the whole resource envelope rather than treating it as a residual. (iii) Tracking expenditure allocation and its effectiveness in reducing poverty: Considering that the government of Tanzania has targeted its allocation of resources to poverty reducing activities/sectors under the PRSP, under the PER FYOI a study was implemented to track such expenditures to ascertain whether they reach service delivery units and have the expected impact in terms of improved service delivery and poverty reduction. The cverall aim of the study was to assess the efficacy of pro-poor budget execution, focussing on primary education and health care. The study entailed a review of government disbursements and of the flow of materials and supplies, as well as reporting systems. The field survey covered the central government (ministries of finance, education, health, water and regional administration and local government), five local authorities (Babati, Kisarawe and Dodoma rural councils and Mtwara and Kigoma urban councils) and service units. (iv) PER FYOO update work, costing of PRSP Targets and development of MTEFs: The WG provided support to the government in preparing seven sector MTEFs for the priority sectors (education, health, water, roads, agriculture, the judiciary, and lands) and the cross-sector MTEF. The work involved oversight and peer review of sector PER update work on priority expenditure needs and priorities right from the terms of reference to the final reports to feed into the development of improved sector MTEFs and the budget for FY02. In addition, the PER FYOI process focused particularly on poverty reduction. As part of the PER update work, the WG also prepared generic guidelines to be used by sectors and consultants to prepare full costing of priority sector requirements in four steps: Step 1: Identification of targets for the medium term (2001/02 -2003/04) from the PRSP for the priority sectors and activities (health, education, water, rural roads, agriculture, judiciary and HIV/AIDS). Step 2: Estimation of the requirements for meeting the targets and costs of implementing the action programs identified in the PRSP using projected /updated unit costs. Step 3: Generation of scenarios for achieving the targets subject to the projected resource constraint (per Budget Guidelines and with additional 10 percent increase in the resource envelope) and implementation capacity constraints, while pointing out the assumptions and trade-offs (including efficiency gains) involved in the different scenarios. In assessing the scenarios, sectors were asked to spell out clearly what the government of Tanzania would need in the light of what other players (private sector, NGOs, religious institutions etc.) are doing/providing and, additional resources required to meet the optimal levels of service provision. Step 4: Recommendation on the optimum expenditure frame and any suggestions for revisiting timing and phasing of activities; and identify resource and capacity gaps for achieving the most desired long term scenario. The first round of checking for consistency of the full costing requirements with the overall resource envelope, and recommendations on any additional financing strategies were subsequently done at the aggregation stage during the preparation of the cross-sector MTEF. (v) Study on HIV/AIDS: This study addressed issues related to effective financing of HIV/AIDS activities by different sectors and organizations in Tanzania. The study covered the experience of financing HIV/AIDS activities during FYOI with the view to propose improved mechanisms for the effective financing of HIV/AIDS activities and effective ways of disbursing, utilization and accounting of resources and funds according to government budgetary standards. The study also developed an MTEF for the multi-sectoral HIV/AIDS response and made recommendations on strengthening planning and budgeting of HIV/AIDS in the context of the future MTEF rounds. (vi) Study on local authorities indebtedness: In many local authorities (LAs) the financial requirements to meet their operational requirements far exceed the resources available. This has contributed to the build-up of arrears with private suppliers, staff and statutory bodies. It is believed that this build up of "debts" militates against the ability of LAs to provide better services to the local population. The objective of this study therefore was to examine the current level of indebtedness of 9 local government authorities and their material effects on councils' operations. The study also considered the longer term aspects of LAs viability and sustainability, including possibilities of debt relief, grants for disadvantaged areas, revenue sharing between the center and LAs, and workable measures for stimulating and providing enabling conditions for expanded local activity. (vii) External evaluation of budget performance: This is a yearly exercise which is undertaken by donors led by the World Bank. It is a standard output of the annual PER. The major objectives of the FYOI external evaluation exercise were to undertake a review of budget performance including: (i) an analysis of the extent to which actual spending corresponds to government plans and budget in terms of both levels and composition of spending; (ii) an assessment of the efficacy of government spending; (iii) the identification of the major concerns and issues in the area of public expenditure management; (iv) an review the poverty focus of the MTEF in line with the poverty reduction paper (PRSP) targets; and (v) an assessment of the budget and financial management capacity and constraints in local governments. 2.3 ACHIEVEMENTS OF THE PER/MTEF PROCESS 2.10 Although it is difficult to establish a direct causal link between improvements in public finance management and the PER process, it is nonetheless acknowledged that significant improvements have taken place in a number of areas which have been a central focus of the PER process. In general, the government owned and led PERIMTEF process, which is highly participatory and transparent, has become a key instrument to inform decision making and priority setting on important budget issues. Specific achievements include: * Improved prioritization and progressively increasing funding for the priority sectors: Recommendations derived from the PER work have guided strategic resource allocation. Consequently, the share of total spending on priority sectors (dominated by the social sectors) as well as allocations to expenditures on operations and maintenance or other charges (OC) have increased significantly over the last three years compared to before. There has also been a significant re-orientation both in terms of allocation and outturn of development spending towards the social sectors, regions and local governments. • Improved strategic resource allocation through the adoption of an MTEF approach: The government has maintained good progress towards the use of the MTEF as a strategic allocation instrument over the last two years and is now turning its focus to adapting the MTEF formulation to the PRSP process. Strategic resource allocation has benefited from the feedback provided through the PER process and the greater transparency and openness in the budget process that were facilitated through the PER process. . Strengthening of budget management systems and institutions: Expenditure planning, execution and control has improved significantly through the adoption of the MTEF, implementation of IFMS and enactment of the Public Finance Act 2001 and the Public Procurement Act 2001. The PER process has been used as one of the instruments to follow- up implementation and roll-out of IFMS to all ministries, regions and local authorities. * Flexibility, predictability and budget accountability: There has been significant improvement in the predictability and availability of funds for OC and development expenditure compared to previous years. The discrepancy between budget allocations and actual expenditures has narrowed down. This is the result of various factors, including more realistic budget projections, better expenditure estimates, and a broader consensus on expenditure priorities which reduced the need for post-budget reallocations. In response to the policy dialogue conducted under the PER, government has commitment and implemented a three months expenditure commitment system and quarterly releases of OC to all priority sectors since December 2000. In addition government has began to publish all allocations in the news media by ministry, local authority and for the key priority sectors. 10 * Overall fiscal sustainability: The PER process has broadened the discussion on fiscal stability through analytical work that outlines an expanded range of fiscal policies that are consistent with overall fiscal sustainability. As a consequence, the policy dialogue has now evolved from focusing on simple deficit targets to a more complex framework that tries to reconcile funding requirements for basic social services under the PRSP with the objective of maintaining fiscal stability. * Poverty focus of public expenditures: Analytical work done under the PER has highlighted the point that government expenditures have a significant impact on poverty reduction and hence the need to make improvements in the quality of expenditure a government priority. In this regard, the PER process has provided important inputs for the design, update and monitoring of the Tanzania PRSP, which has become the national framework for implementing poverty reduction measures and a benchmark document for prioritizing expenditures. * The PERIMTEF process as a coordinating toolfor various other initiatives: The PER/MTEF process is now widely regarded as the main framework for coordinating and harmonizing other initiatives including PRSP, Country Financial Accountability Assessment (CFAA), and Sector Development Programs (SDPs). The PER/ MTEF process has also given more impetus to the Poverty Reduction Budget Support (PRBS) approach as an instrument to ensure that adequate resources are allocated to priority sectors through the budget in support of the poverty reduction strategy. * Improved information flow and aid coordination: The opening up of the budget process through the PER process has led to improved information flows between Treasury, sectoral ministries, donors, and other stakeholders. Furthermore, improvements in foreign resource disclosure and reporting on aid flows under the PER process have considerably strengthened the information base for government budgeting. The PER process and measures to improve budget management have also improved donor confidence and motivated a greater share of donor assistance provided in the form of budget support or channeled through the budget, thus limiting transaction cost and enhancing government ownership. In that regard, the MTEF process associated with the PER now offers an instrument for aid coordination and for ensuring consistency with macroeconomic stability. The influence of the PER in shaping recent macroeconomic developments is significant. For example, once the foreign resource projections were agreed to by the WG, the Ministry of Finance adopted the numbers as indicative resource flows to Tanzania. These were re-confirmed with individual donors after making some corrections for marginal errors based on past experience. The government also used the information generated by the PER process to counter-check the authenticity of development budget data on donor project finance expected by ministries and local authorities. Similarly, in firming-up the FY02 budget around May 2001, the government made adjustments to overall sector allocations as per budget guidelines to take into account not only the latest information available then but also full costing of PRSP requirements done under the PER. Consolidation and deepening of the PER/MTEF process: The PER has now been adopted as the main routine instrument for annual open review of budget performance and identifying critical strategic issues for improving efficacy of public spending programs. The medium term approach to strategic allocation, priority setting, and integration of sector programs into the overall budget frame have become increasingly routine. Bringing systemic fiscal issues into the domain of public policy discussions: The analytic work undertaken under the PER/MTEF process has been instrumental in raising pertinent systemic fiscal issues for attention by the authorities and suggesting feasible options for their resolution. Some examples include improvements in cash budget management and tackling the problem of arrears as a result of recommendations from the PER process in favor of front- loading of donor budgetary support to ease cash flow constraints, use of bridging finance and 11 preparation of forecasts of cash flow requirements by spending units. The analytic work has also been very useful in terms of revealing malpractices and weaknesses in budget management including short-term borrowing from from the road toll account for other uses; problems with accounting and disbursement of funds from the health basket fund; inadequacy of financing the road program from the Road Fund alone; potential undermining of strategic resource allocation through the mushrooming of earmarking of revenue; the plight of the development budget; and multiplicity of donor financing mechanisms * Focus on the status of budget andfinancial management in local governments: As part of the external evaluative function of the PER FY00 and FY01, an on the ground assessment was made of the status of the budgeting process and management in LAs, problems and weaknesses, as well as efforts underway to strengthen them. The findings of this work have been instrumental in bringing the local government reform issues on the table for policy discussion and to help in charting out the way forward. Work was also done to assess the gravity of local authority debts and to come-up with suggestions on how to help highly indebted LAs while avoiding contraction of new debts. 2.4 OUTSTANDING BUDGET ISSUES AND CHALLENGES FOR FUTURE PERs * Poor accounting and reporting of donor finance: In spite of the improvements made under the PER with respect to the reporting of projected donor finance over the last two years, the accounting and reporting of donor finance remains far from satisfactory. Continued dialogue under the PER process, together with improvements in financial management, governance more broadly, and accountability are key in resolving this problem * Low operational efficiency: The continuing problem of low operational efficiency in public spending mainly due to: (a) gross under-funding, partly reflected in continued variance between budgets and actual expenditures, leading to constraints in prioritized service delivery - the problem is more acute for maintenance as investment programs appear to face a softer financing constraint; (b) low predictability of resource availability partly constraining application of the performance budget; (c) personnel incentive problems as civil service pay has declined in real terms over the past three years which underlines the urgency of dealing with affordability and financing constraints to the implementation of the Medium Term Pay Reform (MTPR); and (d) continued weakness in financial accountability as brought out in the CAG reports particularly regarding accounts of regions and local authorities. * Weaknesses in financial management: There remain weaknesses in financial management and the related institutional frarnework. These include: incomplete roll-out of IFMS particularly to the regions and LAs which limits coverage, timeliness, and cohesion of financial reporting; disconnect between the central and local government financial reporting system despite large budget subventions to LAs and major responsibilities for basic service delivery; and the yet to be implemented link between IFMS and MTEF using the multi-year budget module in the EPICOR; as well as poor donor coordination. * Capacity constraints: Low capacity particularly at the sector level and at the regional and district levels remain as critical constraints in improving budget formulation, implementation and reporting at those levels. Weaknesses in the preparation of full costing of PRSP targets and sector MTEFs as well as poor budgets of most local authorities were largely the result of limited capacity and poor work incentives. This requires making an assessment of absorptive capacity at the relevant levels and working out systematic capacity building / training programs in budget management while revisiting the role of external consultants in the process. * Limited dissemination andfollow-up of PER findings: Although the PER/MTEF process has produced a good number of reports/findings and recommendations, the follow-up mechanism 12 is still lacking and needs to be worked out. Possible considerations include MOF issuing directives to implementing agencies to provide formal responses and follow-up actions during the PER consultations. The PER technical studies could also include a summary of main conclusions and policy implications for wider distribution to the general public, policy makers, local authorities, Parliament and civil society. * Phased-extension of the PER/MTEF process to lower levels of government and Zanzibar: So far the PER/MTEF process has not covered Zanzibar and it would be important to explore the feasibility of expanding the PER process to Zanzibar and work out how this should be done. Aanalogously, a phased-extension of the PER/MTEF process to lower levels of government is needed, particularly in view of the government's firm decision to devolve more powers and decision making to the LAs as the main arm of the government which provides basic social services to the population. In the interim, this calls for broader participation of LA officials (District Executive Directors, District Treasurers, Association of Local Authorities of Tanzania (ALAT) etc.) in the PER process. * Corruption: The fight against corruption needs to be intensified, particularly in infrastructure investment and social service delivery. * Fight against HIV/AIDS: Decisive and immediate action is needed against the HIV/AIDS pandemic. * Other challenges: Other challenges that need to be addressed include: the continued problem of poor revenue performance; development of improved forecasts of cash requirements by spending units; the need to conduct regular or formalized service delivery and expenditure tracking surveys or benefit-incidence studies; and the need to pay greater attention to the link between poverty and the environment. 2.5 SOME LESSONS FOR THE FUTURE * Maintaining macroeconomic stability is key: Macroeconomic stability is key for increasing growth and reducing poverty. Thus it is extremely important that the macro-economic gains achieved so far in Tanzania are sustained. However, in addition, it needs to be ensured that such gains are translated into poverty reduction. In order to realize growth which benefits the poor, sustained growth in agriculture and investment in basic social services and rural infrastructure is paramount. * Timing of the PER activities to dovetail well with the budget cycle: The work program of the subsequent PERs should be developed and discussed early in the PER cycle (July/August) so as to establish clear objectives and facilitate the development of the terms of reference for the analytic work in support of the process and subsequent preparation of the reports and MTEFs. * Strong leadership and ownership of the process by the government and participation by all stakeholders: Strong leadership and ownership of a participatory PER process by the government (central & local government and sector working groups) is fundamental for the process to have significant impact. * Stronger role for the civil society: Civil society has a very important role to play, particularly in monitoring government performance. * Transparency in resource allocation and disbursements: If resources are to reach the poor and have impact it is necessary to safeguard transparency in both resource allocation and disbursement. Thus, the intended beneficiaries need to be provided with information about funds made available from the central government. 13 * Monitoring the impact of public expenditure: As progress is being made in sharpening poverty focused priorities and expenditure allocations, it becomes imperative to pay greater attention to monitoring the impact of public expenditure in terms of poverty reduction and social service delivery. This calls for the need to make social service delivery surveys and expenditure tracking part of a routine annual assessment of effectiveness in public expenditure. 14 3. REVIEW OF RECENT MACRO PERFORMANCE 3.1 OVERALL ECONOMIC PERFORMANCE 3.1 Tanzania's overall economic performance in recent years has remained robust. The country is largely on track with regard to meeting macroeconomic objectives and targets of the Poverty Reduction and Growth Facility (PRGF). GDP growth in the last three years has been steadily rising, estimated at 4.9 percent in 2000 compared to 4.8 percent in 1999 and 4.0 percent in 1998, led by strong performance in exports, increased gold production and investment in tourism. The rate of inflation has fallen significantly, reaching single digit in 1999 driven by a decline in both food and non-food inflation rates from the last quarter of 2000. In June 2001 the annual headline inflation rate was recorded at 5.1 percent. Sustained donor support and adherence to the cash budget management system has ensured a modest fiscal surplus (after grants) amounting to about 1% of GDP in FY01. These achievements partly reflect the government's continued efforts at laying the foundation of a modern, diversified economy through accelerated implementation of structural reforms aimed at removing the remaining constraints for private investors, increasing the efficiency of the economy and creating an environment for private sector led growth. Although private sector growth, beyond mining and tourism, has been slow in responding to the liberalization, and the inflation rate remains above that of Tanzania's principal trading partners, the Finance Bill for FY02 has included measures geared to support a continuation of Tanzania's strong macroeconomic performance in FY02 and beyond. 3.2 FISCAL DEVELOPMENTS 3.2 The objective of fiscal discipline and macro economic stability has continued to dominate government budgetary developments. In general, fiscal developments have been encouraging. The fiscal balance improved during FY98 and FY99 recording budget surpluses (0.2 to 0.4 percent of GDP, after grants). In FY00 the deficit after grants was estimated to be around 0.5 percent of GDP. In FY01 the tax revenue collected have been 6 percent higher than target and expenditures remained below the target despite the elections of October 2000. The fiscal surplus for FY01 is expected to be about 1 percent of GDP. The budget for FY01 included further steps in the ongoing tax reform, including introduction of VAT on petroleum products and the elimination of government exemptions on VAT. The budget for FY02 also shows that implementation of all priority spending programs and locally financed development spending will continue as budgeted. However, although there was substantial repayment of budgetary arrears from previous years, continuing weaknesses in public spending control led to some new arrears. 3.3 In the areas of financial management and accountability, government has rolled out the computerized integrated financial management system (IFMS) to all ministries and all sub-treasuries. All the ministries are on line in a centralized IFMS using the EPICOR software. Effective July 2001, the central government expenditure controls are being implemented using IFMS, both against the exchequer issues and the budget allocations. All payments and reconciliation by Bank of Tanzania are now done centrally. Similarly, 19 sub-treasuries are also using IFMS based on stand alone servers and the EPICOR software. The national budget for FY02 was also prepared in IFMS. 3.4 During FY00 Tanzania also made significant strides in improving its external debt management and qualified for the HIPC interim assistance and Paris Club VI debt rescheduling. The government is on track for reaching the Completion Point during the latter part of 2001 under the enhanced HIPC initiative. Significant steps have also been taken to deal with its domestic debt particularly by reducing government arrears with the private sector and private external commercial debt. As a result of the HIPC debt relief, the EIU forecasts that Tanzania's total external debt will 15 decline from US$7.4 billion in 2000 to US$5.8 billion in 2002, while the debt-service ratio will fall from 18.8% to 12.0% over the same period. 3.3 MONETARY DEVELOPMENTS 3.5 Tanzania's monetary policy has continued to aim at a low rate of inflation, with remarkable achievement in terms of lowering inflation to single-digit level even against a backdrop of higher oil prices. However, data on monetary developments indicate that broad money supply (M3) increased during FY00 by 21.7 % compared to FY99 mainly due to a significant increase in net foreign assets (NFA) in the Central Bank and commercial banks. The increase of NFA by 49.8% in FY00 compared to only15.1% in FY99 largely reflect an increase in foreign investment activities, donor flows and earnings from tourism as well as purchase of foreign exchange by the Bank of Tanzania on the Inter-bank Foreign Exchange Market (IFEM). Although the developments in M3 have to some extent contributed to the recent improvements in the balance of payments, higher NFA has not served to build import capacity except for the component earned from tourism. Furthermore, this development has generated an exogenous pressure on monetary expansion which needs to be managed to ensure competitiveness of the Tanzania shilling at a time of sustained donor inflows so as to smooth any sharp falls in the exchange rate. A related problem concerns the fact that the ratio of M3 to GDP has continued to decline, leaving excess liquidity in the banking system when the private sector remains starved of credit. Trends in interest rates have not changed much over the last two years. The spread between lending and deposit rates continue to be high and widening, indicating lack of competitiveness in the banking sector and high interest risks of repayment/default. As regards financial services in the economy, the rural sector remains deprived of financial services with most of the private banks operating only in a few urban centers. During FY00 commercial bank credit to various sectors increased by only 10.9 percent compared to 40.3 percent for the preceding year. The major beneficiaries of credit included trade, mining and manufacturing sectors. Nevertheless credit extended to the productive sector remains low and declining and thereby acts as an impediment to growth. In particular, although overall credit to agriculture showed a slight increase during FY00, the amount directed to the purchase of agricultural products continued to decline following the liberalization of crop marketing and the decision by the government to stop providing guarantees on commercial bank credit to cooperative unions. 3.4 EXTERNAL SECTOR DEVELOPMENTS 3.6 In the external sector the export performance has been strong, specially due to the start of the gold exports and resumption of fish exports to the European Union following lifting of the ban on such exports in 2000. Exports of merchandise goods increased by 21.9% in 2000 over the 1999 eamings of US$543.3. Exports of services also increased by 2.6% over the same period. The exchange rate depreciated by about 6 percent during the end of January-February, 2001, mainly a market corrective response to a strong US$ but has remained stable since then around T.Sh.890 per U.S. dollar. Despite sharply lower international prices for most traditional products, exports grew and the external position improved. By end-2000, gross official reserves had reached the equivalent of 5.3 months of imports of goods and non-factor services. The current account deficit fell substantially from an average of about 15% of GDP during the 1990s to about 4% of GDP in 2000 mainly because of a fall in the level of imports and a rise in current transfers. 3.5 STRUCTURAL REFORMS 3.7 Substantial progress in structural reforms has been made in the last few years aimed at bolstering market efficiency and private sector led growth and reducing the involvement of the public sector in commercial activities. A majority of the banking sector now is in private hands with the sale of NBC to ABSA and the increased participation of the international banks in the country, with Barclays Bank being the newest addition to the more than 23 commercial banks now operating in the 16 country. Having divested/liquidated a large number of state owned enterprises in the manufacturing sector, the focus of the privatization program has now moved to the restructuring and divestiture of large utilities, especially power and water, and provision of appropriate regulatory framework to govern the operation of these liberalized sectors in a competitive environment. 3.6 Focus ON POVERTY REDUCTION 3.8 Having made significant strides in macro stability, fiscal sustainability and debt management, the government concerns have re-focused on poverty reduction. With 57 percent of the rural population living below the international poverty line of US$1 a day, poverty is predominantly a rural phenomenon. GOT has embarked on an aggressive poverty reduction strategy, which emphasizes both higher growth and more effective public service delivery. The poverty reduction strategy focuses on four strategic areas, including (i) rural development; (ii) improvement in social services; (iii) private sector and infrastructure development; and, (iv) public sector reform and institution building. The strategic prioritization areas are reflected in budget for FY01 and the budget for FY02 presented to Parliament in June 2001. Both emphasize further increasing public spending in the priority sectors of primary education, primary health care, agricultural research and extension, rural roads, the judiciary and HIV/AIDS. In addition, the budget for FY02 contains tax policy measures that will provide cost relief for all essential drugs and equipment to treat communicable diseases (including HIV/AIDS), provide incentives for private investment in education and in small scale enterprise. The latter measure aims at expanding income earning opportunities for the poor. 17 4. REVIEW OF BUDGET PERFORMANCE 4.1. INTRODUCTION 4.1 This chapter presents a review of fiscal developments and expenditure management issues and covers the fiscal year 1999/00 and the first eight months of fiscal year 2000/01. The analysis focuses primarily on issues related to aggregate fiscal discipline and strategic allocation. Specific sectoral issues relating to the efficiency of public expenditures have been analyzed in sector specific PER updates and these reports can be obtained from the PER secretariat. Within the framework of the PRSP and HIPC, the monitoring of pro-poor expenditures receives increased attention. To ascertain that increased resource allocations and disbursements for pro-poor expenditures, undertaken by the Ministry of Finance lead indeed to increased resource availability at the point of service delivery, a pro-poor expenditure tracking study was undertaken in the framework of the PER. The results of this expenditure tracking study are presented in section 4 of this chapter. The chapter concludes with a few observations on accessibility and transparency of information on public expenditures in Tanzania. 4.2. AGGREGATE FISCAL PERFORMANCE Domestic Revenue and Foreign Inflows 4.2 Total revenue as a percentage of GDP continued its decline from the peak of 13.5% of GDP reached in FY97 to 12.0% in FY98 and to 11.3% in FY99 and FY00. In the first eight months of FYOI revenue collection had slightly recovered to an annualized rate of 12.0% of GDP. This improvement is mainly due to the imposition of VAT on petroleum products as a deliberate tax administration measure aimed at curbing tax evasion on petroleum products. Underlying the decline in tax revenue over the past few years are substantial reductions in external taxes, 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 from the shrinking parastatal sector. In FY99, the bulk of the revenue shortfall was accounted for by a sharp decline in revenue from other taxes, which declined from 1.8 percent of GDP in FY98 to only 1.2 percent of GDP in FY99. This is mainly the result of the streamlining of the tax system and the elimination of a number of nuisance taxes. It may therefore be important to recognize that the tax system is still in transition such that there are temporary declines in revenue as nuisance taxes are eliminated. The expectation is that this declining trend will be reversed as the supply side responds, albeit with a lag. However, increasing tax revenue remains an important challenge in the area of fiscal management, given the substantial expenditure requirements that arise in connection with government's role of providing infrastructure services and basic social services in support of economic growth and poverty reduction. The program of strengthening tax administration is ongoing and expected to yield tangible results over the next few years. In addition to administrative improvements, the streamlining of tax exemptions is another potential source for additional revenue generation. 18 Table 4.1 Government Revenue and External Grants (as % of GDP), FY96 - FY01 Central Government FY96 FY97 FY98 FY99 F Y00 F'0l* Operations Total Revenue 13.2% 13.5% 11.9% 11.3% 11.3% 12.0% TaxRevenue 11.3% 11.9% 10.9% 10.1% 10.3% 10.9% Taxes on imports and 3.6% 3.8% 3.5% 3.6% 3.4% 4.9% exports Sales and excise taxes on 2.8% 3.1% 2.7% 2.6% 2.6% 4.9% local goods Income taxes 3.1% 3.0% 2.9% 2.7% 3.1% 2.5% Other taxes 1.9% 2.1% 1.8% 1.2% 1.2% 2.4% Nontax revenue 1.9% 1.6% 1.0% 1.2% 1.0% 1.1% * annualized estimate based on prel. outturn for July-Feb. 2001 Source: Tanzanian authorities 4.3 The relatively weak domestic revenue performance has been offset by increases in official development assistance in the form of grants and concessional loans (net of amortization). Since FY96, grants and foreign loans have seen a continuous increase, with data for the first eight months of FY01 indicating that donor's provide now about 5.5 % of GDP in resources to the public sector. This steady increase in concessional foreign resources presents the donor community's response to domestic policy reforms which create a favorable environment for enhanced aid effectiveness. Along with the general increase in foreign resources came also significant changes in the composition of foreign assistance. Among donors who provide grant financing there is an increased use of budget support through facilities such as the multi-lateral debt fund (MDF) and now the poverty reduction budget support (PRBS). Disbursements on concessional loans have also increased significantly, reflecting accelerated project implementation. Table 4.2: Foreign Inflows - Grants and Loans (as % of GDP), FY96 - FY01 Central Govemment FY96 FY97 FY98 FY99 FYO0 FY1O* Operations Grants and Loans 1.0% 3.1% 4.0% 4.3% 4.7% 5.5% Grants 2.2% 3.6% 3.0% 3.9% 3.9% 4.0% Program 1.0% 1.8% 0.7% 1.2% 1.7% 3.1% Project 1.2% 1.8% 2.3% 2.7% 2.2% 0.9% Foreign loans (net) -1.2% -0.5% 1.0% 0.4% 0.8% 1.5% Foreign loans (loan 0.2% 0.8% 2.0% 1.6% 2.1% 2.6% disbursements) Program loans (import 0.0% 0.5% 1.3% 0.6% 0.8% 0.8% support) Development project loans 0.1% 0.3% 0.8% 1.0% 1.3% 1.9% Amortization -1.4% -1.3% -1.1% -1.2% -1.3% -1.1% * annualized estimate based on prel. outturn for July-Feb. 2001 Source: Tanzanian authorities Government Expenditures 4.4 Government spending increased in FY00 to 15.6 percent of GDP compared to 14.7% of GDP in FY99. Expenditures during the first eight months of FY01 are estimated to be below the level 19 recorded in the previous year. The decline is mainly on account of reduced development expenditures. However, in past years government expenditures during the last quarter of the fiscal year have typically been higher than expenditures in the other quarters which could result in expenditures above the currently estimated 15 percent of GDP. Table 4.3: Government Expenditures (as % of GDP), FY96 - FY01 Central Government Operations FY96 FY97 FY98 FY99 FY00 FY01 * Total expenditure and net lending 17.6% 15.1% 14.7% 14.7% 15.6% 15.0% Recurrent expenditure 14.0% 12.5% 10.9% 10.7% 11.8% 12.0% Wages and salaries 4.6% 4.7% 4.2% 3.6% 4.2% 4.0% Interest payments 3.3% 2.6% 2.2% 1.6% 1.5% 1.6% Domestic 2.3% 1.7% 1.0% 0.6% 1.0% 1.0% Foreign 1.0% 0.9% 1.3% 0.9% 0.5% 0.6% Other goods and services and 6.1% 5.1% 4.5% 5.6% 6.1% 6.3% transfers Development expenditure and net 3.6% 2.6% 3.8% 3.9% 3.9% 3.0% lending o/w Expenditure financed 0.2% 0.5% 0.5% 0.3% 0.3% 0.2% domestically * annualized estimate based on prel. outturn for July-Feb. 2000 Source: Tanzanian authorities 4.5 The economic breakdown of recurrent expenditures shows that about one third of recurrent expenditures is spent on wages and salaries, 14 percent are spent on domestic and foreign interest payments, and more than 50 percent of recurrent expenditures are spent on other goods and services. While the problem of insufficient funding for operations and maintenance persists, during recent years there was some improvement in terms of increasing the share of the budget spent on operations and maintenance and containing the share going to the payment of interest and wages and salaries. 4.6 Since average salaries increased in nominal terms only by 1.3 percent in FY99 and the number of civil servants was reduced by another 2.7%, expenditures on wages and salaries remained more or less constant in nominal terms, implying a decline in expenditures on wages and salaries as a percentage of GDP from 4.2 percent in FY98 to only 3.7 percent in FY99. Salary increases below the rate of inflation have led to real income losses for civil servants of up to 35 percent since FY96. However, after further real wage losses in FY99, in FY00 government started the implementation of it Medium Term Pay Policy by granting salary increases ranging between 16 and 65 percent, with the higher increases being accorded to technical personnel and middle and upper management. In addition, the reduction in staff numbers was only 1.3 %, significantly less than in previous years. These large pay increases combined with only small reductions in the staffing levels led to a projected increase in the wage bill in FY00 by more than 30 percent and an increase in expenditures on salaries and wages from 3.7 percent of GDP in FY99 to 4.3 percent of GDP in FY00. In FY01 civil service employment increased by almost 7000. However, since no salary awards have been granted in FY01, expenditures on wages and salaries are estimated to decline to 4.0 percent of GDP. 20 Table 4.4: Civil Service Employment, FY98-FY01 (December of each year) Salary Scale FY98 FY99 FY00 FY01 FY98 FY99 FY00 Staff in absolute numbers Percentage change TGOS 35651 34806 35001 -2.4% 0.6% TGS 63787 63310 59994 -0.7% -5.2% TGTS 122215 118868 119566 -2.7% 0.6% TPSW + TGPSW 36190 34821 36448 -3.8% 4.7% OTHERS 12785 11381 8837 -11.0% -22.4% TOTAL 270628 263186 259846 266,718 -2.7% -1.3% 3.0% Source: CSD Table 4.5: Civil Service Average Salaries, 1996-97-FYOO Salary Scale FY97 FY98 FY99 FY00 FY98 FY99 FY00 TGOS 33707 37703 38212 47220 11.9% 1.4% 23.6% TGS 47099 57313 57625 78067 0.5% 35.5% TGTS Included 56314 57073 80162 1.3% 40.5% in TGS TPSW + TGPSW 65002 60506 61400 70925 -6.9% 1.5% 15.5% OTHERS 35344 115878 108011 167861 227.9% -6.8% 55.4% TOTAL 47498 54684 55387 74143 15.1% 1.3% 33.9% Source: CSD 4.7 Expenditures on other goods and services have increased continuously from a low of 4.5% of GDP in FY98 to an estimated 6.3% of GDP during the first eight months of FY01. Expenditures on interest payments declined from 2.2 percent in FY99 to 1.6 percent in FY00, as a result of a fall in both domestic and foreign interest payments. During FY00, foreign interest payments fell from I percent of GDP to 0.5 percent of GDP. However, this decline in foreign interest payments is offset by an increase in domestic interest payments to one percent of GDP. 4.8 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. However, problems with the proper integration of development expenditures in the budget continue. Data on official development expenditures collected by UNDP indicate that official development assistance is about 12 percent of GDP. Although a significant share of ODA goes directly to NGOs, the private sector, and the Bank of Tanzania, it is likely that the information on donor financed projects implemented by sector ministries and local authorities is incomplete. On the other hand, development expenditure in the appropriation accounts is only about one third of the Development Expenditure as a Share ofGDP, FY96-FY01 7.0%M 6.0'%,- DCGO 3 4.0% * Budgct Estimate °,3.()"e, l_ - ttU~~ffl4jII~~jjIIft1II~~~4~I1 0 Approp. Amts. 2.0% FY96 FY97 FY98 FY99 FY00 FY01 21 figures shown in the central government operations tables. Deficit Financing 4.9 Following the introduction of the cash budget in 1996, Tanzania recorded continuous budget surpluses (after grants) in the past five years with the exception of FY00, when small budget deficit in the magnitude of 0.6 percent of GDP was registered. During the first eight months of FY01, data show a surplus after grants of about 1 % of GDP. Taking into account net inflows from foreign loans, which also constitute development assistance and contain a significant grant element, the deficit for FY00 turns into a small surplus of 0.2 percent of GDP and for the first eight months of FY01 a surplus of 2.2 % of GDP is registered. These surpluses have allowed government to reduce its outstanding debt with the domestic banking and non-banking institutions and also repay arrears to a significant extent. The fiscal discipline enforced through the cash budget system was a key factor in restoring macro-economic stability in Tanzania. However, given the significant public expenditure needs identified in the PRSP in key poverty related areas, the issue of the appropriate fiscal position gains relevance and is discussed in the chapters on fiscal sustainability and on key systemic issues. Table 4.6: Financing of the Fiscal Deficit (% of GDP), FY96 - FY01 Central Government Operations FY96 FY97 FY98 FY99 FY00 FY01 l Overall balance before grants -4.3% -1.6% -2.8% -3.4% -4.4% -3.0% (checks issued or commitment basis) Overall balance after grants -2.2% 2.0% 0.2% 0.5% -0.5% 1.0% (checks issued or commitment basis) Overall balance after grants -3.0% 1.9% 0.2% 0.3% -0.5% 0.7% (checks cleared or cash basis) Foreign loans (net) -1.2% -0.5% 1.0% 0.4% 0.8% 1.5% Overall balance after grants and -4.2% 1.4% 1.2% 0.7% 0.3% 2.2% foreign loans (checks cleared or cash basis) Domestic (net) 3.3% -0.7% -0.4% -0.2% -0.1% -1.9% Bank 2.7% -0.4% -0.9% 0.0% -0.1 % -1.7% Nonbank (net of amortisation) 0.6% -0.3% 0.5% -0.1% 0.0% -0.2% Privatization Funds 0.5% 0.3% 0.1% 0.2% 0.0% 0.5% Change in arrears 0.4% -0.9% -0.8% -0.8% -0.2% -0.9% * annualized estimate based on estimate for July-Feb. 2000 Source: IMF, Tanzanian authorities 4.3. STRATEGIC RESOURCE ALLOCATION 4.10 This section presents the functional analysis of public expenditures for the period FY96- FY00 using data obtained from the appropriations accounts. In addition, data for the first nine months of FY01 are presented on an annualized basis based on exchequer release data from the flash reports. However, since under the cash budget system monthly releases are determined by monthly revenue collection outturns and foreign concessional inflows, these annualized figures for FY01 should not be interpreted as an estimate for the likely outturn for FY01. It is worth noting that beginning with the third quarter of FY01, the Ministry of Finance has provided quarterly exchequer 22 releases to the priority sectors, facilitated by the newly set up Poverty Reduction Budget Support facility. Table 4.7: Composition of Public Expenditures (as % of GDP), FY96-FYO1 FY96 FY97 FY98 FY99 FY00 FY01* Recurrent Expenditures 12.5% 13.5% 13.1% 12.8% 12.8% 12.7% Debt Service 3.7% 5.0% 4.9% 3.8% 4.2% 3.4% Recurrent 6.5% 6.2% 6.0% 6.9% 6.3% 6.7% Central Recurrent 2.3% 2.3% 2.3% 2.1% 2.3% 2.6% Regions Development 0.5% 0.9% 1.6% 1.8% 1.5% 0.3% Expenditure Total Expenditure 13.0% 14.4% 14.7% 14.6% 14.3% 13.0% * annualized expenditures based exchequer releases for the first nine months of FY01 Source: Appropriation Accounts (FY96-FYOO), Flash Reports (FY01) 4.11 Overall, recurrent expenditures as a percentage of GDP remained fairly stable over the period FY99-FYOI at around 12.8 percent of GDP.' However, as the share of recurrent expenditures used for debt service payments declined from about 37 % in the mid nineties to around 30 percent in recent years, a greater share of recurrent expenditures is used to fund ministerial and regional supply votes. Although debt service payments show a declining trend in the medium term, year to year fluctuations in debt service payments are relatively large with corresponding variations available to fund salaries and operations and maintenance.2 It is also interesting to note that during the past five years resources allocated for the delivery of decentralized services in the priority sectors covering education, health, water, and agriculture have remained fairly constant at around 2.3 percent of GDP. Based on expenditure releases during the first nine months of FY01, resources for decentralized service delivery appear to have increased to 2.6 percent of GDP which would be in line with the implementation of the government's decentralization policy and the prioritization of expenditures for poverty reduction. Table 4.8: Sectoral Recurrent Expenditures (actuals, as a %age of GDP) Sector IFY96 IFY97 |FY98 |FY99 |FYOO |FYO1* Administration 2.1% 1.4% 1.6% 2.2% 2.1% 2.1% Defence and Security 2.4% 2.5% 2.2% 2.2% 2.0% 2.0% Social Services 3.5% 3.5% 3.6% 3.7% 3.6% 4.1% Economic Services 0.2% 0.6% 0.4% 0.4% 0.7% 0.9% Productive Services 0.5% 0.4% 0.3% 0.5% 0.3% 0.3% Supply Votes 8.7% 8.5% 8.2% 8.9% 8.6% 9.3% Consolidated Fund Services 3.7% 5.0% 4.9% 3.8% 4.2% 3.4% Total Recurrent 12.5% 13.5% 13.1% 12.8% 12.8% 12.7% Expenditures * annualized expenditures based exchequer releases for the first nine months of FY01 Source: Appropriation Accounts (FY96-FY00), Flash Reports (FY01) The 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. 2 The figure for FY00 for expenditures under the supply votes obtained from the appropriation accounts appears to be almost one percent of GDP less than the comparable figure obtained from the Ministry of Finance on Central Government operations. Clarification is being sought from the Ministry of Finance. 23 4.12 Table 8 presents the functional classification of expenditures funded from the central government budget for the period FY96-FYOI. Although debt service payments have been on the decline during the past five year, until FY00 they claimed the largest share of recurrent expenditures and the government spent more on debt service payments than on the social sectors. The social sectors which include education, health, water, and the ministries for community development and women's affairs and labor and youth development. Expenditures in this sector were fairly constant during the past five years receiving about 3.6 percent of GDP with only a slight upward trend in recent years. 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 the central payment of electricity since FY99, 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. Taking all this into account, expenditure pattems are broadly in line with the objective of increasing social sector spending faster than other spending plans. Nevertheless, given that under the extremely tight resource constraints which Tanzania faces and the often commented on inability to fund operating cost for many 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. Nonetheless, there is a clear recognition that some specific administrative services such as funding for the Office of the Controller and Auditor General are essential for the proper functioning of government. Expenditures on defence and security have declined from 2.5 percent of GDP in FY97 to 2.0 percent in FY00. Table 4.9: Social Sector Recurrent Expenditures (actuals, as a %age of GDP) Sector FY96 FY97 FY98 FY99 FY00 FY01 * Education 0.3% 0.3% 0.4% 0.3% 0.3% 0.3% Water, 0.0% 0.0% 0.0% 0.1% 0.0% 0.1% Health 0.3% 0.4% 0.5% 0.6% 0.5% 0.4% Science, Technology & Higher Education 0.5% 0.5% 0.4% 0.5% 0.4% 0.5% Regions 2.3% 2.3% 2.3% 2.1% 2.3% 2.6% Total Social Services 3.5% 3.5% 3.6% 3.7% 3.6% 4.1% * annualized expenditures based on exchequer releases for the first nine months of FY01 Source: Appropriation Accounts (FY96-FYOO), Flash Reports (FY01) 4.13 Expenditures in the social sectors have been increasing continuously from 3.5 percent of GDP in FY97 to 4.1 percent of GDP in FY01. This reflects the priority status accorded to these services by government and additional support received from the multi-lateral debt fund, which was linked to enhanced funding of the social sectors. 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 PRSP, which can be expected to have the biggest impact on poverty reduction and economic growth. In this context, the benefit incidence of post primary education and tertiary health care should be guiding principles as to whether government should retain its current role in providing and financing these services. 24 Table 4.10: Tanzania: Priority Sector Spending - Other Charges. Actual Expenditure on OC as Percentage of Budget Estimate FY00 FY01 Education 97% 148% MOEC 100% 132% MSTHE 95% 132% Teachers Service Com. 110% 81% Local Government 98% 185% Regions 101% 103% Health 97% 137% MoH 97% 131% Local Government 98% 161% Regions 101% 103% Judiciary 66% 122% Judiciary 66% 122% Roads 81% 80% Road Fund 80% 80% Other Roads 98% Water 137% 103% MoW 158% 98% Local Government 98% 109% Regions 465% 0% Agriculture 41% 131% MOAC 41% 131% Extension Services Land 101% 110% Land 101% 110% Note: OC for FYOI cover July - December only. Source: MoF 4.14 Allocations for OC in the priority sectors have increased significantly in FY01. Available data for the first half of FY01 indicate that exchequer releases for the priority sectors were also significantly higher than budgetary allocations. For the last two quarters of FY01 the Ministry of Finance has introduced quarterly exchequer releases for the priority sectors and the priority sectors have received their full OC allocation at the beginning of the third and fourth quarter. 25 Table 4.11: Sectoral Development Expenditures (actuals, as a %age of GDP) Sector IFY96 FY97|FY98 FY99 FY00 Administration 0.1% 0.4% 0.4% 0.0% 0.3% Defence and Security 0.0% 0.0% 0.0% 0.0% 0.0% Social Services 0.2% 0.2% 0.4% 0.9% 0.5% Economic Services 0.1% 0.2% 0.6% 0.7% 0.6% Productive Services 0.1% 0.0% 0.2% 0.2% 0.2% Consolidated Fund 0.0% 0.0% 0.0% 0.0% 0.0% Services Total Recurrent 0.5% 0.9% 1.6% 1.8% 1.5% Expenditures * annualized expenditures based on first nine months Source: Appropriation Accounts (FY96-FYOO), Flash Reports (FYOI) 4.15 Recorded development expenditures show a continuous upward trend increasing from 0.5 percent of GDP in FY96 to 1.8 percent of GDP in FY00. This increase in development expenditures is partly due to a greater share of donor funded expenditures passing through the exchequer system, 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. Foreign Aid: From the Central Bank to the Office of the Auditor and Controller General 4.16 Tanzania benefits from huge aid flows which amounted to about External Project Support, Average FY99 and US$1.2 billion in recent years. Less FY00 than 20 % of the aid flows registered in the balance of payments are in the I- A form of program or general budget wo, _ - support and as such included in the _____ government's budget. The bulk of foreign assistance is however still in Or/. the form of project financing - _ - through grants and loans. The amount of project financing that 20%/. _ __ enters Tanzania is about 8% of -. -- _ f i GDP, an amount almost similar to P.im ODA (B0P) D"I UEP. D- 40-p. (Appop. the total of Tanzania's recurrent (B.dgt/CGO) Aces.) budget. However, little more than one third of project assistance to Tanzania are recorded in the Central Government Budget estimates. The other two thirds of project assistance are essentially outside the budget process and any official scrutiny. Of the one third of project assistance that is recorded in the budget estimates at the beginning of the fiscal year, again only about one third passes through the exchequer system and is finally recorded in the government's appropriation accounts, which are audited by the Office of the Auditor and Controller General The reasons for providing assistance outside the budget by directly working with sector ministries, local authorities, communities, or NGOs are manifold and partly related to Tanzania's history of poor management and accountability for public funds. However, huge aid flows outside the budget clearly undermine the budget process by weakening the incentives for good budget management if significant resources for essentially public tasks are available outside the formal budget process. The fact that only 16 % of project assistance is recorded in the appropriation accounts and audited raises the question about the accountability of the remaining 84 % of project assistance. Accountability mechanism for these funds generally aim at satisfying the 26 funding donor, with relatively less attention to accountability to local stakeholders. As the Tanzanian authorities make progress in improving public finance management and accountability mechanisms, it seems desirable to have development assistance linked to the public sector also properly integrated in the budget. 4.4. VARIATION BETWEEN THE APPROVED BUDGET AND ACTUAL EXPENDITURES 4.17 Previous PERs have noted significant variations between budget allocations as approved by the National Assembly at the beginning of the financial year and expenditures as recorded in the appropriations account. There are three principal reasons for these differences with respect to recurrent expenditures: > reallocations from the contingency account at the Ministry of Finance in the course of the budget year; > differences between projected and actual resources available; > under/overspending my ministries compared to the approved budget. 4.18 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 undertake reallocations across votes throughout the fiscal year. The budget speech presented to Parliament usually indicates how the contingency allocation will be used. At the end of the fiscal year, a statement of reallocation is presented to Parliament for approval. This contingency fulfills at least three purposes: > reduce the risk of having to borrow in case revenue collection and foreign inflows are less than projected; > keep aside funds for emergencies; and > cover increases in the wage bill which are typically not determined prior to the approval of the budget. 4.19 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 4.12: Contingency allocation retained by Ministry of Finance FY98 FY99 FY00 FYOI TSh. Billion 38.0 29.3 66.7 44.5 % of Supply Votes (Ministerial and 6.7% 4.7% 8.8% 5.1% CFS) 4.20 In recent years, the amounts allocated for contingencies has varied considerably, ranging from TSh. 29.3 billion in FY99 to TSh. 66.7 billion in FY00 (see Table 12). The relatively large size of the contingency in FY00 was necessary to provide for the planned implementation of Phase 1 of the pay reform and the parallel introduction of a medical insurance scheme for civil servants. 27 Table 4.13: Actual Reallocations to PE and OC FY98 FY99 FY00 Total 43,182,423,677.00 91,732,932,979.00 66,780,763,015.00 PE 20,388,229,000.00 9,679,534,453.0 53,208,424,454.00 OC 22,794,194,677.00 82,053,395,906.0 13,572,338,561.00 Source: MoF, Reallocation Warrant 4.21 Table 13 shows total reallocation undertaken by the Ministry of Finance during the past three years. In FY00, 80% of the contingency were used for reallocations to PE to fund the implementation of the first phase of pay reform. In contrast, in FY99, only 11 percent of reallocations were needed to fund salary adjustments, while 89 percent of the reallocations were used to fund OC and increased debt service payments. In FY98, the reallocations were almost evenly divided between OC and PE. 4.22 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 determnining 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. Table 4.14: Reallocations and Expenditure Outturn, FY00 Approved Reallocation Budget after Actual Actual Approved Budget Reallocation Expenditure Expenditure Budget after as % of Reallocation Approved as % of (Tshs. (Tshs. (Tshs. (Tshs. Budget after Approved Billion) Billion) Billion) billion) Reallocation Budget Ministry of 128.4 (67.4) 61.0 48.5 79.6% 47.5% Finance Administration 94.2 10.4 107.2 94.2 87.9% 113.8% Social Services 222.6 48.8 271.4 263.2 97.0% 121.9% Economic 54.4 2.1 56.5 23.0 40.7% 103.8% Services Productive 26.8 (2.1) 24.7 18.2 73.7% 92.1% Services Total Supply 526.4 (8.2) 520.7 447.1 85.9% 98.9% Votes CFS 266.3 29.0 295.3 287.7 97.4% 110.9% GRAND 792.7 816.1 734.8 90.0% 103.0% TOTAL Source: Budget books, appropriation accounts, and statement of reallocation. 4.23 Table 14 shows the allocation of this contingency in FY00 to the various sectors. The bulk of it went towards the social sectors, especially to local authorities to provide for salary increases at the service delivery level. A significant re-allocation took also place from the Ministry of Agriculture (productive services) to the local authorities to complement the shift of extension staff from the Ministry of Agriculture to local authorities. Consolidated fund services received a special requisition of TSh. 29 billion which was used for debt service payments in excess of the budgeted amounts. 4.24 Table 14 also shows actual expenditures in TShs. and as a share of approved budget estimates after reallocations. For the supply votes, actual expenditures were only 85.9 percent of 28 budgeted amounts. As statutory payments, salaries, and priority sectors including health, education, water, roads, the judiciary, and lands are protected from cuts, these expenditure shortfalls affected disproportionally OC expenditures of non-priority spending units. Spending units in the social services sector, which comprise most of the protected expenditure units, received 97 percent of the budgeted amounts. Spending units in the administrative sector received 88 percent of their budget. Spending units belonging to the economic services and the productive services were hardest hit by expenditure short falls, receiving only 49 and 74 percent of their budgets, respectively. As was observed in previous PERs, this underfunding of OC expenditures for non-priority ministries, which is compounded by unpredictable and fluctuating exchequer releases throughout the year, makes rational budget implementation virtually impossible. Mechanisms to alleviate this situation and to restore the credibility of the budget need to be pursued vigorously and are described in the discussion of systemic issues. 4.25 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. 4.26 Firstly, the problem of appropriately capturing planned donor project disbursements in the budget submitted to Parliament still persists. A proper system to capture donors' disbursement plans is still not in place and there is scope for improving information flows on disbursement plans between donors and government. While donors who provide planned disbursement figures to the government often find that these planned disbursements are not reflected in the budget as indicated, many donors also do not provide adequate information to the government. Further work is required to improve the capturing of donor disbursement plans appropriately in the budget. 4.27 In addition, even for donor funded projects that are contained in the approved budget, expenditures are often 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 4.15: Development Expenditures by Sector, Actual Expenditures as a Share of Budgeted Expenditures, FY96-FYOO FY96 FY97 FY98 FY99 FYOO ADMINISTRATION 8.3% 29.2% 17.5% 6.4% 101.5% DEFENCE AND SECURITY 18.6% 25.0% 34.4% SOCIAL SERVICES 12.2% 35.3% 24.6% 54.1% 32.6% ECONOMIC SERVICES 20.0% 1.5% 33.9% 52.7% 59.9% PRODUCTIVE SERVICES 66.6% 16.4% 91.4% 31.1% 36.5% GRAND TOTAL 14.0% 6.1% 27.5% 45.9% 52.2% Source: Appropriation Accounts 4.28 Table 15 shows the ratio of budgeted to actual development expenditures for the period FY96-FYOO. In recent years there has been a significant increase in the share of budgeted development expenditures actually recorded as expenditures. Between FY97 and FYOO, the ratio increased from 6.14 percent to 52.3 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. For example, a recent country portfolio 29 performance review undertaken by the World Bank indicated that the disbursement rate of World Bank credits almost doubled in the last 2 years increasing from 15% to 28%. 4.29 Provisions for counterpart funds in the amount of TShs. 3.01 billion made in the budget for various projects (e.g., the national agricultural extension project, Madibira smallholders rice scheme, rehabilitation of schools and colleges, Lake Victoria, environment management program, etc.) were reallocated for the set up of Television Tanzania. There is concern about the implications of this reallocation for the implementation of the affected projects. 4.5. PRO-POOR EXPENDITURE TRACKING STUDY 4.30 Allocating resources to pre-identified pro-poor expenditure programs is necessary but not sufficient for guaranteeing that public spending contributes to poverty alleviation. It is also necessary that allocated funds flow indeed to service delivery units and the intended beneficiaries. Since 1998, various expenditure tracking studies were undertaken to ascertain whether allocated funds reach the intended beneficiaries. With the implementation of the Local government Reform Program these expenditure tracking studies gain on importance, as greater responsibility for the delivery of basic services in priority sectors is being moved from central government ministries to local authorities. In the framework of the FY01 PER process another tracking study was carried out by ESRF and REPOA with the objective of assessing the efficacy of budget execution and the intention of drawing lessons that could be used to improve implementation of pro-poor public expenditures. The study focused on primary education and health, two of the eight sections/activities that have been identified to receive government priority in the fight against poverty. Rural roads and water which were to receive equal treatment in the study have featured only marginally because they have not established institutional linkage at the sub-district levels. 4.31 The study reviewed government procedures and channels for disbursing funds, reporting systems and the flow of funds from one level to the other. In addition to the flow of funds, an assessment was made on the flow of materials and supplies. 4.32 Both desk work and field work were undertaken. In the former a review of documents was made. In the latter visits and interviews were made in the relevant government ministries and local authorities. Five local authorities were visited, three rural and two urban, namely Babati, Kisarawe and Dodoma rural councils and Mtwara and Kigoma urban councils. In the councils documents were reviewed and interviews were made with staff. In addition service centers were visited to provide a feel about supply and availability of materials and supplies. At this level interviews with the beneficiaries were also held. 4.33 A number of observations have been made in the main report. Below we present the salient ones and draw policy implications. Firstly, the cash budget system has in principle relegated OC to a residual position. This has implications on the timing and level of OC disbursement. Lack of predictability on the disbursement of OC has promoted leakages especially at the sub-national levels. Other things being equal, increased predictability would enhance transparency and minimize leakages. 4.34 Secondly, the existing system of disbursing funds is in principle alright. However, it assumes adequate transparency in information sharing and transmission. The present review has found that in situations of scarce resources ownership of information and misinformation are important for redistributing resources to suit the owners of information. In the case of local authorities information contained in the exchequer issue notification has normally not been transmitted to sectoral heads. This has opened room for re-allocations, without the knowledge and consent of sectoral heads. The introduction of IFMS at the ministries level has improved albeit marginally the speed of OC disbursements. However, it has not reduced reallocations at the local authorities because it has not been implemented at that level. Since it will take a long time for the IFMS to be implemented at lower levels efforts should be made to build capacity to manage the existing manual system and 30 promote transparency there. On the latter, it is proposed that sectoral heads be given copies of the exchequer issue notification to minimize misinformation. 4.35 Thirdly, perhaps in response to budget cuts, at both the Treasury and the council, sectoral heads have tended to re-allocate the OC they receive in favour of activities that benefit the council staff at the headquarters, at the expense of service units. In this case traveling and vehicles have been favoured at the expense of school materials and health drugs. 4.36 Fourthly, councils have managed to get away with observations two and three above because financial reporting requirements have accommodated too much aggregation. It is important that an acceptable format be developed that can easily be related to the exchequer issue notification. 4.37 Fifthly, sectoral departments are the last point of cash flow in the disbursement system. There is no flow of cash beyond council departments. Only supplies and services flow to the service units. However, as stated above very little materials and supplies originate from the councils. Most of the materials distributed by councils originate from sectoral ministries. In this context survey results show that the supply of primary materials is limited and ad hoc. The lack of predictability may promote leakages. This is different from the supply of health drugs which is more reliable in both timing and quality. The perceptions of beneficiaries on availability of school materials and health drugs support the above observation. Beneficiaries are more satisfied with availability of health drugs than school materials. A closer look at survey results show that both the quantity supplied and reliability play an important role in influencing beneficiaries perceptions. It is important that the Ministry of Education work on improving both quantity and reliability, i.e. there should be a known schedule on the distribution of materials that is followed. 4.6. TRANSPARENCY AND ACCESSIBILITY OF FISCAL INFORMATION 4.38 One of the cornerstones for public accountability in fiscal management is readily available, reliable, and easily understandable information on public expenditures. As government continues to open up the budget process to various stakeholders inside and outside of government and as more donors are considering to either provide direct budget support or at least channel their assistance through the budget, these issues gain additional importance. 4.39 At present, fiscal information is not easily accessible and often difficult to interpret. In addition, there is also the perception that data from various sources are often inconsistent and pubic debate on fiscal issues often lacks a solid informational foundation. 4.40 With the introduction of the computer based Integrated Financial Management System and the reclassification of the budget according to the GFS, the transparency of the budget should improve and it is thus important to proceed with this initiative. 4.41 In addition, it is recommended that active efforts are undertaken to make information accessible and to provide easily understandable infonnation on key fiscal issues. It is thus recommended that the PER process in close collaboration with local economic research institutes pays greater attention to the dissemination of information generated by the PER process to the public. It is also recommended that in parallel government undertakes to provide information on budgetary performance to the public on a quarterly basis. 31 5. BUDGET SUSTAINABILITY AND FULL FINANCING OF THE PRIORITY SECTORS3 5.1 The main aims of this chapter are (i) to make an assessment of the resource envelope available for the FY 01-02 budget and for the MTEF period. This is approached through a review of the prospects for the major sources of resource availability - domestic revenues, external grants and loans and domestic financing; and (ii) to present estimates of the cost of fully financing the PRSP priority sectors under the 2001/2 to 2003/4 MTEF. The report also comments briefly on some aspects of expenditure control and the budgetary management systems in place to improve expenditure control (the use of first charges). 5.1. THE RESOURCE ENVELOPE Domestic Revenue 5.2 Tanzania's revenue performance (measured by the revenue/GDP ratio) is poor in relation to the average for Sub-Saharan Africa and has declined significantly in recent years, reflecting the substantial reduction in external taxes, relatively large tax incentives for new investments, and the continued downsizing of the parastatal sector coupled with informalisation of the economy. The recent decline in revenue performance (1996/97-1998/99) has been attributed to the growth in exemptions (OPM, 2001) and streamlining of the tax system and the elimination of a number of nuisance taxes. 5.3 There is significant scope for substantially raising domestic revenue in Tanzania. However the changes required, particularly with respect to necessary improvements in tax administration, are expected to take time. Thus, unless the government pursues a policy to significantly reduce exemptions in the short term, even with significant tax effort, we do not expect to see any substantial improvements in domestic revenue during the life of the current MTEF. These observations support the cautious revenue projections included in the Budget Guidelines. External Grants and Loans 5.4 The analysis suggests that the Budget Guidelines are conservative in their estimates of the likely flow of external grants and loans over the MTEF period. This is resulting primarily from the methodology used by the MOF (and the supporting donor community) which tends to identify only fairly firmly confirmed donor financing opportunities. This is not necessarily the appropriate budgeting practice, which should be based on most likely outcome (with allowance for risk). Domestic Borrowing 5.5 Prudent budgetary policies in the past 5 years have led to a reduction in the domestic debt burden, measured by domestic debt as a proportion of macro aggregates (govermnent revenue or GDP). Successful inflation control has also led to a potentially sharp reduction in the cost of servicing the outstanding debt as the yield on Treasury Bills has declined to low nominal and real rates. 3This chapter is the executive summary of a report on "Budget Sustainability and Full Financing of the Priority Sectors" prepared by Emerging Market Economics Ltd. and funded by DFID as a contribution to the PER process. 32 5.6 The commercial banking system remains very liquid, with liquid assets accounting for per cent of total commercial bank assets, opening up the possibility of some renewed recourse to domestic borrowing, should this be appropriate There are several explanations for the high degree of commercial bank liquidity which combine structural and transitory factors. The restrictions on lending by NBC during restructuring and high liquidity arising from expansion of the mining sector are among the transitory factors, while the limited outreach and lack of experience in private sector lending are more structural features. The interest rate structure is characterized by a divorce of commercial lending rates from the BOT's discount rate or the TB rate. 5.7 In this situation it is implausible to argue that modest net government borrowing would crowd out private sector lending. Equally implausible is the frequently cited view that low government borrowing can crowd-in private sector lending, given the structural market constraints and the lack of an integrated interest rate structure. 5.8 However, there are important reasons for continuing to take a cautious approach to renewed net borrowing: any borrowing from the commercial sector would be short-lived as the overall volume of excess liquidity would be rapidly absorbed by any significant shift in government policy; any significant net borrowing is likely to be at a high budgetary cost as a rising TB yield progressively affects both new debt and rolled over debt; until such time as expenditure arrears have been cleared from the system, any new borrowing would in effect be financing the build up of arrears. 5.9 There is a need for changes in the reporting of TB market transaction by the BOT to enable researchers to analyze adequately the determninants of yields in the TB market. BOT should publish information that allows the determination of net TB sales on a monthly basis. 5.10 Domestic debt policy should be based on a range of considerations, rather than observance of a rigid "target" in the form of the ratio of debt or debt service to some macro aggregate. The study's views on the main elements to be taken into consideration at present include: cost-effective management of the outstanding debt stock, private sector crowding-out issues, and the need for expenditure side controls to be fully in place before there is any renewed recourse to net domestic lending. 33 5.2. THE EXPENDITURE SIDE Full Financing Costs for the Priority Sectors 5.11 The study attempted to calculate the cost of fully financing the requirements for achieving the PRSP targets in the priority sectors, drawing primarily on the sectoral reports prepared by consultants during the first quarter 2001. The main purpose of this exercise was to identify the degree of financing gap between the full financing requirement and the resource envelope. 5.12 A review of the sectoral consultants' reports revealed several types of obstacle to calculation of the desired type of full financing estimates: * for some sectors the definition of targets in the PRSP is either unclear, insufficiently detailed or only loosely linked to sectoral investment or service delivery programmes; * some sectors have not explicitly adopted the PRSP targets, preferring to continue with their own sectoral strategies which have not been fully integrated with the PRSP process; * some sectoral programmes, at least as judged from the sectoral consultants reports, have not yet been conceived or detailed to the point where they provide a plausible approach for meeting sectoral PRSP targets either at all or in the timeframe envisaged by the PRSP; * almost all the sectors have so far failed to define detailed action plans and activities which form a basis for a disaggregated and quantified approach to PRSP target achievement. This is true even for the best organized sectors; * the basis for costing of achievement of PRSP targets remains poor in most sectors, largely because of the lack of detailed activity plans, which are a necessary condition for preparation of scaleable cost estimates; * there remain important implementation and absorption capacity constraints in most sectors which mean that the realistic upper bound for effective expenditure in support of the PRSP targets remains below the level of the proposed full financing level of expenditure; * per contra, in some sectors the estimate for full financing has been based too closely on the sectors' expectation of likely resource availability as indicated in the Budget Guidelines, leading to full financing estimates which substantially understate the resources which could and should be made available to meet PRSP targets. This appears to be true of the Judiciary and Water sectors. 5.13 As a result of these problems and shortcomings the full cost estimates need to be interpreted carefully for each sector on the basis of the stage which the sectoral ministries have reached in strategy and budget formulation and in the light of the pervasive implementation and absorption capacity constraints. 5.14 Aggregation of the estimated full financing costs for the 8 priority sectors suggests an under- financing in the Budget Guidelines of Tsh 233bn in 2001/2 and Tsh 166bn in 2002/3. This would constitute roughly 1.5 per cent of GDP and of a magnitude which it might prove possible to accommodate in the resource envelope. 5.15 However, attention is drawn to the severe implementation constraints which are prominent in most sectors which will place a ceiling on the effectively utilizable funds in the PRSP priority sectors at least in the near term. Key implementation constraints arise from: * Limited procurement capacity (especially in the Roads sector - both MOW and MORLG); * Limited capacity at the district level to administer the roll-out of sectoral service delivery programs; 34 * Constraints in the staff remuneration and incentive systems which limit the ability of the sectors to achieve a rational deployment of staff to areas of greatest need for poverty reduction; this implies that even if additional funds are provided for Personal Emoluments, the effectiveness of the spend for poverty reduction will remain muted; * Constraints to the release of approved funds arising from the cash budget and expenditure management systems (including the regulations in place for the Health Sector Basket). Specific recommendations arising from the review of full financing costs are: Recommendation 1: The anticipated revision of the PRSP should be undertaken as a matter of urgency on completion of the current budgetary cycle, with a view to establishing a firmer basis for sectoral strategies and plans in the lateryears of the MTEF. Recommendation 2: In order to permit less tightly constrained growth in the later years of the MTEF high priority should be afforded to measures which address and reduce existing constraints for programme roll-out in the PRSP priority sectors. Recommendation 3: The PER Committee should review the adequacy of the existing and prospective provisions for the sectoral ministries to apply sufficiently remunerative packages to attract key staff required for programme roll-out in remote and poor districts, with a view to ensuring that this constraint, which needs a systemic rather than sectoral solution, is adequately addressedt 5.3. ASSESSMENT OF FISCAL SUSTAINABILITY 5.16 Sustainability is treated as a composite and probabilistic concept. At this stage of the PER a key aspect of the assessment of sustainability relates to the confirmation of the plausibility of the projections set out in the Budget Guidelines, and a substantial part of this report has been concerned to assess and comment on those projections using our own independent estimates of low, medium and high scenarios for each of the major aggregates and their principal elements in the resource envelope. 5.17 The overall conclusion drawn at this stage is that the Budget Guidelines (December 2000) provide a robust basis for budgetary planning. However, this result depends on a number of off- setting elements contributing to this conclusion: we believe that the Budget Guidelines are appropriate for domestic revenue, but conservative for contingent liabilities and for external grants and loans. 5.18 The full financing cost estimates, such as they are, point to some scope for increased PRSP target achievement through the allocation of incremental resources in addition to those anticipated in the PRSP and the December 2000 Budget Guidelines. This scope, is, however, limited by the poor state of several sector's preparedness to implement the PRSP and the strong implementation constraints faced in most if not all sectors. 5.19 There is, in principle, scope for the renewal of net borrowing from the commercial sector given the current liquidity of the commercial banks, and the reduction in the burden of domestic debt service in recent years (which has resulted from GDP growth and falling interest rates). However, this option should probably not be pursued till firm control of arrears and expenditure planning has been achieved. 35 6. REVENUE ISSUES IN THE LONGER-RUN4 6.1. THE CURRENT POSITION 6.1 The share of domestic revenue in GDP, at around 11.7%, is low in Tanzania relative to that in several other Sub-Saharan countries. The Budget Guidelines state that "raising the revenue to GDP ratio will be given high priority", but accept that over the next three years the budget share will be nearly flat5. The Guidelines thus reflect a tension between a normative approach to the budget (what we would like to see happen) and a positive approach (what we actually think will happen). 6.2 In the budgeting context, the positive approach must be right. Building-in implausible revenue forecasts would expand the perceived resource envelope and hence the magnitude of planned expenditures beyond what was likely to be financeable in practice. Unless there were offsetting errors, it would therefore lead to a built-in bias so that, in practice, expenditures would have to be curtailed below the budgeted amounts6. Since different components of spending are differently compressible, this is not only inefficient, but is also likely to lead to unplanned shifts in composition. 6.3 However, while the normative consideration should not be allowed to disrupt the budgetary process, revenue remains a serious issue. It appears to be an implicit rule of thumb that governments in poor countries can usefully deploy resources at least of the order of 20 per cent of GDP, and this seems now to be accepted as an appropriate broad target for Tanzania. With a revenue share of less than 12 per cent, this leaves a very large gap for financing. In view of Tanzania's high current approval rating with donors, they seem happy to provide the bulk of the requisite flow. However, it is one thing to do this against a background where the gap is being narrowed albeit slowly by continued relative growth in domestic revenue: it is another matter if the gap appears to be stationary and hence indefinite. It does seem important, in consequence, for the government to take a serious look at the issue of stagnant revenue: whether it is desirable to raise it, and if so, how that might be accomplished; if not, how that can be squared with the longer run issues of size of government operations and aid dependency. 6.4 There have been a number of reforms of the tax regime, and overall, the problem does not seem to rest in the design or rate structure of the system. Also, neighboring countries have very different revenue performances from their somewhat similar systems. Performance in Uganda is very similar to that in Tanzania, at around II%-12% of GDP, whereas Kenya's performance is much stronger, with the tax share averaging roughly twice as much. It is tempting to attribute this to Kenya's higher per capita income and more developed manufacturing sector, but it is necessary to be cautious about this. Recent cross-country analyses of tax revenue performance tend to suggest that Tanzania should be capable of generating substantially higher revenues, even when her low per capita income and disadvantageous economic structure are controlled for. As an illustration, we can feed Tanzania's (1999) characteristics into the tax equations estimated for 39 Sub-Saharan African countries by Ghura'. Using his base regression, which includes only per capita income and economic This chapter was prepared by David Bevan as part of his report on "Budget Sustainability and Full Financing of the Priority Sectors." It is projected to remain at 11.7% over the next two years, and then rise to 11.9% in 20023/04. 6 In practice, there is a potential offsetting error; this is the tendency to underestimate future donor flows because they are not yet committed, discussed in the text. However, it does not seem appropriate to budget on the basis "two wrongs [may] make a right". 7D. Ghura, Tax Revenue in Sub-Saharan Africa: Effects of Economic Policies and Corruption, IMF working Paper, WP/98/135. 36 structure, Tanzania should apparently be capable of generating a revenue-to-GDP ratio of nearly 18% (or 19.5% if Tanzania's status as a non-oil mineral producer is taken into account). These regressions should certainly be taken with a large pinch of salt, but they do suggest that Tanzania's relatively poor performance cannot be fully attributed to her unfavorable initial conditions. 6.5 Taking the longer view, it is worth noting that the growth in the revenue share would be (in the Ghura model) only around 3 percentage points even if per capita income were doubled, something which would require twenty years of growth in real GDP at rates in excess of 6% per annum. Once again these estimates should be treated with considerable caution. But they do indicate that growth alone is unlikely to resolve the revenue problem. 6.6 There is also the reverse connection to consider, running from taxation to growth. Tanzania has recently succeeded in raising the real GDP growth rate to over 5% and plans to raise it further to (and maintain it at) over 6%. However, the investment share in GDP has been quite low in recent years (around 15%); to achieve a sustained growth performance without increasing this would imply an incremental capital output ratio of only a little over 2, which is unlikely to be achieved over a long horizon. The continued high growth of GDP on which the government is relying is therefore likely to require a substantial increase in private investment. While international evidence suggests that a high tax share is not inimical to rapid growth, this evidence relates to countries where broadly based taxes at moderate rates raise large revenues. It certainly would not sustain a recommendation to raise revenue share by setting high rates on narrow bases. 6.7 The long-run spending implications of these unresolved domestic revenue issues can hardly be over-stressed. At present, the government's recent record as a good macroeconomic manager, and evidence of its commitment to a comprehensive reform agenda, mean that donors are prepared to underwrite something close to a 'normal' spending programme, despite Tanzania's relatively modest revenue performance. It is of course possible that they will be prepared to do so indefinitely. However, it would not be wise to plan on this eventuality. More immediately, there must be some risk that a failure to address the revenue issue will begin to tarnish Tanzania's other achievements, and might conceivably lead to an earlier decline in net aid flows than would otherwise occur. This would involve something rather like a 'matching grant' phenomenon; an extra shilling of domestic revenue would be accompanied by an enhanced aid flow, and vice versa. It should be stressed that this argument is both speculative and (probably) long run. But there are real risks nonetheless. 6.2. THE REVENUE PROFILE IN THE MEDUM TERM 6.8 This section discusses considerations that are relevant to the choice (and attempted implementation) of the revenue profile in the medium term. Since it utilizes a somewhat idiosyncratic, though quite natural, use of terms, this is spelt out first. 6.9 'Revenue effort'8 is usually taken in the tax literature to be the number here called 'revenue performance', typically the share of revenue in GDP. The common usage implicitly assumes that the input (which is here called effort) is exactly measured by the output - a low level of achieved revenue is taken to reflect a proportionately low level of the effort to raise that revenue. Without wishing to deny that there is a positive relation between the two, it seems implausible that this relationship should be linear, as opposed to reflecting some sort of diminishing returns. We make the plausible assumption, instead, that increased effort would lead to increased revenue, at least over a range9, but 8 The discussion is equally valid for the closely similar but somewhat narrower concepts of tax effort and tax performance. 9 Beyond that range, the converse will be true. The idea that there is a revenue-maximizing rate for any tax, so that any further rate rise actually reduces revenue, has been made familiar in the concept of the 'Laffer curve'. 37 at a diminishing rate. Devoting an extra ten percent of resources to collection, or raising rates by ten per cent, would increase revenue, but by less than ten percent. 6.10 It also seems plausible, and there is some evidence in support, that these diminishing returns set in more sharply if the attempt to raise performance is rapid rather than gradual, at least if this is attempted by rate rises. Conversely, lowering rates when these had been widely perceived to be unjustifiably high can improve collections by increasing the system's legitimacy and hence compliance. However, in Tanzania, while there is scope for further design improvements, it does not seem likely that these will of themselves generate substantial gains in revenue, legitimacy or compliance. 6.11 Even in the literature which presupposes a linear relationship between effort and performance, it is widely acknowledged that the slope will differ between countries, reflecting differences between them in the practical difficulties of collecting taxes. (See the discussion of the paper by Ghura above.) Once these difficulties have been taken into account (typically by regression analysis, using proxies for the various recognized dimensions of "tax-collectability"), a normalized revenue performance can be computed for a country, which acknowledges its own idiosyncratic characteristics. Any discrepancy between the normalized and actual performance is then ascribed to differences in its effort relative to the average in the sample. In particular, a low ratio of actual to normalized revenue is taken to imply a proportionately low level of effort. 6.12 There are two problems with this procedure. The first is that it assumes that the residual category "effort" is uncorrelated with the tax-collectability parameters. But this seems very unlikely, and significant correlation would obviously lead to misleading inferences. To illustrate, suppose low- income countries, facing severe difficulties in raising revenues, systematically tried harder than richer ones. Then the coefficient on per capita income would be biased downwards; a low-income country whose effort was high relative to the overall average but low relative to the low-income group would be categorized as low effort relative to both. If the correlation had the opposite sign, effort levels would again be biased, and the regression would now exaggerate the benefits of income growth for revenue. 6.13 The second problem is that the residual category is not really effort at all, but a composite of tax design, tax administration, compliance, and corruption. A low level of "effort" could therefore reflect unduly low tax rates or excessive exemptions; it could reflect lax and inefficient tax administration; it could reflect a culture of non-compliance; or it could reflect a high level of corruption. The latter two difficulties are often treated as two sides of the same coin, but there is a material difference between them. Poor compliance will give a tax administration problems even if it is completely free from corruption. Most tax administrations depend on a high degree of voluntary compliance by taxpayers coupled with a reliable system of enforcement, which itself requires an appropriate culture. 6.14 The effort calculation - for what it is worth - suggests low effort in Tanzania. This poses a choice. It would be possible to discard the calculation, and argue that Tanzania's structure and circumstances make it appropriate to settle for a low level of revenue generation for the time being. This argument would have to rest not on the proposition that a country in these circumstances could not raise substantially more - there are counter examples to show the contrary. It would have to rest on the proposition that such a country should not raise more, on the grounds that the damage inflicted to private sector activities, income and growth would be excessive. By extension, other countries in similar circumstances but with higher revenue collections should be encouraged to reduce their revenue effort. 6.15 The alternative is to accept the calculation, and consider ways of improving performance. Since the structure does not seem to be characterized by unduly low rates, the explanation must lie in one or more of the other components, but these imply very different strategies. Particularly difficult to tackle is the issue of a culture of non-compliance. Also, while the perception that revenue performance is poor seems naturally to imply that revenue effort should be increased, it is far from 38 clear how large an increase in effort would be justified. Indeed, it is unclear how far and how fast the revenue to GDP ratio could be raised, even if the government was to devote all available energies to raising it. In principle, it would be possible to carry out a cost-benefit analysis, weighing the marginal costs (including those borne by the private sector) of increased efforts against the benefits. In practice, the information for such an analysis is lacking. 6.16 In present circumstances, where donors are prepared to fund Tanzania government operations very generously, the pressing issue may be one of absorptive capacity, rather than the contribution of domestic revenue. In other words, even given the low level of domestic revenue mobilization, donor supplementation could be carried to the point at which severely diminishing returns to government spending would set in. In this context, raising domestic revenue may be a low priority. If it were possible to raise revenue performance reliably and quickly by adopting certain clear cut measures to increase effort, then there would be a strong case for leaving the revenue picture alone, and concentrating on other matters, until such time as absorptive capacity ceased to be a problem, and/or donor flows began to subside. However, there does not appear to be any quick fix of this type, so that prudence suggests that the attempt to improve revenue performance should not be delayed until funding shortfalls have already become an issue. 39 7. RECOMMENDATIONS 7.1 SYSTEMIC ISSUES IN BUDGET MANAGEMENT 7.1 One of the main objectives of the Tanzania PER process is to support the improvement of budget management. This goal goes beyond ensuring sustained fiscal balance consistent with macroeconomic stability, which Tanzania has achieved in the past five years. This chapter highlights four key systemic budget management issues in Tanzania, identifies the key challenges faced, and recommends actions to deal with them. Some of these recommendations, which were drawn during FY01 PER missions in December 2000 and May 2001 and presented to the Government of Tanzania, have already been adopted and implemented as shown in Chapter 8. 7.1.1. Strategic Allocation of Scarce Public Resources 7.2 Until FY00, strategic budget allocation primarily targeted the enhancement and protection of expenditure in the social sectors. Tanzania being a signatory to the Copenhagen Social Summit (1995) agreed to pursue the international development targets largely focused on raising and protecting expenditure allocations to the education and health sectors. Development partners likewise prioritized their support, including that provided through the Multilateral Debt Fund, to the same end. Sector development programs were designed without explicit link to overall poverty reduction outcomes or exploiting synergy with other sectors in achieving this goal. 7.3 In FY00, the focus shifted more explicitly to reducing poverty (first with TAS and then PRSP) in all its dimensions, through actions in a wide range of interrelated sector activities. This shift has entailed paying attention to reducing material poverty (income poverty) in addition to the previous emphasis on increasing social welfare. To this effect, the government's strategic expenditure allocation through the MTEF was modified to also explicitly target support for growth, particularly in rural areas, and to increase income earning opportunities for the poor. In this regard, expenditures on health and education were to be valued not only for their impact on improving social welfare but also for enhancing the capabilities of the poor to earn incomes - the human capital dimension. 7.4 There are notably three key challenges in enhancing the cohesiveness of strategic resource allocation under the new approach. One challenge is to harmonize the sector development programs, which were precedent to the PRSP, with sector-specific programming of activities as a means for achieving specific poverty reduction targets. It requires adapting the sector development programs to the poverty reduction goals in the PRSP and seeking greater cross-sector cohesion in activity programming to achieve the desired results. For the health and education sectors, the gap between the sector programs and the requirements for achieving social welfare targets (and human capacity building) in the poverty reduction strategy in broad terms is not wide. However, it will require two types of deliberate actions. One is a clearer mapping of inputs to outcomes, and the other is a more deliberate focus on disadvantaged locations and socio-economic groups. The rest of the sectors can anticipate the need for harmonization as they prepare their development programs. This is particularly urgent for the agriculture and transportation/roads strategies, currently under preparation. 7.5 Second is the need to resolve the emerging tension between sectoral and central agencies over responsibilities and authority for strategic expenditure planning. Sector ministries maintained a significant autonomy in negotiating external financial support with donors, often within a sector development program but largely outside the overall strategic expenditure allocation framework. Project support to sectors was channeled largely outside the exchequer system, weakening the integrity of the budget process and management. The process of integrating external resources into the cross-sector Medium Term Expenditure Frame (MTEF) and budget has changed the relations between the budgetary authorities and spending units with respect to the autonomy of sectors to pursue their own ends. The tension can be resolved partly by delineating and specifying the 40 respective roles of sectors and central ministries in the budgeting process to minimize unproductive competition and formalize the evolved process in well-documented budgeting procedures. Furthermore an inclusive planning process, involving the budget authorities, sector ministries and donors inter alia, as currently being done under the PER Working Group provides a useful mechanism for harmonizing dialogue. 7.6 Third is the need to more explicitly involve the local authorities (LA) in the national budgeting process and strategic expenditure allocation decisions. The share of budget allocated to the LAs is approximately 17 percent (excluding large off-budget donor support). These sub-national entities have the primary responsibility for delivering basic services (e.g. primary education, primary health care, agricultural extension services and rural infrastructure) involving significant shares of total spending in the priority areas (e.g. nearly 55 percent of total spending on education). Prior to the inception of the Local Government Reform Program, the local authority delivery system was essentially a deconcentration of the sector programs under the relevant sector ministry. With more autonomy envisaged for local authorities under the reform program, there is greater need for integrating their budgeting processes into the cross-sector and sector MTEF preparation. 7.7 Two measures are needed in moving towards this direction. One is the harmonization of the fiscal calendars. The LAs operate on a calendar year in contrast to July-June fiscal year for the central government. This difference leads to some disconnect in the planning process and lack of accurate foresight for expenditure planning for half of the operating year. The uncertainty in budgeting is particularly acute since LAs largely depend on the central budget subventions (conditional grants, capital transfers and shared revenue), accounting on average for about 76 percent of their financing sources. New budget guidelines to LAs provide guidance for expenditure planning in an MTEF approach, but have still to be implemented. The other measure is devising a mechanism for involving the 113 LAs in the MTEF process, both in terms of capacity requirements and logistics for participation. To enable grass root participation in the budgeting process, it would be necessary to begin the process early. The regional administration secretariats and PORALG would consolidate the inputs from the sub-national levels and use these inputs in the preparation of the national budget guidelines (for prioritization) and subsequently local authority MTEFs. The budgetary authorities at both levels would need to ensure consistency with both the projected national resource envelope and agreed public service delivery standards. 7.1.2. Enhancing the predictability of resource flows for improved service delivery 7.8 The cash budget system was introduced in FY96 with the primary aim of instilling fiscal discipline in spending units and restoring sustainable budget balance by limiting spending to available cash. There is no doubt about the success of this policy, having helped the country achieve sustained macroeconomic stability. Tanzania has been able to keep its fiscal balance either in surplus or sustainable deficits, since the introduction of the cash budget system. Combined with a tight monetary policy, which precluded government borrowing from the central bank to finance its expenditures, it enabled a sharp reduction in inflation over the same period to the current single digit levels. It is worth noting that the introduction of this system came immediately after Tanzania had gone through three years of very poor fiscal performance (FY92-FY95), during which the government lived well beyond its means. 7.9 Five years after the system became operational, all spending units, at national and sub-national levels, now express concerns about the stringency of the cash budget system. During the FY01 PER mission in December 2000, concerns were raised by all sector ministries and 14 local authorities visited. The main problems cited revolve around uncertainty faced by implementation units in planning their spending as monthly disbursements were based on actual revenue collection and disbursement of donor 41 funds, resulting in under-funding of requirements at the time large needs arise. The pressure from such stringency has led in some cases to commitments outside the official system and the build up of payments arrears. A comparison of expenditure on first claim items (debt service and salaries) with other charges and development expenditure for FY99 showed as expected that first claims have been largely constant and fully funded. On the other hand, development expenditure and operating charges were treated as residual, whenever there were cash shortfalls. In FY00, problems with the technical operation of the cash budget system led to a temporary lack of synchronization between the exchequer releases and availability of cash (from domestic revenue and external budget support) as illustrated in figure 1. There has been a very significant reduction in the gap in FY01. 7.10 There are three main types of budgetary stresses associated with large variations in the availability of cash and the associated under-funding of activities, which contribute to pressures for undermining prudence in budget management. Fig 1: Monthly Fiscal Deficit/Surplus (Tshs billion) 140.0 - 120.0 100.0 Monthly Exchequer Issues .2 80.0 -4-Total Resources Available ,, 60.0 40.0 20.0- 0.0 a s. H t b s i o o oa o~ 0 Ca o 0 0 0 .L 0)0 0)0 0) 0) 0 0 0 000 Month (a) Under-funding of first-charge type expenditures 7.11 The first relates to the overall amount disbursed being far short of requirements for effective delivery of services. Frequently, this problem has been addressed in terms of enlarging the overall resource envelope and sector as well as sub-sector prioritization. However, the budget stress is accentuated by the fact that the budget preparation stage has typically overlooked completeness in activity prioritization (sub-items). Currently statutory payments (CFS) and personal emoluments (PE) are treated as first charge. There are at least two more first-charge type of expenditures, which if under- funded often lead to pressures for making commitments outside the budget and building up of arrears. These are utilities (power, water, telecommunications) and "automatic catering or upkeep needs" such as food and upkeep needs of prisons, security forces, boarding schools and hospitals. Although these do not constitute an exhaustive list, together they account for the largest proportion of all arrears built up. For example based on the Controller and Auditor General's Report for FY99 the Police Force, Prison Service 42 Department, Defense (Ngome), and National Service accounted for 75 percent of all outstanding liabilities (43.2 billion T Shs) of the Central and Regional Governments. The bulk of these new arrears are in the form of suppliers' credit related to procuring these items. Preliminary analysis of arrears accumulated since then shows a similar dominance by these votes. 7.12 The government was encouraged to review the classification of first charge expenditures for completeness and ensure that budget guidelines for future years include instructions for fully funding these. In this regard, all spending units should provide reasonable estimates of these in their MTEF and budget submissions and be held accountable for staying within limits of the budget. (b) Cash Flow Gaps 7.13 The second dimension of the budget management problem relates to the mismatch between the monthly cash requirements and exchequer releases - amounts and timing of release. The lumpiness of some expenditure categories such as road maintenance after the rain season, pension funds, and examinations present major stresses in the overall functioning of the government budget when they occur without provision for cash flow gap smoothing. The pattern of expenditures through the year varies greatly across sectors to render a general rule of releasing 1/12 of the -approved budget each month disruptive to the effective provision of public services. Generally cash flow gaps have emanated largely from sharper variations in monthly expenditure requirements since the revenue patterns have tended to be more stable. 7.14 The monthly cash stress is typically accentuated by unforeseen expenditures triggered by emergencies. The main spending units that have to deal with such crises include Health for epidemics, Defense and Home Affairs for security-related emergencies, and Works and Agriculture for flood disasters entailing emergency repairs of infrastructure and food security. In the last FY it included handling of cholera outbreaks, refugee situations, food relief and security in the game parks. While it is possible to preempt some contingencies, such as dealing with tourism protection through a program of strengthening security presence in the parks, the bulk of other unforeseen events can only be dealt with on a contingency basis. 7.15 The delay in exchequer releases adds to further uncertainty in expenditure planning. Since there is a predetermined rule to base releases on the average of three previous months revenue collection plus projected external program support, it is hard for spending units to understand the delay. One of the causes of delay is a process of conforming to the monthly monetary programming necessitating protracted exchanges with BOT before finalizing decisions for monthly exchequer allocations. Such exchanges could be programmed in a way to enable meet the stipulated date for exchequer releases. 7.16 Even though releases for OC to priority sectors (under the quarterly commitment system) are made a quarter in advance, this does not solve the problem of possible disconnect with the timing of accruing resource flow. To deal with the problem of cash flow management the mission put forward four actions to be considered by the government: 7.17 The Budget Guidelines should require spending units to present a cash flow plan along with their annual budgets, which conform to the respective aggregate ceilings but clearly show the time pattern of their cash requirements. New budget guidelines to LAs now include this requirement. Particular attention needs to be paid to the typical lumpiness in expenditure patterns in the spending units. Currently submissions from some ministries include cash flow needs for some lumpy activities, such as examinations for education and immunization programs from health. The suggestion here is to systematize and broaden the coverage of such submissions to all spending units. Once consolidated the overall budget could also have a cash flow plan to provide a more systematic projection of exchequer releases. 43 7.18 The government was encouraged to consider a cash flow smoothing instrument that respects the overall budget ceilings for the year. Two options or their combination, were discussed with the concerned stakeholders. One is a cash reserve operated as a government deposit that can be drawn down or built up, depending on the net cash requirement position compared with the available cash. There were indications of possible contribution to such a reserve by donors, provided it can be protected from becoming a financing instrument. The other option is to establish a prudent credit line with the BOT, with advances subject to the same rule that the net position as end year be zero. The law allows for such advances up to the value of one eighth of the average revenue collection for the past three years provided they are repaid within 180 days. Prior to the recent up-front disbursement of budget support by donors, the BOT allowed the government to draw down its deposits to pre-finance the shortfall between projected external program support and actual disbursement. The recommendation made here was to allow the use of advances as provided under the law with specific safeguards against potential abuse by turning advances into a financing instrument as it occurred in the past. The previous PERs and this current one established that typically the variance between actual and projected domestic revenue collection is smaller than that for external resource flows. Furthermore, there is a concentration of external inflows in the last quarter of the year, which can be exploited as a safeguard against such abuse. Given the cost of advances, it is in the interest of the government to minimize the utilization of this window for financing reversible cash gaps. Upfront disbursement of budget support and indeed better foresight of aid disbursement plans will help minimize recourse to this avenue. The government was therefore advised to consider the two options or their combination based on comparative cost assessment, more specifically, the opportunity cost of maintaining a cash reserve and cost of credit. 7.19 There are two minor caveats with regard to these devices, or at least as to the detail of their design. First, while they would both work, as intended, to smooth expenditure against intra-year fluctuations in government receipts, they would have the undesired consequence of injecting and withdrawing liquidity via reserve money as they did so. In principle, this would simply enter the BOT's calculations of how much liquidity paper would be required to offset these movements (as well as movements in other autonomous factors). However, it might be better to back the government smoothing deposit by an increase in foreign exchange reserves from the outset. Excess spending in one month would then effectively be financed by drawing down foreign exchange reserves, so sterilizing the liquidity impact automatically; and vice versa when receipts exceeded spending. 7.20 The second caveat concerns the end-year position. While it might be believed with certainty that some fluctuations in receipts would be reversed within the current fiscal year, there is more usually a problem of interpretation. Is a downturn a temporary delay, which will be made good within the year? A temporary delay, which will be made good in a later year? A delay, which will only be made partly good later? A loss, which will not be made good at all? In practice, there are bound to be errors of interpretation, in both directions. Hence there will need to be some facility for carrying forward a limited positive or negative balance, provided there is a mechanism to prevent this process being cumulative. 7.21 A further step along the path of graduation from the strict cash budget is to move away from an annually balanced domestic budget to one in which the government may choose to run a domestic deficit. The prudent and efficient level of deficit would have to be chosen in the light of a number of factors. These would include the circumstances of the private sector, including the domestic credit position, the demand for real balances, and the saving and investment rates. They would also include the government's own debt position, the inflation, interest and exchange rates, and the value to be attached to additional government spending. Finally they would include a view of the extent of complementarity between public and private activities, and the absorptive capacity of each sector. 10 See Annex 2 for more discussion of this. 44 7.22 Once a level of deficit had been set, all other features of the budgeting process would operate as at present. In particular, protracted shortfalls in receipts would still require corresponding reductions in spending. In addition, the factors listed in the previous paragraph would have to be monitored continually, and unanticipated changes might require revision of the deficit target. The government will have to take a view on when it will be prudent to take this second step. 7.23 It was further recommended that, a separate line item be provided for contingencies under Vote 50 to minimize the stress from unforeseen expenditures on the overall functioning of the government. The current provision under Vote 50 predominantly provides for salary adjustments and regular shortfalls in OC. The FY01 PER mission heard from stakeholders that the existing arrangement does not permit to address these contingencies on a timely manner. To help speed up responses to emergencies, the government was advised to consider having reasonable estimates of emergency requirements based on past experience and to develop and agree on automatic triggers for release based on predetermined criteria. 7.24 Finally it was recommended that exchequer releases be issued on the stipulated date of the 5th of each month so as to minimize uncertainty on available cash. The timing of the meetings/exchanges between Treasury and BOT for ensuring conformity to the monetary program could be set within the stipulated timing target. (c) Payment Arrears 7.25 The third area of budget management stress relates to payment arrears. There are two dimensions to this problem - dealing with the accumulated stock of arrears and preventing the building up of new stocks of arrears through appropriate incentives. 7.26 Preliminary un-audited data on the stock of arrears (see Tables 8 and 9) show that there has been a sharp build up in the stock of arrears after the bulk of past stock were extinguished with the help of a program funded by EU (SASP IV) to reduce domestic debt. Audited stock of arrears are typically smaller with the difference largely accounted for by over-pricing of supplies and exaggerated commitments through collusion between some officials and suppliers to short change the government. Table 7.1: Structure of Consolidated Arrears- By Category July 1998 - December 2000 (percent share) VOTE VOTE HOLDER GOODS & UTILITIES OTHERS TOTAL SERVICES 28 Police Force 20.0% 12.7% 37.9% 20.3% 29 Prison Services 12.0% 10.7% 17.2% 12.3% 38 Defense 20.6% 47.7% 5.1% 25.6% 46 Education & Culture 1.2% 0.6% 3.8% 1.4% 52 Health 0.6% 0.6% 1.6% 0.8% 39 The National Service 2.5% 6.9% 0.9% 3.4% All Other Ministries/Departments 39.0% 20.4% 32.9% 33.6% All Regions - RAS 4.0% 0.3% 0.5% 2.7% TOTAL 100.0% 100.0% 100.0% 100.0% Note: Utilities include water, telephone and electricity; and Others include PE and allowances 7.27 Table 8 shows that the Police Force, Prison Service Department, Defense (Ngome), and National Service accounted for 62 percent of all central government new accumulation of arrears with the bulk of them being due to suppliers credit (upkeep and utilities). The structure of arrears also point towards the 45 dominance by suppliers' credit with the combination of goods and services and utilities together accounting for nearly 88 percent of total (Table 9). More than 60 percent of total payment arrears build up is on account of goods and services and can be controlled through the centralized system of issuing local purchase orders (LPO) instituted in early FY01. It is instructive to note that the bulk of new arrears in this category have occurred in agencies that had not been brought into the Integrated Financial Management System (IFMS), through which the centralized commitment system operates. About a quarter of the arrears are on account of utilities, which are outside the purview of the centralized commitment system until such time as the prepayment system will become operative. The rest of payment arrears (about 12 percent) are debts to employees related to personal emoluments and allowances. Table 7.2: Consolidated Arrears- As a Proportion of Total Arrears July 1998 - December. 2000 (percent share) VOTE VOTE HOLDER GOODS & UTILITIES OTHERS TOTAL SERVICES 28 Police Force 12.6% 3.2% 4.5% 20.3% 29 Prison Services 7.5% 2.7% 2.1% 12.3% 38 Defense 13.0% 12.0% 0.6% 25.6% 46 Education & Culture 0.7% 0.2% 0.5% 1.4% 52 Health 0.4% 0.2% 0.2% 0.8% 39 The National Service 1.6% 1.7% 0.1% 3.4% All Other Ministries/Departments 24.5% 5.1% 3.9% 33.6% All Regions - RAS 2.5% 0.1% 0.1% 2.7% TOTAL 62.8% 25.2% 12.0% 100.0% Note: Utilities include water, telephone and electricity; and Others include PE and allowances 7.28 The latest data on payment arrears owed by LAs is for end year 1999 (CAG report). The total stock of arrears based on the 1999 CAG report amount to Tshs. 5.9 billion. This is made up of 94% suppliers credit and 6% statutory remittance. In 1998 the stock of arrears stood at 6 billion Tshs. Three quarters of this stock was due to councils' creditors and a quarter of it due to non remission of statutory deductions from employees for contribution to Local Authority Provident Fund. Outstanding imprests and advances are owed to the LAs, and should therefore be netted out to obtain a net stock of contingent liabilities. Although unpresented checks do not constitute a fresh claim on future budgets, to the extent that they have to be financed out of inadequate flows of cash in the following year, they do represent a strain on cash management. It is noteworthy that the value of unpresented checks has more than tripled between 1998 and 1999. 7.29 The gravity of the problem varies across local governments. For some of the poorer authorities the FY01 PER mission visited, the stock of arrears make up over 50 percent of their own revenue collection, representing a major potential drag on financing their development programs. Most better-to- do and fewer poorer local authorities visited by the mission have begun, on their own, to reduce the stock of debt and often at the expense of much needed funding for public services in the poorer ones. 7.30 There is no operation for relieving the local governments of their debts. A major issue in such debt relief programs is the moral hazard problem and preventing the recurrence of build up of arrears. The mission discussed with those met appropriate disincentives to building up new flows and possible approaches to tightening the financial management system against recurrence of unsustainable stock. 46 7.31 It was recommended that the government should consider a relief program to clear up the stock of debt of poor Local Authorities (Highly Indebted Poor Local Authorities - HIPLA) to give them a fresh chance under the local government reform program. The PER Working Group commissioned a separate study for a more in-depth analysis of this problem and recommendations for way forward., 7.32 Following the central and local government debt relief operations, it was deemed necessary to prevent recurrence of the unsustainable stock of debt The FY01 PER mission recommended measures that would ensure that this is not happening. One of the measures is that new payments arrears should be integrated into the following year's respective spending unit's budgets as first charge from their allocations. This used to be the practice in the past aimed at enforcing discipline in budget management. It would, however, require timely compilation of information on arrears (in the EPICOR system where available) and verification. To prevent to abuse of extra-budgetary commitments, it was suggested that there was need to enforce the centralized system for issuing LPOs. Extended coverage of IFMS sub- treasuries and local governments will facilitate the control of extra-budgetary commitments. Furthermore, it was suggested that there was need to enhance enforcement of discipline by the accounting officers (AOs) to ensure that no invoices are held outside the financial reporting system. Strengthening the internal audit functions in ministries, regions and local governments would help minimize collusion with suppliers. It should also be widely publicized that the government would not honor any commitments made outside the IFMS. Private suppliers that violate this warning should bear the costs due to non-payment for goods and services supplied inappropriately. 7.1.3. The Integrated Financial Management System 7.33 The government has introduced the Integrated Financial Management System (IFMS), which when fully operational would permit improved transparency of public financial operations through real time information, and better controls through centralized payments and procurement processing. The EPICOR (formerly PLATINUM) is the software currently used to operate the IFMS. The Exchequer Ordinance has been revised inter alia to provide a legal basis for operating this new system effectively and accepting its reports for fiduciary accountability. To date all ministries in Dar-es Salaam are using the IFMS. However, two are not on line. In practice this means that all expenditure transactions are now being executed through the system. In the case of the two remaining institutions (President's Office and the Ministry of Defense and National Service and agencies under it), although not yet on line, the IFMS and EPICOR software have now been implemented. 7.34 Central government agencies located in regions and districts are served through 19 Sub- Treasuries. Manual operation of IFMS has been in place since last fiscal year. During the financial year 1999/2000 all payments were processed at the central payment office in the sub-regional treasuries. The structure of the chart of accounts in use is the same as that in use at the Central Payments Office but all payments and revenues are captured manually. Following failure of the first attempt to automate the system, on line links are being established in the 19 sub-treasuries from where payments and commitments are being executed for all central government agencies located in the 19 regions. An earlier attempt at automating IFMS in the sub-treasuries using the PLATINUM software suffered technical hitches in hardware and software services making connectivity inoperative. Furthermore, the system used dial up technology for linking with Dar es Salaam. Consequently the links were subject to telecommunication interruptions. The government is in the process of reinstalling equipment and linking the sub-treasuries with the center via satellite technology. Sub-treasuries connected to the IFMS center can now operate on line and the task to connect all the sub-treasuries is expected to be completed by end August 2001. Training of staff to operate the new system and modules is ongoing. Software services will subsequently be provided on line from the center. Provisions are also being made for stand alone servers to enable early operation of the EPICOR system in each sub-treasury while connectivity via satellite is being setup. This provision also caters for managing risk of failure of the central server. The 47 shift to satellite-based communication will avoid operating malfunctions associated with a dial up system and will enable technical support services on the software from the center. The revamped roll out of the EPICOR system will help resolve the problem of poor technical supportive services encountered in the earlier roll out initiative. 7.35 IFMS and the EPICOR software are also being rolled out to 28 local authorities on a pilot basis under phase I of the LGRP initiative. About half of the 28 LAs can operate the software effectively. 42 more have been targeted for phase II which begun in January, 200 1. 7.36 All ministries, Sub-Treasuries and local authorities visited by the FY01 PER mission expressed their appreciation for improvement in efficiency and controls when the EPICOR is operative and strongly urged for revival and expansion of its operation. However, some key transitional concerns that need to be addressed urgently were noted. 7.37 As of April 2000, a centralized LPO system became operational in Dar-es- Salaam obviating the need for continued use of printed LPOs. The presumption was that since many of the Sub-Treasuries had the EPICOR system, the need for printed LPOs would subside. Due to the fact that the initial rolling out of the Platinum/EPICOR system to the sub-treasuries was broadly not operative, commitments continued to be made using the old LPO system. Since these were largely supplied through central agencies, who had themselves migrated to the automated system, a shortage of LPOs emerged which endangered the integrity of the procurement system in areas where EPICOR is not operative. It was therefore recommended that there was an immediate need for ensuring that adequate supplies of printed LPOs was availed to all those in need of them prior to the effective migration to the EPICOR system in the sub- Treasuries. 7.38 It was also suggested that the government should also ensure availability of functioning equipment, effective supportive technical services, and adequate training to operating staff. More fundamentally, however, was the need to enhance the skill profile of the operating and accounting staff and incentives for retaining qualified accountants for whom the government faces stiff competition from the private sector and independent public agencies. 7.39 Currently, the OCAG conducts its audits around the computerized system. The new Finance Act enables them now to audit within the computerized system, which will facilitate more timely completion of audits and ongoing monitoring of integrity in the payments system. To do so, however, the OCAG will require major retraining of its staff to use the automated audit system. It was therefore suggested that there was an urgent need for drawing up and implementing an IFMS and EPICOR training program, as well as building up skills for automated audits. 7.1.4. Budget and Financial Management in Local Governments 7.40 The PRSP and a variety of other reports emphasize the need for improving strategic budget preparation and financial management at local government level, as Local Authorities increasingly assume the primary responsibility for the delivery of essential public services. Efforts are also under way to strengthen the capacity for financial management and accountability at this level of government. The FY01 PER mission, building on previous reports, visited 14 LAs to undertake an on the ground assessment of status of the budgeting process and financial management, and reviewed progress in strengthening them. More specifically the aims were: (i) to assess the effectiveness and capacity for bottom-up approach in expenditure planning; (ii) the frequency and quality of quarterly and annual financial reporting and feedback from supervisors; (iii) the efficacy of the supervisory functions of RAS, PORALG, OCAG and Ministry of Finance in budgeting and financial reporting/accountability systems of the local authorities. The following is a synopsis of the main conclusions and recommendations from the detailed assessment. 48 (a) Budget Processes 7.41 Budgetary stress: It was noted that there is an apparent difference in the extent of budgetary stress between rich and poorer local authorities (LAs). These differences are manifested in the relative size of payment arrears, own revenue as percentage of total revenue, own revenue per capita and the extent of under-funding of essential services. Deliberate effort was being pursued under the PRSP to identify and provide higher support to poorer districts. The mission suggested that there was need to mainstream this approach in the annual budgeting process and draw up concrete cross-sector programs that relatively favors the poorer areas. 7.42 Differential treatment of LAs in exchequer releases and funding: By and large under-funding is still a problem both in relation to needs and approved allocation. Some districts receive less resources than approved especially for the funding of other charges. There are marked variations in the extent of under-funding between regions and between districts with some districts receiving far less than their approved allocations. It is not clear what criteria are being used in deciding the allocation, timing of actual releases and disbursements since some LAs do not receive releases as expected. It was thus recommended that there was a need for a review to determine the extent of inequity of treatment and transparency in the criteria used for differential treatment. 7.43 Cash flow plans: There are indications from the LAs that cash flow planning would help ease cash flow problems and reduce uncertainties in budget implementation. Some LAs already prepare cash flow plans while some better to do LAs have been able to use cash reserves to bridge cash flow gaps and smoothen cash flow over time. Generally, councils indicated their agreement to including the requirement for preparing cash flow plans in the Guidelines for the budgeting process. As such it was suggested that councils should be encouraged to pursue this approach. 7.44 Arrears: Most LAs and in particular the poorer LAs have accumulated huge payment arrears. Usually in the preparation of the budget, LAs include repayment of arrears as one of the expenditure items. In practice, however, it is only the better to do LAs that have been able to spare enough resources for this item. The inclusion of arrears in the expenditure side of the LA budgets has instilled a sense of responsibility to the LAs making them wary to incur more debts. It was recommended that such practice (integrating debts in their budgets) should be formally adopted in the guidelines for budgeting at all levels of government. 7.45 Role of the Regional Secretariat (RAS Office): The RAS office's advisory role is currently very limited and ambiguous. This has led some to think that the Regional Secretariat is an unnecessary overhead to the government. However, given the distance between ministries and the LAs, there is definitely a need for an intermediate body, which will monitor and ensure that the national standards in service delivery and policies are adhered to. The Regional Secretariat is well placed to perform such a function. It was suggested that there was need for GOT to clarify the role of the Regional secretariat vis-a-vis the sector ministries in monitoring and evaluation, and in enforcing national standards. The national budget provision for these functions, it was also suggested, should be shared between the sector ministries and the Regional Secretariats. (b) Financial Management 7.46 Capacity for Financial Management: The four key positions - Treasurer, Expenditure Accountant, Revenue Accountant and District Planning Officer are filled by staff with the requisite professional qualifications. However, the Intemal Audit departments are weak in capacity, because most of the auditors are new (72 recruited in 2000) and lack job experience to man Internal Audit department, while others are not qualified for the job but only seconded from the Finance Department to do the job. 49 Positions below the accountants are also manned by accounts assistants, most of whom have not gone through a formal training on accounting. The finance management capacity is even more worrying and totally lacks at the ward level where, apparently most of the revenue collection in the Councils takes place. The mission recommended a need for further training of accounts clerks and orientation of audit staff to the financial accounting and management of LAs, as well as training of finance and audit staff in computer skills to cope with the introduction of the Epicor System. To help more effective supervision by the councilors, it was suggested that there was a need for occasional seminars for the councilors to sharpen their understanding of financial statements. The LGRP and other complementary initiatives should pay particular attention to this need. Internal Audit reports and their usefulness: Internal audit reports are essential in uncovering irregularities and ensuring that regulations are properly followed in the course of performing day to day financial management operations. The reports are important not only as a financial management tool for the management, but also in guiding and simplifying the work of external auditors. It was therefore suggested that these reports should be regularly produced and submitted both to the management and the Council. In particular, the Councilors should demand to read such reports on a quarterly basis and subsequently ensure that the management responds to queries contained in these reports. 7.47 Frequency and quality of quarterly and annual financial reporting and feed-back from supervisors: Monthly, quarterly, and annual financial reports are prepared by the LA management and submitted to the Finance Committee and the full council. Quarterly and annual reports are copied to the RAS and PORALG. However, none of the reports go to the sub-council level directly. The assumption is that the contents of the reports would reach the sub-council levels through their representatives. The mission recommended that both financial and audit reports (translated into Kiswahili) be displayed /made public at ward and village levels. 7.48 The practice existing up to the early 1980s of sending copies of the quarterly reports to Treasury was very useful for accountability / feedback. This practice, it was suggested, should be re-instituted and a desk officer should be designated in MOF to deal with such reports. 50 8. THE WAY FORWARD 8.1. Government Response to PER Recommendations 8.1 The Government of Tanzania made a substantial effort in addressing various issues that were identified during the PER main mission in November/December 2000. Although relatively little time had passed by June 2001 since the formulation of the recommendations in December 2000, it was heartening to learn that in a number of areas steps had already been taken to implement the mission's recommendations while in other areas the mission's findings had stimulated a debate among the key parties involved which were likely to lead to concrete steps to implement the recommendations of the PER mission. 8.2 Cash Budget: Although the cash budget system was instrumental in re-establishing fiscal discipline, there were rising concerns about the negative effects of the cash budget system on strategic resource allocation and operational efficiency. To soften the negative impacts of the cash budget system and pave the way for its abandonment in favor of a more orthodox budget management system the mission had proposed to incorporate monthly cash requirement plans in the budget process. This recommendation was taken up in the budget guidelines and ministries are now required to submit monthly cash requirements in addition to overall annual budgetary requirements. A second recommendation concerned the synchronization of monthly cash requirements with monthly cash availability through bridge financing by the Bank of Tanzania or alternatively through a cash reserve. There is now an understanding in principle that the Bank of Tanzania is prepared to provide bridge financing in the form of cash advances or through marketable instruments such as treasury bills. The preference of the Bank of Tanzania is to use marketable instruments since this would provide a safeguard against the possibility that bridge financing would turn into effective long-term Central Bank financing of government expenditures. government remains concerned about the cost of such financing and would prefer to use instruments whose time structure corresponds to the temporary nature of the financing needs. 8.3 Arrears: With the introduction of the cash budget system, a central payment system, and an integrated financial management system, government has implemented a system of hard budget constraints for spending units. Hard budget constraints increased budgetary stress for many spending units and arrears have emerged as an important problem undermining the budget process and fiscal discipline. The PER main mission identified several categories of expenditures, in particular utility payments and "automatic catering and up-keep needs" as well as expenditures in the purview of sub- treasuries where the LPO/IFMS was not operational as main areas of arrears. The mission recommended immediate action to close loopholes at the sub-treasury level by restoring the functionality of the LPO/IFMS system at that level and appropriate action has been taken by the government. With respect to utilities and automatic catering and up-keep needs, the mission recommended that these expenditures should be given the status of first claim expenditures similar to debt service and salary payments. This recommendation is still under discussion at Treasury. 8.4 Integrated Financial Management System: The PER main mission had noted problems with the implementation of the IFMS at the level of sub-treasuries and at local authorities. The problem at sub- treasuries was due to the incompatibility of computer hard- and software as well as problems with data transmission to the Treasury through the telephone network. This problem is being addressed with the Office of the Accountant General deploying new workstations (and new software - EPICOR) to the sub- treasuries to equip each sub-treasury with 4-5 new workstations. Connectivity with Treasury is being re- established through the use of a wide area network using virtual private network technology. It is 51 expected that the new system will be operational by end August 2001 with all sub-treasuries being online with the central servers at Treasury. 8.5 Problems with the roll-out of the computerized IFMS to local authorities are also being addressed by the government also through re-equipping them with a new software. The system was implemented to 28 local authorities, but was only operational in half of them. The main constraints at the level of local authorities are limited human resources and required expertise, which creates a need for sustained efforts in capacity building. 8.6 Local Authorities: The PER main mission had visited 14 local authorities to gain first hand knowledge of financial management issues at the local level and in particular the implications of the local government reform program which involves substantial fiscal decentralization. One of the key problems affecting the budgeting process of local authorities is the difference in financial years of the central government (July-June) and local authorities (January - December). The mission had follow-up discussion with the LGRP and Treasury on this issue. While there is broad agreement on the desirability of synchronizing fiscal years of central and local governments, no steps have as yet been taken to effect the necessary changes. The mission's recommendation of considering a debt relief scheme for local authorities has received positive response and a study has been done to establish the actual magnitude of the local government debt problem. 8.7 The Budget and MTEF focus on Poverty Reduction: The government more strongly used the MTEF as a strategic allocation instrument during 2001 by adapting its formulation to the PRSP process. Box I summarizes the pro-poor budget measures outlined in the 2001/02 budget speech by the Minister for Finance. Other actions taken so far and plans for the future are summarized in the subsequent matrix. 52 Box 1: Recent Measures on Poverty Reduction as Highlighted in the 2001/02 Budget The budget for FY02 is focused on poverty reduction through a combination of measures to raise overall growth and measures directly targeted to the poor. Apart from tax policy-related measures aimed at promoting private investment and reducing the cost of business operation, the budget contains measures that lower the cost and support improved access to basic services. The link with the PRSP is particularly sharp both on the tax policy side and expenditure allocation to priority sectors. However, the budget is too donor dependent (36.8 percent of total expenditure financed by foreign grants and loans) and calls for the need to further raise revenue effort. Substantial tax cuts have been proposed with revenue losses ensuing from them being covered largely through removing tax exemptions to Government/public institutions, whopping new taxes levied on the game of chance/gambling, and further efforts in curbing tax evasion. The main theme of the 2001/02 Tanzania Budget is poverty reduction. Broadly the budget targets a growth rate of 5.9% as against 4.9% of the previous year. The expenditure plan is supportive of the PRSP priorities, viz: education, health, water, rural roads, agriculture, justice, lands, and HIV/AIDS. The budget proposes to increase non-wage expenditure for priority sectors by 28 percent. The fiscal deficit before grants will be 3.5% of GDP as compared to 1.2% of GDP in the previous year (with no net domestic borrowing and hence largely financed by increased external funding). The increased expenditure allocations will primarily be targeted to priority sectors for poverty reduction. The government will continue to request donors to channel their assistance through the exchequer system and specifically to increase their budget support to consolidate gains made in the past two years. The Government has also decided to adopt/continue with the policy of quarterly rather than monthly cash releases for priority sectors. For non priority ones the monthly cash release system will continue but with commitments indicated for a full quarter. A wide range of tax measures for enhancing growth, which is targeted at 5.9% in 2001 and 6.2% in 2002, are outlined and hinge mostly on tax reforms to: promote private sector investment and growth, enhance competitiveness and transparency of the import regime, and reduce the burden of the poor. Tax measures directly targeted to the poor include: Health * Abolished all taxes for drugs used by all suffering from HIV/AIDS, tuberculosis and Malaria (communicable diseases) to make the drugs more affordable; and * Abolished VAT on hospital equipment. Education * Allocated Tshs. 5 billion to the newly introduced Education Fund for sponsoring post primary education of the children from extremely poor families; * Abolished VAT on capital goods for investment in education project, to promote the expansion and improved quality of the private education sector; and * Removed VAT on computers, printers and their accessories to promote use of IT in schools and other education institutions. Agriculture * Removed all stamp duty charges on agricultural and livestock produce; and * Will enforce the 5 percent cap on crop and livestock cess. Non Governmental Organizations * Resumed exemption from taxes for these to enable service delivery to the poor. The Government will tighten enforcement against abuse. Deal with Nuisance from Taxation and promote small/micro business * Abolished advance payment of income tax to promote the growth of small and micro businesses; * Will enforce the legal requirement that Minister Responsible for Local Government will consult with Minister for Finance before making any to legislation on levies and fees for Local Governments; * All licensing for small and micro-enterprises have been transferred to District Councils; and * Generally the issuance of licenses will now be substantially simplified. There will be a one stop center for issuance of such licenses at the village. district. reeion and ministerial levels. There will be less imoediments for 53 8.2. Impact of the PER Process 50. In addition to the review of fiscal performance and progress in addressing issues identified during the main PER mission, there is also a need to constantly review the effectiveness of the PER process, as all involved parties invest substantial amounts of staff time and financial resources in the process. Although it is difficult to establish a direct causal link between improvements in public finance management and the PER process, it is nonetheless evident that significant improvements have taken place in a number of areas which have been a central focus of the PER process. They include: * Improved prioritization and poverty focus of public expenditures; * Strengthening of budget management systems and institutions; * The strengthened MTEF produces more realistic budget projections and thus smaller deviations between budgeted and actual expenditures; * PER analytical work provided important inputs for the design of HIPC and the PRSP; * Improvements in donor reporting on aid flows under the PER process have considerably strengthened the information base for GOT budgeting; * The PER process and measures to improve budget management have improved donor confidence and a greater share of donor assistance is provided in the form of budget support or channeled through the budget thus limiting transaction cost and enhancing government ownership; and * The opening up of the budget process through the PER process has led to improved information flows between Treasury, sector ministries, donors, and other stakeholders. 8.3. Challenges for the Future 8.8 Overall the government has maintained good progress towards the use of the MTEF as a strategic allocation instrument and has this year turned its focus to adapting the MTEF formulation to the PRSP process. The key challenge is to further strengthen links with Sector Development programs, particularly where these were developed prior to the advent of the PRSP and further improving the costing of realistic requirements (including consistency with absorptive capacity) for achieving poverty reducing outcomes. 8.9 The core pillar for improving public financial accountability remains the implementation of the Integrated Financial Management System. Very good progress has been made in rolling this out to all central government agencies, which together accounts for nearly 80 percent of all resources mobilized through the Exchequer. All central government agencies, including those in the regions should be operating this facility on real time basis by end of August, 2001. Rolling out the IFMS to all LAs would also go a long way in improving financial management at the sub-national level. This is a challenge that has to be faced by GOT. 8.10 Results from the expenditure tracking studies and from the participatory CFAA underline the value of making service delivery surveys and expenditure tracking part of a routine annual assessment of the effectiveness of public spending. This is particularly important in view of the growing role of sub- national authorities and local communities in delivering/organizing essential services without adequate capacity and systems for financial management. 8.11 A long step forward has been an agreement and initial steps to address the vexing problem of tight cash budget and its impact on limiting predictability in expenditure planning. The government 54 began quarterly releases and commitments for all priority activities since December 2000 and has committed to stay with this practice. The government has also embarked on a system of cash flow planning and adopted measures for smoothing cash availability. The extension of this practice to non- priority sectors and preventing the build up of new arrears will reduce pressures on cash flow management. 8.12 The PER process and dialogue seems to be paying off in terms of improving both predictability and flexibility of budgets. A larger proportion of recorded loans and grants is now being provided as budget support and a larger proportion of the external resources are integrated through the Exchequer (including the development budget). There is still a long way to go but progress has been substantial and the dialogue is at least taking place on the same platform. This raises the stakes for improving the integrity of financial management and result-orientation to provide the necessary comfort to make further strides. 8.13 Finally, it would be pertinent that in the future the PER working group undertakes a full review of the strengths and weaknesses of the PER process with a view to learn from past experience and continuously improve the process. One widely shared view is that it would be useful to start the PER process with the drawing up of a plan and identification and commissioning of studies earlier than in the past to ensure that all inputs are adequately processed during the government's budget cycle. 55 Table 8.1: Summary of Impact of Public Expenditure Process on Public Expenditure Management in Tanzania Objective Recommended Action Actions Taken Remarks and Next Steps Institution Responsible A. Building a cost Use MTEF as a strategic MTEF is now mainly used as a Need to strengthen links MOF/Priority effective and better allocation instrument. strategic allocation instrument between MTEF/PRSP and Sectors. quality system of and its formulation in 2001 was SDPs; and the PER public service Increase allocation of adapted to focus on the PRSP process delivery. resources to priority process. Allocations to priority sectors. sectors have been increased Improve further the costing substantially. of realistic requirements for achieving poverty reducing outcomes. Undertake value for money Pro-poor expenditure tracking MOF/PER WG audits and review. study was undertaken under the Undertake value for money FY01 PER process. GOT has audits and review also begun the system of annually, and bring their publicizing (in newspapers) OC outcomes into public releases to all levels of domain through the PER govemment. process. Develop strategic plans Strategic annual plans and All GOT and performance budgets. performance budgets have been Agencies/Depts developed for MOAFS, Need to extend to other MOWLD, CSD, Health, MOF, ministries/Depts and to POPP, Education, and Water. adhere to strategic plans and performance budgets so as to enhance efficiency and quality in public Strengthen competencies MTEF training to service delivery. MOF/LGRP and institutional capacity planning/budget staff in for managing public ministries. Extend MTEF and other service programs. capacity building trainings to regions and LAs; and more explicitly involve LAs in the national budgeting process and strategic expenditure allocations decisions. B. Improving public Implement the Integrated IFMS using EPICOR software Roll out IFMS to all LAs MOF/LGRP finance management Financial Management has been rolled out to all so as to improve financial and accountability. System (IFMS) at all levels ministries and central management at sub- of government. govemment agencies. All national level, where the central government agencies in responsibility for delivery the regions will be operating this of essential services facility (through sub-treasuries) increasingly rests. on real time basis by end of Strengthen LA capacity to August, 2001. Only 28 LAs implement IFMS and run have so far been connected to EPICOR. IFMS. Adopt a centralized LPO Strengthened system of MOF system to ensure that all All payments and Bank centralized LPO is commitments are made in reconciliation are now done expected to go a long way line with approved centrally; and a centralized LPO in containing payment allocations. system is operative and all arrears. commitments are now made in line with approved allocations. Enact new procurement Enforcement of new All GOT law and regulations that New Procurement legislation procurement regulations Agencies/Depts permit greater transparency was passed by Parliament in and transparency in and integrity in February 2001. Regulations tendering are essential both procurement of public were prepared and the law in strengthening financial works and goods and became effective by end June management and reducing services. 2001. Results ofpublic tenders corruption. are also published quarterly in the local press. 56 Objective Recommended Action Actions Taken Remarks and Next Steps Institution Responsible Strengthen reporting There is need for LA system for accountability Under IFMS financial reforms reports to be copied to the purposes. are produced online. Various Treasury; and also made audit and financial reports accessible (translated in to (quarterly and annual) are Kiswahili and displayed) at produced at LA level for the sub-council level. management and the council. C. Addressing the Move to quarterly releases Quarterly releases and Extend this practice to MOF vexing problem of and commitments. For commitments for all priority non-priority sectors and tight cash budgeting this to be possible, embark activities was begun since curb the build up of arrears and its impact on on a system of cash flow December 2000. so as to facilitate reduction limiting foresight in planning and measures for of pressures on cash flow expenditure smoothing cash GOT has also embarked on a management. planning. availability. system of cash flow planning and adopted measures for smoothing cash availability. D. Improving both Request donors to provide Larger proportion ofrecorded More effort needed toward MOFI predictability assistance through the loans and grants are now being mobilizing resources Ministries and flexibility of budget (budget support) provided as budget support and through the budget. budgets. rather than through a larger proportion of the Dialogue and bilateral projects, to avoid external resources are integrated consultations; and duplication of effort and through the Exchequer improved financial for greater accountability (including the development management, purposes. budget). accountability and transparency in resource Request donors to provide use are essential for this to upfront indications of Donors are more willing to happen. support (budget/project) disclose their indicative within the PER and MTEF commitments. framework. Ministries/Departments no longer cut their deals with donors without involving MOF. 57 POSTSCRIPT 8.14 This report serves as a record of the findings of the Bank-led, external (to government) evaluation of fiscal developments. A large number of other reports that were produced during the FY01 report and which are listed in Table 2.1 of this report can be obtained from the PER Secretariat. The contents of this report have been discussed by the PER Working Group, with Government, and also been presented at the consultative PER meeting which took place in May 2001. In order to make the technical information contained in this document accessible to a wider audience, the PER Working Group has commissioned the production of a "popular" version of the PER, 8.15 The Tanzania PER being an annual exercise, by the time this report appears in print the PER FY02 is well underway. One of the main objectives of the FY02 PER external evaluation exercise will be to track progress on many of the issues discussed in this report. In addition, the FY02 PER will focus on the following issues: 8.16 General assessment of sources of fiscal risks: 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 will focus on identifying specific 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. In addition to identifying the sources of risks, where data is readily available the mission will undertake preliminary quantification of these, and assess the efficacy of the existing mechanism to contain risks. The PER FY02 review will also develop an agenda for further work on managing fiscal risk in Tanzania. 8.17 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. The FY02 PER will revisit 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. 8.18 Debt management: With the reaching of the completion point under the enhanced HIPC initiative, there is now increased urgency to ensure that the fiscal flexibility gained through debt relief is not eroded through weaknesses in the debt management system. The FY02 PER will review (i) the roles of MOF, BOT, and other actors such as sectoral ministries in debt contracting, debt servicing and consolidation of the information in the fiscal accounts, (ii) the adequacy of IFMS and of the provisions under the Public Finance and Procurement Acts in tracking and disclosing all investment spending financed through debt, (iii) and government systems in place that ensure that legal provisions in the Public Finance Act and the Public Procurement Act are adhered to. 8.19 Particular emphasis of this review will be on the integration of debt management with the budget process and the capacity of the Ministry of Finance to adequately monitor new debt commitments. In the past, there have been concerns that debt financed expenditures that are spread out over several years or 58 implemented by several ministries are not explicitly mapped to funding sources. The recently developed IMF/World Bank guidelines on debt management will provide the framework for carrying out the assessment of debt management. 8.20 Definition and monitoring of poverty reducing expenditures: Previous discussions of Tanzania's definition and monitoring of poverty reducing expenditures have revolved around two, to a certain extent conflicting, objectives. On the one hand, there is the desire to define and monitor a set of expenditure items that are clearly identified as "poverty reducing" expenditures in the budget. In particular with the expansion of budget support by donors, there is some pressure to clearly define and label such "poverty reducing" expenditures that eventually could be matched to donor budget finance. On the other hand, the PER and government have adopted a wider concept of poverty reducing expenditures. This approach considers poverty reduction in the broader context of the budget where government identifies pro-poor priority areas which receive priority allocations and whose expenditures are largely protected from expenditure cuts in the case of revenue shortfalls. The external review will revisit these issues taking into account information requirements of the providers of budget support. 8.21 A particular problem with respect to poverty reducing expenditure has been the consolidation of thematic priorities, such as funding of HIV/AIDS activities, which are implemented by various ministries and expenditures for which are spread across various ministries, departments and agencies. Additional constraints to the reporting on thematic poverty reducing expenditures are that donor support for these activities is not fully captured in the budget and government accounts as well as remaining GFS classification problems in the development budget. The external evaluation will assess to which extent these constraints limit the usefulness of budgetary reports on poverty reducing expenditures and review progress in overcoming these obstacles for the seven priority areas. 8.22 Internalization of expenditure tracking and service delivery surveys: During past PERs, the PER working group has commissioned expenditure tracking studies which analyzed the flow of resources from the central government to service delivery units. The government plans to implement an expenditure tracking system that would internalize this expenditure tracking function in order to provide the Ministry of Finance and sector ministries with timely information on the flow of resources. The external evaluation will review the status of the implementation of the government internal expenditure tracking system and make recommendations as to the adequacy of the system and the potential need for continued external expenditure tracking as a complement to government internal expenditure tracking. Particular attention will need to be paid to the tracking of transfers in kind, particularly in heath and education sectors. 8.23 The expenditure tracking system is to be complemented by service delivery surveys which would provide feedback on the impact and quality of public services. Government has committed to implementing service delivery surveys on a routine basis as part of the implementation of the performance budget approach. The external evaluation will assess progress with the implementation of the performance budget approach in general and pay particular attention to progress with the roll out of service delivery surveys. 59 ANNEX 1: PROCEEDINGS OF THE PER FY01 CONSULTATIVE MEETING, KARIMJEE HALL 7th & 8th MAY 2001 A. OPENING, PRESENTATIONS AND DISCUSSION DAY ONE Morning Session: Welcoming Remarks Mr. Peter Ngumbullu (Permanent Secretary, Ministry of Finance) welcomed all participants to the 2001 PER consultative meeting and thanked them for accepting the Government's invitation. In particular, the Permanent Secretary (PS) thanked the Hon. Daniel Yona (Minister of State, Vice President's Office) for accepting to come to officially open the consultative meeting. The PS noted that the meeting draws participants from a wide range of stakeholders including Parliament, central and local government, private sector (financial/banking sector; manufacturing, agriculture and trade) civil society organizations/NGOs, donors (bilateral and multilateral), academic and research institutions, faith institutions and the media. The PS observed that the theme of the 2001 PER consultative meeting is the adaptation of the PER process for poverty reduction and aims at consolidating the partnership approach in implementing the PER process began three years ago. Mr. Ngumbullu also pointed out that the 2001 PER consultative meeting is expected to feed into the finalization of the Poverty Status Report towards the HIPC completion point as well as the preparation of the 2002/02 budget and the Consultative Group (CG) meeting which will be held later in the year. The PS concluded his opening remarks by recognizing the co-chair of the consultative meeting Mr. James W. Adams, Country Director for Tanzania and Uganda - World Bank, as well as the presence of Mr. Jurgen Reitmaer - Assistant Director, African Department of the International Monetary Fund (IMF). Official Opening Hon. Daniel Yona - Minister of State, Vice President's Office - opened the meeting on behalf of the Hon. Basil Mramba, Minister for Finance who was on duty abroad. The Minister expressed gratitude to all the stakeholders and in particular the PER working group which has been overseeing the PER work. He also noted that the Government was encouraged by the strong participation and support by all the key stakeholders which has greatly enriched the Government budget process. He requested for more ideas from the participants on how the PER process can be improved further. The Minister highlighted important PER-related developments that have taken place since the last PER consultative meeting to include (i) The adoption of the PRSP in August 2000 as a national framework for implementing poverty measures and benchmark document for prioritizing expenditures in Tanzania (ii) formulation of strategies on rural development, agriculture and primary education (iii) initiation of the Poverty Reduction Budget Support (PRBS) as an instrument to ensure that adequate resources are allocated to priority sectors through the Government budget in support of the poverty reduction strategies. (iv) Further strengthening of budgeting and financial management systems through the Integrated Financial Management System (IFMS) and enactment of the Public Finance Act, 2001 and the 60 Public procurement Act, 2001. (v) Improving the cash budget system by introducing a quarterly commitment system to improve expenditure planning as well as release of OC to priority sectors on a quarterly basis since December 2000 and, (vi) carrying out a study on Country Financial Accountability Assessment (CFAA) to make an assessment of the strength of financial accountability in both public and private sectors in Tanzania. The Minister mentioned three issues that need further attention in the PER process: (i) the need to revise and update sector programmes to reflect the requirements of the PRSP to address the apparent disconnect and inconsistencies between the PRSP and sector strategies and programmes. (ii) problems in budget management and accountability due to the fact that a big part of donor support particularly project type continued to be administered by donors outside the Government budget and (iii) need to explore the possibilities of rolling out the PER process to the local government authorities. The Minister concluded by informing the participants that Tanzania is largely on track with regard to the macroeconomic objectives and targets of the Poverty Reduction and Growth Facility (PRGF). Vote of Thanks Mr. James Adams - Country Director for Tanzania and Uganda, World Bank - thanked Hon Minister Yona for the articulate opening address and also for his work in strengthening the PER process from the time that he was holding the portfolio of Minister for Finance to his current position as Minister of State in the VPO responsible for Poverty reduction. Mr. Adams applauded the Government for providing strong leadership in the PER process which is now transparent and highly participatory. The Country Director underlined the importance of the annual PER consultative meeting noting that (i) government expenditures have the largest impact on poverty reduction such that we need to keep watch to ensure that the quality of expenditure improves and, (ii) there are many outstanding budget issues and challenges including the need to address the concern raised by the Government on accounting and reporting of donor resources, as well as continued variance between budgets and actual budget out-turns etc. Mr. Adams subsequently challenged the participants to use the two days consultative meeting constructively and productively to among others, discuss how to further improve the PER process and also address the challenges and other budget issues generally. Issues Emergingfrom Discussions of Various Papers: Paper 1: Adaptation of the PER Process to Poverty Reduction * The Government was congratulated for good attendance in the PER Consultative Meeting. * There is a need to look into how priorities are set in some sectors. For example, transport as a priority should be taken in the broader sense to include other modes of transport (marine, air, railway etc) not just roads as is currently the case. Papers 2, 3 and 4: Fiscal Performance & Pro-poor Expenditure, Budget Sustainability and Systemic Fiscal Issues * Ministry of Finance was urged to ask donors to pass their funds through the exchequer system and to spend their monies according to Tanzania's priorities and not the way they see it fit. Donors were 61 asked to trust the system and pass their funds through the budget. However, the quality of control and monitoring of public funds needs to be strengthened and improved. Disciplinary measures should be taken against accounting officers who overspend Government funds. Loss of revenue should not be left to go on unchecked. * Ministries were asked to revisit their administrative costs as they are seen to consume large shares of ministerial expenditures. Large administrative expenditures within the ministries have been one of causes of the donors passing their funds off-exchequer system. For instance when one revisits the funds utilised at ministerial level for the past few fiscal years, it comes out clearly that billions are used for administrative purposes only. * The revenue effort is still very low in Tanzania. The Government has to find alternative ways on how to improve revenue flow. However, this can be done through connecting potential tax base to actual tax base, improvements of tax administration and getting rid of unnecessary tax exemptions. * The Government was urged to improve implementation capacity of the Cash Budget System to foster growth and be able to meet the PRSP and Vision 2025 targets. * Despite the fact that IFMS is operating at both ministries and Government departments, it has been noticed that the process of releasing funds still takes a long time. * Utilisation of domestic sources of funds - It was observed that credits from the banking system to private investors are still very insignificant despite the fact that Government does not borrow from the banking system. Moreover, it is estimated that about 20% of the bank reserves are invested outside the economy. The Government should therefore continue to set-up a conducive environment that will encourage the utilisation of domestic sources of financing such that economic agents can access more credits instead of leaving idle funds accumulating within the banking system. * Following decentralisation policy where service delivery units have been placed under the direct responsibility of the LAs, there is a need for Government to improve capacity within LAs. * It was noted that there is a great shortfall between budgeted expenditures and actual expenditures (specifically development expenditure) The MoF was urged to disburse public resources as approved by the Parliament. * The Government was asked to set a conducive environment for petty traders to obtain credit from the banks, and also revisit the present tax rates. For instance, the tax rate has to consider the issue of location implying that the highly taxable petty traders should be those located in the city centre where business is relatively brisk. Analogously, those located in the periphery of the city should bear much lower tax rates. * The current inflation rate is worth preserving. Pushing inflation further down from the current 5% may be harmful to the economy. High and very low rates of inflation are not good for growth. * There is need to consider issues of risk when addressing the issue of the optimal level of borrowing, otherwise HIPC objectives (to reduce debt ratios) may be eroded. 62 Afternoon Session: Papers 5 and 6: Financial Accountability * Participants expressed general satisfaction regarding the Government's efforts to further strengthen budgeting and financial management systems by implementing the Integrated Financial management System (IFMS) to improve expenditure planning, execution and control and adoption of the Medium term Expenditure Framework (MTEF). * However, participants underlined the need for the Government to fully implement the IFMS including rolling over the IFMS and MTEF to local governments to further improve financial management. The rate at which the Government puts the system into use will determine how donors channel their resources through the Exchequer. * Participants urged the Government to do more to address the problems in public financial management pointed out by Controller and Auditor General (CAG) the year-in year out. * Participants also underscored the importance of capacity building to enable the office of the Controller and Auditor General (CAG) to perform its audit functions around the computer in line with the IFMS. * The role of civil society in monitoring Government performance was emphasized * Donors were asked to route their support through the exchequer system since it allows for tracking down of expenditures down to service delivery units, something that donors may not be able to do on their own. Papers 7 to 11: MTEF Cross-Cutting Issues * It was suggested that PRSP Programmes be targeted at community level and the programmes should be owned by communities themselves. Communities should be made aware of what the PRSP programmes are and how to go about dealing with the whole issue of poverty reduction. * The Government was urged to put more effort toward fighting HIV/ AIDS in the country and check what is really missing in the entire drive/war against HIV/ AIDS. The Government should put in place implementable programmes and also mobilise more resources for fighting the pandemic. * LGRP was urged to prepare a plan for providing training programmes that will assist the informal sector operators to familiarise themselves with laws and regulations, and cope with the ongoing reforms. LGRP should also design strategic training programmes for councillors to make them aware of their responsibilities given ongoing economic reforms. Furthermore, LG institutions should be given a mandate to hire and fire the executive officers who misuse the public resources/government funds. * There is need to put more focus on infrastructure investment, particularly on roads, so as to avert losses. Loss due to bad roads is estimated at Tshs. 600 billion or 9% of GDP per year. 63 * A lot of work still needs to be done on how the LGRP is going to incorporate local authorities in a harmonized budget process. + There is a strong urgency to address the issue of corruption in policy making and investment. 64 DAY TWO Morning Session: Paper 12: Cross-Sector MTEF The Government and Donors work together to improve the predictability of external finance. As far as the poverty reduction strategy is trying to get rid of poverty it was suggested that everybody and each household in the country must share this vision and commitment to eradicate poverty. To that effect, PRSP Programmes need to be owned by everyone. Poverty eradication must be seen to be the central focus to which people at all levels are aware of. Public education will have to play a vital role to make the general population aware of what is happening at the national level and on their roles/ responsibilities to improve their daily lives. The Government was advised to ask for debt cancellation and not just to bank on slow-processed debt relief. This is because debt cancellation is a permanent solution compared to debt relief which is a temporary solution. Debt relief to heavily indebted LAs (HIPLA) was highly supported. The Government was asked explore the possibility of reserving privatisation funds to pay for retirement benefits for civil servants. It was pointed out that the issue of governance has been marginalized in the Cross Sector MTEF. It was emphasized that governance is a cross cutting issue that needs to be highlighted and evaluated in all sectors. Issues related to corruption and good management of resources need to be seen as an integral part of governance. The Government was asked to scale-up the drive to improve governance in all public institutions and across sectors. Participants asked the Government to clarify on apparent over funding noted in the case of Trunk and Regional Roads in the cross-sector MTEF numbers considering that under funding was said to be a problem in the Roads sector PER study. It was commented that the future of Tanzania lies on education. There is a need to make sure that Tanzania achieves quality education for the betterment of its economy. It was emphasized that although the PRSP has put much emphasis on primary education only, it is also the case that primary education alone cannot cope with the ongoing technical changes and development issues. In this case, the Government emphasis should also be put on higher education. Government needs to include the growing problem of unemployment, particularly for the youths, among its priority areas to act upon. A concern was raised that there was no participant in PER Consultative meeting from the Government of Zanzibar. For the future, an attempt should be made to have representation from Zanzibar so as to facilitate dissemination of deliberations of the meeting to the people of Zanzibar. 65 Afternoon Session: Papers 13 and 14: Financing (Indications of Domestic Financing and Donor Support FY02-FY04, Dealing with Financing Gaps - Domestic & External) There is a need to find out the actual resource gaps, broken down into recurrent and development, since development resources which are largely from foreign sources are not fungible and the aggregate may blur gaps in recurrent budget. It was suggested that automatic finances from the Central Bank to the Government should be discouraged. But auctioning of Government securities by the Central Bank is allowed and encouraged. It was pointed out that beginning year 2002 about 1.5 million students have to be enrolled in primary education, but the resources available are inadequate for purposes of reaching this target. This suggests that borrowing could be one option to achieve the goal. It was underlined that increasing growth and reducing poverty requires macro stability. The Tanzania Government was therefore cautioned not to ignore macro stability as the country focuses on how to finance and achieve PRSP targets. DONOR SUBMISSIONS/REMARKS Sweden a) The Ambassador of Sweden to Tanzania expressed his pleasure for being one of the participants in the FY01 PER Consultative meeting, noting that he had enjoyed the discussions. He commended the PER process and its participatory nature, and the progress that Tanzania has made especially in attaining macroeconomic stability. His other remarks were as follows: b) There is a need to review the PER process itself to allow for (i) more participation of other national stakeholders, and (ii) the Government to run the process itself. c) The concerns that a significant amount of donor funds are not included/channeled through the exchequer are genuine. Sweden will mobilize efforts toward addressing this issue. d) Further development hinges on capacity improvement in the Tanzanian institutions and existence of absorptive capacity. e) The existing financial management system lacks transparency and is vulnerable to leakage. There is a need to involve stakeholders by transparently stating to them the amounts of resources that are released to the sub-national levels, in order to strengthen accountability at those levels. f) Supports the idea of strengthening donor resource projections by (donors) providing information ahead of budget. g) The importance of HIV/AIDS fight is now appreciated - as evidenced by the formation of TACAIDS. However, much more still needs to be done toward this. 66 h) The bottom-line for continued flow (or increase) of resources is the improvement of general policy of the Government, not only financial management but also resolving issues of conflict in Zanzibar and involvement of Zanzibar in the PER (or creation of its own mini-PER) process. European Union (EU) a) The Representative of EU commended the PER consultative meeting, noting that it was interesting both in terms of presentations and comments raised. Other remarks were as follows: b) EU was negotiating a new program of support to Tanzania based on the PRSP framework. c) Urged that there was a need to convert macro stability and growth into poverty reduction. EU is keen in providing assistance to the key social sectors for both poverty reduction and macro stability. d) EU is ready to pass more resources through the exchequer. However, this will depend on GOT putting in place good control institutions. e) Emphasized the importance of putting the correct numbers in the MTEF, especially for the Road sector. As presented, Road sector MTEF numbers were inconsistent PER study numbers. f) Stressed the importance of building new roads so as to reduce the transportation cost and time. g) Stressed the importance of indicators of results and transparency in resource allocation and disbursement to ensure that resources reach the targeted group at lower level. h) Capacity at the local level needs to be addressed/enhanced if funds are to be efficiently channelled there. i) Tanzania can get a bigger share of the resources based on performance. In this regard the following need to be done: (i) corruption has to be controlled, and (ii) clear sector policies have to be put in place so as to guide decisions on funding, especially through program assistance. j) The focus of EU will be in supporting local government and civil service reforms. k) Commitment and Disbursement Forecast by the European Union (2001 -2004) are summarized in the table below: 67 European Commission: Grants - Commitment and Disbursement Forecast (million EUROs), 2001 - 2004. . Project Aid Budget Support Total 2001 Commitments 61 75 136 Disbursements 40 38 78 2002 Commitments 165 165 Disbursements 80 37 117 2003 Commitments 60 110 170 Disbursements 119 37 156 2004 Commitments 20 20 Disbursements 96 37 133 United Kingdom (UK) a) The UK High Commissioner to Tanzania praised the PER consultative meeting for its participatory nature. His other remarks were as follows: b) UK supports PRSP as a prioritizing tool and MTEF as a vehicle for ensuring the resource allocation. c) UK will continue to support PRBS (above current support of £ 130 million) if results could be made explicit. d) It is important to ensure a strategic link between PRSP, MTEF and SDPs. e) More achievements beyond fiscal performance are essential, especially at sector level. f) Improved management of resources and service delivery is crucial. In this case taking stock of what has been delivered/achieved (generally and in the context of PRSP targets) is essential. g) There is a need to ensure great transparency on budgeted resources and how they are intended to be used and how they are ultimately used at local level. h) UK would like to see growth that benefits the poor. As such there is a need for sustained growth in agriculture, which benefits the majority who depend on it. Denmark a) Denmark representative praised the FY01 PER process, noting that it has resulted in very good findings that will guide future resource allocations. Other remarks were as follows: b) Denmark's support will move toward budget support. However, this will depend on improvement in GOT financial management, including predictability and transparency in government spending. 68 c) Denmark will support local government reform. The support will also be directed on (i) Agriculture (ii) Roads (iii) Health (iv) Private sector. African Development Bank (AfDB) a) The representative of AfDB commended GOT for attaining macro stability in the country. Noted, however, that there was a need to review various instruments to ensure that a judicious balance is maintained in the economy. Other remarks include: b) There is a need to move PRSP and budget processes closer to the people. c) There is a need to analyze various indicators (e.g. on education) at disaggregate level so as to unearth the diversities between regions and districts. In this case there is a need to consider and give priority the most disadvantaged areas of the nation in terms of provision of social services. d) AfDB have taken note of priorities and are ready to support the (Education) sector. AfDB will also support Roads and Health sectors; and balance of payments (BOP). UNICEF/UN System a) UNICEF/UN System representative commended the PER process in Tanzania and Tanzania's tremendous effort and achievements in macro stabilization in the past five years. b) Focus of UNICEF/UN System support/participation will be on: (i) Maintaining macro stability (ii) Poverty monitoring (iii) HIV/AIDS (iv) Basic Education (v) Primary Health (vi) Food security and livelihood (vii) Local Government Reform - to help reforms come down to the people. 69 B. CONSENSUS AND SUMMARY OF MAJOR EMERGING ISSUES The theme of the PER FY0 I Consultative Meeting focused on realigning the budget and the MTEF with the PRSP which is now the overarching national framework for implementing poverty reduction measures in Tanzania and which has become the benchmark document for prioritizing expenditures in Tanzania. A two-volume PER report which will contain updated versions of the papers presented at the meeting as well as a summary of the discussions will be prepared after the Government budget for FY02 has been finalized. A key feature of the 2001 PER consultative meeting was the broader participation both in terms of the substantial increase in the number (around 250) and composition of participants, compared to the last two PERs. Participants were drawn from the Government (Ministries, Independent Departments, Local Authorities), the Parliament (Public Accounts Committee, Local Authorities Accounts Committee, Parastatal Organizations Committee, Finance and Economic Committee, Social Services Committee), the private sector (financial/banking sector, manufacturing, agriculture and trade), donors (bilateral, multilateral, UN System), civil society (NGOs, Association of Local Authorities, gender organization), faith institutions, academic and research institutions, and the media. However, during the course of the consultations, a need was expressed to extend the PER process and consultations to involve Zanzibar as a mechanism both to support and facilitate development and political dialogue. In previous years, the consultative PER meeting was usually followed by a Consultative Group meeting. Since the 2001 CG meeting has been moved to the second half of the calendar year, the consultative PER meeting took up some of the issues typically reserved for the CG meeting. In particular, in the last session of the meeting several donor representatives expressed their intention to maintain and in several cases also increase substantially official development assistance to Tanzania (see donor submissions above) as long as progress is made in the key areas of: * Budget management; * Prioritization and focus on poverty reduction in line with the Poverty Reduction Strategy Paper; and * Governance, including issues related to Zanzibar. Donors also indicated that progress in these areas would also be critical to sustain the shift from project to program support. In the closing of the meeting, the World Bank Country Director James Adams presented a summary of the main emerging issues from the meeting. Major Emerging Issues 1. PER Process (i) Mileage Gained under the PER Process: There is a consensus that the PER process has enriched the Government budgetary process. Remarkable progress has been made on progressively increasing funding to priority areas. In addition, the PER has become an increasingly GoT owned and led process as well as highly participatory and transparent. All this needs to be maintained and strengthened. 70 (ii) Adaptation of the PER Process to Poverty Reduction The PER process has in addition to providing inputs into the budget process and development of the cross-sector and sector MTEFs adapted to focus more on poverty reduction. The subsequent PER therefore will need to pay greater attention to the poverty focus. (iii) Extending the PERIMTEF Process to the Sub-national Level: Flow of funds to LAs and service delivery units is of concern and GoT needs to put appropriate monitoring mechanisms in place with adequate follow-up. This will require, among others, (i) rolling-out the PER/MTEF process to the lower level (Local Government Authorities) as these levels do deliver the primary services. This will improve planning, execution and control of budget at that level. In addition this will strengthen local ownership and local participation. (ii) spending more resources on building capacity at all levels of government but more so at the sub-national levels. 2. Maintain Macroeconomic Stability There is a broad consensus that the current success of the GOT in maintaining macroeconomic stability should be maintained. 3. Further Prioritisation to Foster Strategic Expenditure Allocation There is need to continue work on prioritisation to foster strategic expenditure allocation using the MTEF instrument, including revisiting the priorities, to focus on sectors, such as the transport sector (broadly defined), which can make a real impact in terms of poverty reduction. It is important that ongoing interventions targeted at realising sustainable development in the rural sector and delivery of high quality primary education are revised and updated to reflect the requirements of Rural Development, Agriculture and Primary Education for poverty reduction. 4. Concern on the Low and Drifting Down of the Revenue Effort Low and declining revenue effort remains an important concern. The revenue effort is seen to be driven by taxes on international trade while performance of other taxes does not show much improvement. Efforts are thus needed in improving compliance and strengthening tax administration. 5. Concern on Continued Channelling of Donor Resources Outside GOT Budget More attention needs to be put on improving the reporting and accounting of donor resources. Administration of donor support outside the Government budget creates real problems in budget management and accountability considering that over 30% of the budget is donor funded (over 80% for development budget). More donor resources need to go through the budget so as-tosupport the poverty reduction strategy. Addressing this problem requires promoting/enhancing trust and collaboration between the Government and Donors. It also requires the GOT to continue demonstrating resolve to improve budget management, transparency and accountability. 6. Need to Harmonize SDPs and PRSP There is an apparent disconnect between Sector Development Programs (SDP) and the PRSP. There is therefore a need to harmonise the SWAPs/SDPs with the PRSP for orderly implementation of the poverty reduction strategy. This implies that Sector development programmes and strategies should be reviewed to ensure consistency with PRSP to ensure a strategic link between resource allocation and poverty reduction. 71 7. Options on Full Financing of the PRSP Full financing of PRSP targets/requirements has implications on resource mobilization vis-A-vis attainment of the targets. Various financing options will need to be considered if the estimated resource gap of 3% of GDP is to be mobilised. The options include: reducing the PRSP targets or lengthen the period over which targets are to be achieved, or raise more domestic resources or through borrowing (external/internal) but without prejudicing macroeconomic stability. 8. Dealing With the Problems Associated with the Cash Budget System The cash budget system has allowed Tanzania to achieve macro stability and increase discipline in budget management. However, the system denies flexibility in terms of informed planning. The cash budget creates a considerable level of stress to the budget, taking no account of lumpiness of expenditure or the ability to implement programmes in an environment of uncertain/irregular/delayed funding. Government should explore possible feasible options in dealing with the problem of cash flows. The challenge facing Tanzania is how to move from the cash budget system to a flexible budget system hinged on credible government institutions. There is a need to address issues of budget management stress from the cash-budget system by: (i) revisiting prioritization by type of categories of spending and bringing to fist charge items like utilities and automatic catering and upkeep; (ii) cash flow planning, and (iii) creating a disincentive for accumulation of arrears. 9. Need to Address Weaknesses in the Financial Sector for Poverty Reduction The financial system in Tanzania is not sufficiently deep, even by the stand order of developing countries. So that the impact of financial intermediation on the growth of the economy and poverty reduction is still low. 10. Strengthening Public Financial Management While the establishment of the IFMS has greatly enhanced the Treasury's capacity in managing Government finances, lack of ownership of the system threatens its sustainability in the long term and makes the system susceptible to potential breakdowns and/or sabotage. More training is needed together with the roll-out of IFMS and enhancing staff motivation of staff in Government generally. 11. Monitoring of Expenditure Impact As progress is being made in sharpening poverty-focussed priorities and expenditure allocations, increased emphasis will need to be paid to the monitoring of the impact of public expenditures. This should include further tracking studies and service delivery surveys as well as participatory monitoring and evaluation. 12. Enhance Transparency and Information on Expenditure Releases to Recipients and Beneficiaries A key measure to improve accountability for the use of resources is to provide the intended beneficiaries with information about funds made available from the Central Government for specific purposes and beneficiaries. 72 13. Greater Attention to Local Authorities It should be ensured that the public expenditure management process takes reforms under the Local Government Reform Program adequately into account. Providing some form of debt relief to local authorities who are burdened with high levels of arrears and debts should be seriously considered. 14. Need to pay Greater Attention to the Close Link Between Poverty and the Environment There is a realization that poverty and the environment are intricately linked, so that greater attention to this close link deserves greater attention in the poverty reduction drive. 15. Decisive and Immediate Action Needed Against HIV/AIDS Decisive and immediate action against HIV/AIDS is necessary and should be fully funded by GOT and donors. 16. Continue Mainstreaming of Gender in the Government Budget 17. Implementation of the Pay Reform Requires More Impetus Progress is being made regarding the Public Service Reform Programme. Donors were asked to discontinue LCC and support the SASE. 73 ANNEX 2: ASPECTS OF BUDGET SUSTAINABILITY11 In some respects, budget sustainability is a wider concept than fiscal sustainability, and is subject to a variety of interpretations. Here, it is helpful to distinguish between three of them. * The first, which is the most familiar, concerns itself with whether a planned expenditure path can be sustained without implying infeasible or unacceptable financing consequences - fiscal sustainability. In effect it asks whether a spending programme is financeable, without enquiring whether it is well designed or particularly desirable. * The second concerns itself with whether a planned expenditure path can be implemented, given available finance; it raises the issue of absorptive capacity as a potential independent constraint on spending - absorptive sustainability. * The third concerns itself with the congruence between planned expenditures and the government's underlying objectives. In the absence of absorptive difficulties, would the budget suffice to deliver the intended outcomes? This third interpretation is more concerned with the sustainability of the objectives, given the level (and composition) of the budget, rather than the sustainability of the budget per se - target sustainability'2. Full sustainability requires that all three component types hold simultaneously. However, a finding that the government's programme was unsustainable in any one of these three senses would have very different implications for policy. If it is sustainable in the fiscal and absorptive senses but not in the target sense, then the first issue is whether the problem can be resolved by shifting allocations between sectors. If not, can it be resolved by raising expenditure without violating either fiscal or absorptive sustainability? If not, it will be necessary to adjust aspirations/targets downward until they are sustainable within existing constraints If it is sustainable in the target and absorptive senses but not in the fiscal sense, then the targets are again infeasible, and downward revision is required until all three constraints are satisfied. In this case, it is likely that the downward revision should be reasonably uniformly spread across sectors. This assumes that the original targets reflected a relatively uniform degree of prioritisation, in the sense of the perceived marginal benefit per shilling expended across sectors. This annex was originally section 2 of a report prepared by David Bevan "The Fiscal Deficit and Sustainability of Fiscal Policy" for the FYO1 PER. 12 Notice that this definition of target sustainability is essentially hypothetical. It asks: "if neither funding nor implementation were a problem, would this level of budget suffice to deliver the planned targets?" It is the level of "full financing" associated with a given set of targets. 3In practice, policy typically does not imply fixed (undated) target levels, but some path over time to a desired level. "Revising a target downward" is therefore a shorthand for a more complex process which revises the (early) path downward, possibly with a subsequent acceleration to make good the lost ground within an unchanged horizon, or possibly with an extension of the horizon. 74 If it is sustainable in the target and fiscal senses but not in the absorptive sense, then the implications are rather different. Difficulties of implementation and absorption are likely to be sector specific; downward revision of some sectoral spending levels within a financeable total opens up the possibility of raising targets in other sectors which are constrained by their budget but not by absorptive capacity. Absorptive difficulties also require specific policy responses to relax them, and these may involve a different set of expenditures to those already budgeted. So far, the discussion has presumed that sustainability is problematic in one dimension or another, so that changes have to be made until all constraints are satisfied, the most stringent of them just so. But what if the budgeted programme appears to be sustainable in the target sense and strictly within sustainable limits for the other two? In other words, it would in principle be possible to finance and implement a more ambitious set of targets than is currently programmed. In an industrialised country with a history of fiscal prudence, a mature tax administration, and a deep financial system, this would be expected to be the norm. The share of government spending in the economy would not be constrained at a boundary, but (ideally) would be an 'internal' choice reflecting a balance between the marginal cost of public funds (including the distortionary 14 15 impact of the tax system) and the marginal benefit of public spending The case of a country like Tanzania is less clear-cut. While, unlike many other countries in Africa, it does not have a high level of domestic debt relative to GDP, it does - even by regional standards - have a very undeveloped financial system and a low rate of tax collection. It has in consequence proved very difficult to mobilise additional domestic resources. One interpretation of this might be to say that the marginal cost of public funds is very high - the distortionary impact of higher tax collections (or higher domestic borrowing) would be very damaging to private sector development, and the gain from lower collections correspondingly high. In consequence, even though the marginal benefit of public programs may be very high, it is still appropriate for them to be very circumscribed'6. If the currently programmed targets can be financed within the available budget, it might be better to lower taxes (or reduce borrowing) rather than to raise targets. Another interpretation might be to say that the difficulty in raising revenues is primarily an administrative matter - the marginal cost of these funds is not spectacularly high, but there is - in the short to medium term - a constraint against increasing them. In that case, it would be appropriate to spend all available resources. Targets should simply be raised until the available 14 To the extent that there are different sources of funds, access to these can be varied so that they each have the same marginal cost. 5 Of course, views may differ radically as to where this balance should be struck, and systematic mistakes may be made in where it actually is struck. 16 This is in respect of domestically financed public spending. As regards external grants or concessional loans, the marginal cost of funds will be very low so it is appropriate to make as full a use of these as possible. 75 budget is exhausted. In other words, matters should be arranged so that the fiscal constraint always binds'7. In the past, these two beliefs have often both been held simultaneously, one in respect of domestic taxes, the other in respect of domestic borrowing. Thus in Tanzania, in recent years, it has been felt appropriate to maintain the pressure to raise tax revenues as a share of GDP, while lowering domestic debt as a share of GDP (and indeed in real terms also). This is a perfectly coherent position to adopt, but it is important to see that it does imply a very specific judgement, namely that the marginal (social) cost of domestic taxes is substantially lower than the marginal (social) cost of domestically borrowed funds. These issues are pursued in more detail in the following section. 17 This assumes that, while absorption constraints may in the short to medium term limit what can be spent in particular sectors, this is unlikely to be the case across the budget as a whole. Hence, given reallocations, the fiscal constraint will typically bind in aggregate before the absorption constraint. 76 ANNEX 3: FISCAL SUSTAINABILITY"8 The notion of fiscal sustainability offered in annex 2 was the concern whether a planned expenditure path could be sustained without implying infeasible or unacceptable financing consequences. While this may seem a natural starting point, it is worth setting it against a more comprehensive, if somewhat stylised, approach to budget design. This section begins by sketching this idea, and then returns to why we may be interested in the more limited concept of fiscal sustainability. 1. Stylised budget design In general, the various sources of government funds are not subject to rigid upper limits; in each case, it is possible to mobilise increased volume at an increased cost. The government, like other economic agents, faces an upward sloping marginal cost of funds. Similarly, we may suppose 19 that the marginal benefit from government spending falls as the volume of spending rises In principle, optimisation of the budget would then involve three steps. First, recourse to different sources of funds would be governed by their respective cost; low cost sources would be expanded and high cost ones contracted until the marginal cost of each was equalised. Second, the composition of spending would reflect the differential benefits of different sectors; high benefit sectors would be expanded and low benefit sectors contracted until the marginal benefit was equalised. Third, if the common level of benefits exceeded the common level of costs, the budget would be increased and vice versa, until marginal benefit was equated to marginal cost. In this process of optimisation, sustainability is not really an issue. An unsustainable course of action would be excluded from consideration because it would have unacceptably high costs (relative to the realisable benefits); the optimum choice would be made from within the set of sustainable programmes. It would be the best of this set, and would in general lie inside it. In other words, it would have been possible to programme a bigger budget without this being unsustainable, but it was preferred not to do so. This idealised scenario makes a number of strong assumptions: * First, it assumes that it is politically and technically possible to vary all the relevant magnitudes at will. In practice, it may be a slow process expanding some sectors and contracting others, for example; then there will be persistent discrepancies between sectoral 8 This annex was originally section 3 of a report prepared for the FYO I PER by David Bevan "The Fiscal Deficit and Sustainability of Fiscal Policy". 19 This is of course an idealised account. In practice, powerful but undeserving interests may pre-empt part of the core budget, so that higher priorities are further down the queue. In that case, the marginal benefit of public spending might actually rise with an increase in volume. Similarly, there may be a great deal of waste, which could in principle be eliminated with no loss of benefit at all. But the whole thrust of the PRSP, PER, MTEF etc. is to reduce the likelihood of these perverse cases. 77 marginal benefits. Similar considerations apply on the resources side. Thus there may be extended frictional effects; however, this does not undercut the principle involved. * Second, it assumes that the relevant costs and benefits are susceptible to ready identification and calculation. In particular, it assumes that this can be done in respect of the same money metric on each side of the budget. In practice, a government may be able to prioritise between different activities, and hence decide on budget composition, without being able to place a cash value on the (relatively uniform) benefits. * Third, it implicitly assumes a rather static type of problem. It fails to do justice to the complicated dynamics that are likely to be involved. For example, low domestic debt may make the marginal cost of domestic borrowing low, but if this results in increased borrowing, 20 the marginal cost may rise over time * Fourth, it implicitly assumes that the supply of funds schedule is steadily upward sloping, rather than being rationed, or otherwise suddenly becoming very steep at some point. If the latter is a more accurate description, and the marginal benefit schedule meets the cost of funds schedule along this steep section, it is no longer appropriate to think of adjusting the volume to equate costs and benefits; it is more that the available funds are rationed across competing uses, and marginal benefit is determined by this rationing process rather than the actual cost21. The second, third, and fourth of these points all have the effect of making sustainability an issue. The second point implies that it may be possible to make progress on getting an efficient financing mix and an efficient spending pattern without it being possible confidently to align the two, i.e. to decide on how big the budget should be. Given this indeterminacy, a natural starting point is to form a view on the location of the practical upper bound, even though it may not be appropriate to utilise all of it. Discussion could then focus on whether it would indeed be desirable to set spending at this limit or at some lower value. Clearly, this would require an exercise of judgement, in the absence of calculation, as to the respective costs and benefits at this limit. The third point emphasises that budget design must be carried out as an inter-temporal exercise, and that - in outline at least - this must be done over a long horizon. Changes to current budget flows must either envisage offsetting changes in the future, or have cumulative stock consequences, or usually both. For example, consider a shift in the conventional domestic budget from balance to a deficit, with revenue stationary, so that expenditure can be raised 22. If this deficit is continuing, it will lead to a continuously rising domestic debt and debt service. If instead the budget returns to balance, domestic debt will stabilise at the new higher level, but debt service will have risen so that the primary surplus associated with conventional balance will 20 Related issues are discussed more extensively in Annex 2. 21 In the jargon of economics, the "shadow price" of funds is equal to their opportunity cost, which is their value in the current marginal use, i.e. the marginal cost is determined endogenously to equal the marginal benefit. Thus if a grant of $100 is intended for a use whose benefit is valued at $200, the cost of these funds, if diverted to another use, is not $100, let alone zero, but the $200 benefit that would be foregone. 22 For simplicity, this illustration considers a stationary economy. In a growing economy, the same issues arise but in a more complex form. See section 3.2. 78 have risen, and expenditure on goods and services will have to fall. If these expenditures are to be maintained, taxes will have to rise. Alternatively, the conventional budget could be shifted temporarily into surplus until the increased debt had been retired; spending could then revert to its original level, but only after a period of more severe contraction. The point of this illustration is that we cannot shape an informed view on the true cost of a current increase in borrowing without knowing something about how the consequences will be handled. In this respect, it is quite unlike, say, the purchase of a good where the cost can be considered in isolation23. This poses something of a problem. In principle, the earlier paradigm of optimal budget design, given these inter-temporal complications, would require the whole sequence of budgets to be solved forward into the indefinite future. Only when we knew what happening in the future could we assure ourselves that what was planned for the present was optimal. Since even economists acknowledge that this is impractical, we have to find a way of finessing this problem. What we cannot do is to ignore it. This would imply behaving completely myopically and ignoring all future consequences of our actions. What is required is some procedure that acknowledges the knock-on effect of present actions, but does not require implausible degrees of sophistication or clairvoyance. This approach brings issues of sustainability to the fore. 2. Fiscal sustainability24 One way to implement this requirement for a degree of practicality is to proceed in four stages. 2.1 Steady state calculations First, form a view as to the very long run rate of real growth and inflation that can be anticipated for the economy, and what this will generate/require in terms of private savings, private investment and other macroeconomic magnitudes. Then examine a range of 'balanced growth' or (proportionately) 'steady state' scenarios. These assume that the economy grows steadily at the projected rate, that other key macroeconomic rates (the interest rate, the rate of inflation, the rate of depreciation of the exchange rate) are constant, and that the main fiscal magnitudes are stationary relative to GDP. Thus, for example, the share of revenue in GDP, and the debt income ratio are assumed constant. Within this structure, a variety of different configurations can be examined. For example a higher ratio of domestic debt to income implies both proportionately more debt to be serviced and also inter alia a higher interest rate at which to service it. The share of government receipts going to debt service will rise for both reasons. Offsetting this, there will also be a higher relative contribution of net borrowing to receipts; however, if the domestic interest rate exceeds the growth rate, this offset will only be partial. Given a realistic view of revenues, some idea of the steady state relation between the debt stock and the feasible level of 23 At least it can in the textbook case of competitive markets and a fully employed economy. 24 In the discussion and illustrations in this sub-section, we concentrate on the domestic deficit and on domestic debt. The resources available to government also include external concessional loans (the external deficit) and the associated external debt. As argued in detail in FPMC the considerations governing these are wholly different. We return to this issue in section 8. 79 expenditures on goods and services can then be obtained25. For example, if the share of revenues in income is t, the debt income ratio A, the interest rate r, the growth rate g, then the steady state ratio of spending in income, e26 is constrained to: e = t - (r - g)A (1) As already noted, we would normally expect that a higher A would also imply a higher r. At some point, this relation may become quite steep, reflecting a very high opportunity cost of increased debt. If this involves substantial crowding out of private sector investment, it may also lower the growth rate g, further lowering e, but having obvious additional costs besides. The relevance of this point depends on a number of factors, such as how integrated the domestic economy is in the world economy, how good is the access of domestic investment to international finance27, and how credible the government's reputation is for prudent policies. The stronger the country's performance in these respects, the less sensitive the domestic interest rate is likely to be. 2.2 Convergence to steady states The second step is to examine the paths for fiscal magnitudes, especially debt stocks (given their actual initial value), if certain fiscal flow variables (such as the conventional budget deficit28) are set at some level relative to GDP and maintained there. For example, suppose the government chose to set and maintain a conventional budget deficit at lOOd%'o of GDP. Then, if the initial debt income ratio was AO, the future debt income ratio evolves over time (n) as: A,, = A0 /(l+g)" +[1-l/(I+g gy'd/g (2) which converges reasonably rapidly towards: A,C =d/g (3) as n becomes large29. It is the conventional deficit that usually figures in concerns about fiscal policy, i.e. the deficit after nominal interest payments. Hence the growth rate should also be 25 Obviously, if a government could by magic be catapulted into any steady state provided it committed to staying there, it would choose the smallest debt possible (or better still to hold the largest asset stock permitted). 26 The symbol e is taken to cover expenditure on goods and services, not including interest payments. For simplicity it will be alluded to in the text by the shorthand 'expenditure'. 27 This might take place through partnerships with transnational firms, via FDI, or via capital account liberalisation and stock market developments. 28 For the remainder of this section, any reference to 'the deficit' can be taken to mean the conventional deficit after grants. For simplicity, external borrowing is neglected. Since, currently, net external borrowing and external interest are both relatively small and more or less offset each other, this simplification does not greatly misrepresent the picture. 29 Indeed, equation (2) can be written as a weighted average of the initial and final values, 80 measured in nominal terms. For an economy like Tanzania, the nominal growth rate is in excess of 10% per annum. With a growth rate of 10%, a deficit to GDP ratio (d) of say 2%, if maintained, would imply an eventual debt income ratio of 20%. If the initial debt income ratio was 10%, close to current values in Tanzania, equation (2) shows that it would rise to 15% (halfway to its eventual value of 20%) in a little over 7 years. However, the more interesting question concerns the altered path of government spending on goods and services (i.e. net of interest payments). This is given by: e,, =t +d-rA,, (4) As an illustration, consider a government which has been running a balanced domestic budget, so d = 0. Since the economy is growing, but debt is not, the debt income ratio has been falling. Now consider three alternative fiscal strategies, each to be maintained indefinitely30. Strategy I is to maintain the balanced budget policy; Strategy II is to start running a conventional deficit which is consistent with a constant debt income ratio. Strategy III is to set some higher deficit, which will see the debt income ratio rise to a higher level. For concreteness, suppose the initial debt income ratio is 10%, the nominal growth rate 10%, and the nominal interest rate 12% (substantially higher than the present cost of domestic funds to government, but the right order of magnitude on average in recent years31). Suppose that total (non-domestically borrowed) government resources are around 18% of GDP (again a typical figure for recent years). Then: * Strategy I has the spending ratio e rising from an initial value of 16.8%32 gradually to a final value of 18%. After 5 years of this strategy, the debt income ratio would have fallen to around 6%, and e would have risen to 17.25%. The reason is that the relative burden of the debt falls over time, lowering the proportion of resources that has to be diverted to pay interest. * Strategy II immediately increases e to 17.8% and maintains it there. Since the economy is growing, the government can finance a deficit without raising the debt - income ratio, provided the deficit is restricted to (g x A) x 100% of GDP, or 1%. The rest of the interest A,, = aA0 + (1- a)A. where a = I/(I+g)n 30 There is clearly no particular merit is the assumption of a fixed strategy, and indeed it is unlikely to recommend itself in practice. But it is a helpful device when (i) it is necessary to think ahead, (ii) it is impractical to think very far ahead in detail, and (iii) it is desired to set off on a track that will not lead to enforced radical subsequent revision. Such revisions as are subsequently made can then reflect appropriate responses to changing circumstances, rather than be predictable consequences of poor short run policy. 31 The weighted average yield on Treasury Bills was only 5.7% during 2002. In 1998 and 1999 it was 13.2% and 11.0% respectively: Bank of Tanzania, Monthly Economic Review, February-March 2001, Table 7. 32 (.18 + 0 - .12 x .1) x 100. The numbers are being taken to represent all resources and all development spending, not just the local component of the latter. 81 burden (1.2% - 1%) has to be met from other resources. Notice that government spending would only reach 17.8% of GDP under strategy I after nearly 20 years!33 * Strategy III is the one sketched earlier, where the deficit is raised to 2% of GDP and maintained there, implying a rise in the debt income ratio to 20%. Expenditure now jumps to 18.8%34 of GDP, then declines steadily toward a final level of 17.6%. Once again, it is nearly 20 years before e reaches 17.8%, this time from above. Two important things to note about these examples are (i) that they all describe sustainable trajectories for the budget35; (ii) that each path for expenditure has the same present value36 as the others. Hence the choice between them should essentially reflect the government's preference for its operations to be an increasing or declining share of total activity. If it was felt, for example, that growth would be fostered by a relatively heavy initial emphasis on education, health, roads (i.e. the priority sectors) there would be a case for 'front loading' expenditure and adopting a strategy like III. If, on the other hand, it was felt that creating maximum space for private enterprise was the early key, then a something like Strategy I might be preferred. The crucial point is that, subject to one important caveat, they are all equally prudent. The caveat concerns their differential downside risk to an adverse change, for example in the growth or the interest rates. Since the impact of each is rather different, we consider them separately. Interest rate changes Suppose the interest rate rose very substantially, to 15%37, and that there was no prospect of raising revenue relative to GDP. In all three cases, either the strategy must be revised, or expenditure must be cut. First, suppose that the change happened just as the new strategy was being adopted, and it was decided to maintain it. Then the consequence for initial expenditure would be the same under More precisely, it is 18.8 years. The cross over between the steady state expenditure strategy and others with a constant relative conventional deficit takes place after m years where m is given by the condition: (I + g)' = rl(r - g). Caution should be exercised in reviewing these calculations, since they are sensitive to the specific numbers used. But the general point is clear enough. 34 (.18 + .02 -.12 x.1) x 100 35 This is conditional on the debt income ratios being feasible: i.e. sustainability really rests on the behaviour of the stocks not the flows, and in particular on there being willing holders for government debt at the postulated level. 36 This follows from the requirement that they must all satisfy the government's inter-temporal budget constraint, and share the same initial debt, revenue profile, and by assumption the same interest rate. The relation would be modified to the extent that the interest rate was affected by the trajectory of spending. To the extent that Ricardian equivalence holds, (private agents are not fooled by the government re-phasing a given tax burden), the interest rate would indeed be invariant to this trajectory. 37 Note this is the rate that is expected to hold over the future long term. Such a shift would require a massive shift, probably on a worldwide basis, such as did occur in the early 1980s, and was implicated in the following debt crisis. 82 each strategy; it would require an expenditure cut equivalent to 0.3% of GDP38 from the various previously planned initial levels. Of course the long-run outcomes for expenditure are rather different; however, as can be seen from equation (3), the long-run debt income ratio is unaffected39. Strategy I still eventually yields a negligible debt income ratio, and e converges from its initial value of 16.5% to 18%. Strategy II continues to maintain the initial debt income ratio, and e remains at the new reduced level of 17.5% indefinitely. Strategy III continues to yield a long run debt income ratio of 20%, but since this is more costly, the level of e falls over time from 18.5% to 17%. The cross over point between the strategies is now somewhat sooner at 11.5 years. However, it is possible that the change in the long run projected interest rate might lead to a revision of strategy. For example, suppose the government preferred to font-load its expenditure, but still did not wish to contemplate e falling below 17.25% of GDP. Given the upward revision to the interest rate, a strategy like Strategy III is now inconsistent with the government's requirements. In fact, the maximum continued deficit is now only 1.5% of GDP; this sees initial spending rising only to 18% of GDP, and falling eventually to the lower limit 17.25% (with the debt income ratio rising to 15%). Second, suppose the revision only took place some years into the new strategy. It is now possible that it would be necessary to reverse the strategy. For example, it was earlier shown in the hypothetical example of Strategy III that the debt income ratio would have risen from 10% to 15% after around seven years. If the interest rate revision took place then, the assumed floor to the expenditure income ratio would require an immediate shift to a strategy of type II; The deficit would be cut from 2% to 1.5% of GDP and expenditure stabilised at 17.25 %. If the revision took place still later, the debt income ratio would have become higher than could be sustained given the expenditure floor. However, in the absence of additional sources of revenue, the only way to reduce the debt income ratio would be by temporarily cutting expenditure further below this 'floor'. If it were felt that there was a very sharp rise in the marginal benefit of public spending below some threshold, (and hence a correspondingly high and asymmetric cost of falling below it), then it would obviously pay to be careful to avoid breaching it. However, this does not seem particularly plausible, and it seems sufficient to deal with this type of adverse change if and when it arises. Growth rate changes The other major form of downside risk in these long run calculations is that it might prove necessary radically to reduce the growth forecast. From equations (2) and (3) it can be seen that, unlike the interest rate change, a growth rate change alters the relation between the deficit and the debt income ratio. For concreteness, consider the same set of scenarios and initial conditions as before, but now suppose that the growth projection has to be revised (dramatically) down from 10% to 7%. Strategy I is unaffected, except that the whole process of convergence slows down; 38 I.e. (.15-.12) x .1 x 100 39 The reason for this is that the deficit which controls the level of debt is the primary surplus. If this is fixed, and there is an increase in the interest rate, then the relation between the primary surplus and debt service is disrupted, and the growth of debt will accelerate. Targeting the conventional deficit instead provides a built-in stabiliser. A rise in interest then requires an offsetting reduction in spending, i.e. an increase in the primary surplus. The budget reacts to the extra cost of debt immediately and continuously. 83 after 5 years, for example, the debt income ratio will fall to 7% of GDP as opposed to the previous 6%. Strategy II requires revision. Expenditure has to be cut to 17.5% of GDP (in place of 17.8%), so the deficit is 0.7% (in place of 1%). If Strategy III is unadjusted (deficit maintained at 2%), then the debt income ratio will rise to nearly 29% (in place of 20%) and e will eventually fall to 16.6%. Thus a 1% deterioration in the projected growth rate has the same consequence under the Strategy II approach as a 1% deterioration in the projected interest rate. They have similar but not identical effects in the other scenarios, with the difference being mainly a matter of timing. 2.3 Other forms of transition So far it has been assumed that the economy in general, and certain key fiscal magnitudes in particular, have been in steady state. This has permitted attention to be restricted to two main 'moving parts', the flow deficit and, given the assumed invariance of the revenue share, spending; and the stock of debt. In practice, many other features of the economy are also likely to be in some process of transition between states. Two of these are of especial interest in the fiscal context. Resources available to government Aside from the domestic borrowing considered so far, the two sources of funds available to government are domestic tax and non-tax revenues on the one hand, and external grants and concessional loans on the other. In the situation of Tanzania, both of these are likely to evolve over time. One plausible long run scenario would see the share of domestic revenues rising over time, from their current low level (see Annex 3); it is conceivable that this increase would be offset by a tapering of the external inflow. Indeed, if the government succeeds in its aim of sustaining real growth at 6% per annum or better, then even sustained growth in the aid flow might see its share in GDP fall steadily. An alternative and more optimistic scenario might suppose that this tapering would be long deferred, or be quantitatively less significant than the growth in the revenue ratio. To illustrate the implications of this type of projection, suppose that the initial level of non- domestically borrowed resources available to the government is again 18% of GDP, but that this is expected to grow steadily to 20% over the next ten years, i.e. at the modest rate of 0.2% of GDP per annum. The other assumptions are maintained as in the original steady state illustrations. How might government spending be planned in these circumstances? We examine three options, two of which are adaptations of the steady state Strategy II. * Strategy IV sets out to preserve one feature of the earlier strategy, the maintenance of the debt income ratio at its initial value of 10%. Expenditure again jumps in the first year to 17.8%, and the deficit to 1%. However, instead of staying at this level, expenditure can be raised in the second year to 18.0%, in the third year to 18.2% and so on. The deficit is maintained at 1%, and this, together with the margin of 0.2% each year between revenues and expenditure, is sufficient to pay the interest at 1.2% of GDP. The 1% deficit is added to domestic debt, which therefore grows at 10%, the same rate as GDP. Clearly, the sustainable level of expenditure is higher than these early values, but it would not be possible to raise expenditure in the early years without breaching the limit on the debt income ratio. 84 * Strategy V preserves the other feature of the earlier strategy, the maintenance of the highest constant share of expenditure in GDP consistent with overall resources and the initial debt. This turns out, on the illustrative figures given here, to be 19.6% of GDP. Domestic debt builds up over the transitional decade to a level equivalent to 19.5% of GDP and then stabilises there. The deficit initially jumps to 2.8% of GDP, and then tapers over the ten-year revenue build-up to a little under 2%, at which it stabilises. * Strategy VI is a hybrid of the previous two. It sets out to bring expenditure forward, reflecting the anticipated rise in resources, and this again requires a rise in domestic debt. However, this rise is designed to be temporary, so that the debt income ratio is again brought back to 10% in due course. This implies that some phased increase in expenditure will be appropriate. Ultimately, expenditure can be raised to 19.8%, since this is consistent with maintenance of the debt income ratio at 10%, but it will initially have to be held below this (and indeed below the 19.6% of Strategy V). Suppose, for example, that the government decided to raise expenditure immediately to 19% and maintain it there until the debt income ratio had reverted to 10%. Given the present illustrative numbers, this would take 15 years. The debt income ratio would peak at 14.25% after 7 years. The expenditure path would jump to 19%, be maintained there for 15 years, and then be increased to 19.8% and maintained there. The deficit jumps to 2.2%, falls steadily over the 15 year transition (1.7% after 5 years, 0.6% after 10, 0.2% after 15), and then reverts to the steady state value of 1%. Financialfeatures These may also be undergoing transitions from disequilibrium states, and this may alter and possibly restrict fiscal policy choices. An important example, which is fortunately not relevant in the Tanzania case, is when there is a very large inherited stock of domestic debt with heavy associated debt service, which dramatically restricts both expenditure and financing options. Tanzania does of course have very large external debts, and is engaged in a variety of processes to relieve them. From the current perspective, the two key features are that current interest costs are relatively modest, and that net external borrowing is also now modest. Other relevant features are the low levels of demand for real balances and probably for credit. These reflect the past lack of a private enterprise culture, and historically high inflation rates; they are likely to take a long time to unwind. 85 ANNEX 4: DATA 86 lFable 1: Tanzania Mlacroeconomic Indicators Indicator nit 199111 19911 19921 19931 19941 19951 19961 19971 19981 19991 2000 PIpiulation/2 Milliolis 23.9 24.6 25.3 26.0 26.7 27.5 28.3 29.1 30.0 30.9 31.9 1ercupita Income/2 1Us$ 160.9 180.7 167.0 149.2 156.2 176.9 210.3 235.6 220.0 240.0 280.0 (il)lD (irowth/2 6.2 2.8 1.8 0.4 1.4 3.6 4.2 3.3 4.0 4.7 4.9 Gross )Doiestic Savings/2 Bit. TZS -4.7 -6.6 -32.3 -53.4 -26.6 25.3 202.2 321.8 229.0 314.3 682.4 Gross investiments/2 Bil. 'IZS 216.9 286.1 373.0 433.5 566.7 597.8 627.2 700.8 902.6 999.6 1281.0 Inflation/2 % 35.8 28.7 21.8 24.0 33.5 27.4 21.0 16.1 12.9 7.8 6.0 Exchanige Racl/2 TZS/US$ 197.6 222.6 301.9 414.5 509.6 574.8 580.0 624.6 664.7 720.0 840.0 External Sector Exports/2 Mil. US$ 470.0 482.2 555.6 747.9 937.6 1265.8 1300.9 1246.4 1141.5 1189.7 1325.8 Imports/2 Mil. US$ 1273.6 1336.3 1438.5 1983.6 1812.6 2140.3 2028.5 1948.2 2360.1 2241.4 2094.4 Current Accouiii1 Balance/2 Muil. US$ -558.9 -736.1 -800.1 -1022.0 -711.0 -647.6 -455.0 -589.1 -993.0 -793.5 -514.5 Balance ofl aynicrtsI2 Mil. US$ -200.0 -260.0 -228.1 -634.4 -446.7 -373.6 -223.9 -355.4 -645.8 -372.8 -336.8 Foreign Reserves/2 NMil. US$ 238.9 301.2 458.5 271.9 431.8 288.5 500.6 638.7 599.0 775.6 974.0 External Debt/I Uil. US$/I 5.6 6.0 5.9 7.5 8.0 8.4 7.8 8.0 7.9 8.0 7.6 Foreign Direct Investmcnt/2 Mil. US$ 0.0 0.0 12.0 20.0 50.0 119.9 150.1 154.7 172.2 183.4 192.8 Tourism Eaamings/2 Mil. liJS$ 65.0 94.7 120.0 146.8 192.1 258.1 322.0 392.4 570.0 733.3 739.1 Monetary Sector Average Dcposit Rate/2 % 26.0 26.0 26.0 24.0 25.0 21.0 16.7 10.0 7.0 6.5 6.9 Average LcndiigRate/2 % 26.0 26.0 30.0 30.0 31.5 35.5 33.5 26.5 21.4 20.4 19.1 Growvth in Money Supply (M3)/2 % 43.3 26.9 42.7 39.3 35.5 32.2 8.7 13.3 10.8 18.6 14.8 Government Finance Total Domestic Rcvenue/l Bil.l7ZS 94.7 137.1 173.6 164.1 242.5 331.2 448.3 572.1 619.1 689.3 777.6 Tax Revenie/I 13il. IZS 81.5 118.3 153.4 146.4 220.4 299.9 383.7 505.4 566.1 616.3 685.1 Non-Tax Reveinuic/l Bil. TZS 13.2 18.8 20.2 17.7 22.1 31.3 64.6 66.7 53.0 73.0 92.5 Total Expenditure/I Bil. T/S 127.9 201.2 223.8 337.9 410.5 453.4 500.1 711.7 856.2 927.7 1168.8 Recurrent Expeiiditure/I Bil. IZS 111.6 163.0 191.2 273.2 335.8 386.6 470.0 587.1 669.6 791.2 808.9 D)evclopmcnt l xpcndiurrc/Il 13iii. TZS 16.3 38.2 32.6 64.7 74.7 66.8 30.1 124.6 186.6 136.5 359.9 Grants/I Blil. I;ZS 27.7 22.9 32.8 58.3 76.9 67.3 46.9 115.4 119.4 169.9 280.3 IFiscal l3alance/I 13iI. TZ7S -33.2 -64.1 -50.2 -173.8 -168.0 -122.2 -51.8 -139.6 -237.1 -238.4 -391.1 /1 F:iscal ycar is used. and it ends in mncitionied year June 30th. /2 Calcndar year is used. and it ends in mentioned year December 3 Ith. Sources: Tlie Planning Commission. The United Republic of Tanzaniia. Economilci Survey. various issues. BureauL of Statistics. Planninig Commissionl. National Accouits ol Tanzaniia. various issues. Ihe Bank of Tanizaniia. Economiic Bulletin. various issues. 88 Table 2: BALANCE OF PAYtMSENTS (in millions of U'S dollars). Items 1990 1991 11992 1993 1994 1 1995 1 1996 1 199? 1998 1 1999 2000 CURRENT ACCOUNT -559 .736 -704 .1.022 -711 -646 -461 -555 -995 -837 -534 Goods -779 -865 -916 -836 .790 -658 -449 -395 -778 -825 -67? Exports (fob) 408 362 401 439 519 683 764 753 589 5433 663 imports (fob) 1,186 1,228 1.317 1.275 1.309 1,341 1.213 1,148 1.366 1.368 1.335 Sersices -157 -158 -169 -390 -85 -217 -279 -306 .443 -227 -96 Receipts 1'll 142 168 311 418 583 537 494 553 646 664 Paviments 288 300 337 701 503 800 816 800 996 873 760 Inicomie -185 -184 -225 -148 -123 -110 -72 .123 -125 -76 94 Receipt-s 6 8 8 21 3 1 32 42 45 48 56 63 Payntienits 191 192 234 169 1 53 142 113 168 173 13? 157 Currenit tranisfers 562 471 606 351 287 338 338 270 350 290 328 Inflows 593 504 641 381 312 371 371 337 377 413 405 Government 538 480 506 371 215 23 6 236 239 261 25? 257 PriNate 54 23 135 1 0 97 135 135s 98 116 162 147 Ottfloss 3 1 33 3 5 30 25 32 3?2 68 27 123 77 CAPITAL ACCOUNT 327 353 298 201 263 191 191 167 300 323 329 Capital trattsfcrs 327 353 298 201 263 191 191 167 300 323 329 Ittflows 37 353 298 201 263 191 191 167 300 3 23 329 Outflows 0 0 0 0 0 0 0 0 0 0 Acqttsition,disposals of nott-produced 0 0 0 0 0 0 0 0 0 0 0 non-ftinattici assets FINANCIAL ACCOUNT 42 118 107 57 -107 140 0 -136 -18 90 -130 Direct iinsestintent 0 0 12 20 50 ISO 149 158 172 183 193 Abroad 0 0 0 0 0 0 0 0 0 0 ( It Tateanaii 0 0 12 20 50 150 149 158 171 183 193 Portfolio twvesttmen 0 0 0 0 0 (I 0 0 0 0 ( Othier investtnent 42 118 95 37 -157 -10 -148 -294 -190 -93 -323 Inflow of fitnaticial resouirces 237 271 332 615 435 437 319 417 436 423 250 Disbuirsernenit of Govenitnetut loans 224 261 268 372 274 244 225 270 274 336 166 Trade credit and otlher finanicial flows 1 3 1 I 64 143 122 142 77 119 1310 83 82 Disbursetnent of loans to banks &, otlier sect- 0 0 0 100 38 51 1 7 28 33 4 Ouitflo%s of finiatucial resouirces 195 153 27,7 578 591 448 467 711 626 516 5 73, Repaytenett of govenimettt loans 169 146 237 498 491 273 362 758 479 405 486 Trade credit and other finaticial flows 26 7 0 68 76 163 43 -108 100 57 36 Repayment of loans by bank-s & other sector 0 0 0 1 2 24 1 3 62 61 47 54 5 I Errors attd onmissions 48 33 46 28 94 -70 25 -32 97 96 -69 OVERALL BALANCE -142 -232 -253 -737 -461 -386 -245 -556 -615 -328 -40t4 FINANCING -142 -232 -253 737 461 386 245 556 615 328 404 Net Reserve assets I- ittcrease) -140 -85 -254 157 -77 60 -165 -84 -11 -121 -138 Exceptsiontal finanlciute 2891 317 506 579 538 326 410 640 626 450 541 Aricars 100 345 207 199 296 269 31 2 131 1311 147 106 Reschediuling 156 0 1319 137 145 0 0 -159 1 751 116 0 Delbt IbrgieNcress 11I 0 25 24 24 0 0 194 151 88 131 Use of Ftuttd credit 14 -27 136 70 31 -IS -28 77 21 45 1 1 Grants borrousing for BOP puirposes - - - 150 74 57 98 156 169 99 304 %letnorantdurn iternts: GI)lntp) Muill. USD) 3.444 3.768 4.042 3.967 4.170 4.866 5.953 6.995 7.594 8.435 8.34 CAB GDP -16 -20 -17 -26 -17 -13 -8 -8 - I3: -10 -6 CAB GD)P level cttrrcnt official transfers) -3? -3-2 -30 -35 -22 -18 -12 -II -17 -IS .u Gross Official Reserves 193 242 394 228 331 271 441 623 599 776 974 Weeks of ltnports 7 8 12 6 10 7 II1 1 7 1 3 1 8 24 S.-rc B.Iw,cc ouf P.Y-nets Deparunueun of the Batk of"Tanmi.oj 89 'I'ahle 3: SilUlMlARY ()OF C N'-RAL. (;OV'ERNMN)IEN'1' Ol'ERII'.'IONS (in billions of 7'anzania shillings) I F |9] V119l1 F 92^1 VY931 FY941 tY95| FY961 IV971 1X981 [ 991 1; Viii lii.ol lotal Rcvcweu 95 133 174 164 242 331 448 572 619 689 778 93 Tax rcvcniue 81 118 153 146 220 300 384 505 566 616 685 82 Nontax rCvCIItIC 13 1; 20 18 22 31 65 67 53 73 93 102 lotal lxpeadittire ad Nt L.ending 126 152 195 305 371 412 461 515 730 819 1169 1307 Rccirrent exp,cndirurc 104 135 162 240 296 346 415 486 544 683 809 1021 Developnefitexpendittireanid 22 16 33 65 75 66 46 29 345 137 360 28 patl of net leniding Overall balanec (checks isstied) -32 -19 -21 -141 -128 -81 -13 57 -111 *130 -3191 -378 Adjustncit th cash and uthicr 4 -9 -2 7 -37 -90 -73 -37 .82 -27 4 -4 items (net) Overall balance (checks-cicarcd) -36 -27 -23 -134 -165 -170 -86 20 -194 .157 -393 -381 Overall balancc(aflcr grants) 4 4 11 -83 -52 -117 -11 102 *37 18 -115 -88 Finanicing: 36 27 23 134 165 170 86 -20 194 157 395 448 Extetnal granms 28 23 33 58 77 54 75 81 156 175 280 293 Foreign Financing, net/I I 9 23 29 48 26 -45 -49 34 -19 105 121 Domestic Borroving 7 -4 -32 47 41 90 56 -53 4 -6 9 -2 Privatisation Frnd . .. . .. .. .. .. .. 7 .. 27 Bank & Parastatal Rccapitalization .. .. .. .. .. .. .. .. .. .. 10 Changes in arreas .. .. .. . Colitigeelcy .. .. .. .. .. .. . .-. As % of i'DP at market nrices Total Revenuc 12.5 13.5 13.6 10.2 11.4 11.8 13.0 13.4 12.1 13.0 11.3 12.8 Tax revetuc 10.7 12.0 12.0 9.1 10.4 10.7 11.1 11.8 11.0 11.7 10.0 11.4 Nontax revenue 1.7 1.5 1.6 1.1 1.0 1.1 1.9 1.6 1.0 1.4 1.3 14 Total Expcndilure and Net Letiding 16.6 15.3 15.3 19.0 17.5 14.1 12.2 12.0 11.1 15.2 17.0 18.0 RecarrTntexpendiltire 13.7 13.7 12.7 14.9 13.9 12.4 12.0 11.4 10.6 12.7 11.7 14.0 Developnient expcnditure aird 2.9 1.7 2.6 4.0 3.5 1.8 0.2 0.7 0.5 2.5 5.2 3.9 part of nct Icnding Overall balanc (checks issrted) -4.2 -1.9 -1.7 -8.8 -6.0 -2.3 0.8 1.3 1.0 -2.2 -S 7 -5 2 Adjustncnt to rash and other -0.5 -0.9) -0.1 0.4 -1.7 1.8 2.7 -0.9 -1'8 -1.2 -0.1 -0.1 items (net) Overall balance (checks-cleared) -4.7 -2.8 -1.8 -8.4 -7.8 -0.5 3.5 0.5 -0.8 -3.3 -5.7 -5.2 Overall balance (after grants) -0.5 0.4 0.9 -5.1 -2.4 1.5 2.2 3.2 1.9 1.0 -1.7 -1.2 Financing: 4.7 2.8 1.8 8.4 7.8 4.5 1.9 -0,5 1.7 2.4 5.7 6.2 External grants 3.6 2.3 2.6 3.6 3.6 3.8 1.4 1.9 0.9 3.2 4.1 4.0 Foreign Finalicing.net/l 0.2 0.9 1.8 1.8 2.2 -1.4 -1.0 -1.1 0.7 -0.5 1.5 1.7 Domestic Borrowing 0.9 -0.4 -2.5 2.9 1.9 2.1 1.5 -1.2 0.1 -0.3 0.1 0.0 1/ Iicltides foleigil gratlis prior to FY87. I prsvissio55al Ssurce: The Tanzanian authorities. 90 Table 4a: BUDGET FRAME FOR 1999/2000 - 2003/04 (ACCOUNTING), (TShs. Billion) 1995/96 1996/97 1997/98 1998/199 1999/2000 2000/01 2001/02 2002/2003 12004 | Atual Atual IActal Acual AItkeal l,ikcly Out-ttirn 1'ro;. 3ruj, r 0j. _ l.lTO'IAL RESOURCES 500 731 856 928 1262 1489 1574 1559 1649 Domcstic revenue 448 572 619 689 778 897 989 1098 1225 PRBS/MDF/BOPS 38 114 82 106 128 246 243 192 188 Project loans and grants 25 64 174 118 340 276 281 211 176 IIIPC interim relief-Multilateral I1 47 54 58 59 HIPC interim relief-Paris Club Non Bank Borrowing -6 4 27 45 Bank Borrowing 63 -48 -24 8 8 Adjustment to cash -85 -23 -45 -2 Privatisation Funds 18 25 - 7 23 7 11. TOTAL EXPENDITURE 500 731 856 928 1262 1489 1574 1559 1649 RECURRENT EXPENDITURE 470 606 670 791 902 1174 1236 1287 1427 CFS 155 188 250 225 266 310 306 264 269 Debt service 120 164 227 202 211 252 246 196 199 interest 66 112 101 91 128 117 116 112 119 amortization 54 53 126 III 82 135 130 84 80 Otlcrs 35 24 23 23 56 58 60 68 70 Recurreit Exp.(cxcl. CFS) 315 384 420 566 636 801 930 1023 1158 o/wv Salarics & wages 156 199 219 232 285 306 356 408 452 Other Charges 153 85 82 171 216 335 370 423 499 Designated Items* 100 119 163 135 160 204 192 207 P'aymcint ol Arrears 34 63 CON'I'INGENCY I 19 21 DEVELOI'MENT EXPENDITURE 30 125 187 137 360 314 319 251 221 Projects 30 84 187 137 360 314 319 251 221 Local 5 20 24 19 19 33 33 35 40 Sonigo songo cnerg Ftind 5 5 5 5 Foreign 25 64 163 118 340 276 281 211 176 Otlier Programme Assistanec 40 Notes * Iicltides Special Expenditure. Road Iunlid, l'arastatati Wages, I'RA and Retenition Schemlie * Startinie 1998/99. Electricity 13ill is inci(led in the Special I'xpenditurc Sourcc:Tanzania Athllorities 91 Table 4b: IJUDGET FRAME FOR 1999/2000 - 2003/04 (ACCOUNTING) AS % OF GDP 1995/96 1996/97 1997/98 1998/99 1999/2000 2000/01 2001/02 2002/2003 12003/2004 Aetuial Actual Actual Actial Actual Likely Out-tuiri ProPoj. |Proj. 1. TOIAL RESOURCES 14.7% 17.3% 17.4% 16.8% 18.3% 19.4% 18.6% 16.7% 16.0% Domestic revenue 13.2% 13.5% 12.6% 12.5% 11.3% 11.7%) 1 1.7%/6 11.7% 11.9% Importsupport/o(il. 1.1% 2.7% 1.7% 1.9% 1.9% 3.2% 2.9% 2.1% 1.8% 1'roiect loan and grants 0.7% 1.5% 3.5% 2.1% 4.9% 3.6% 3.3% 2.3% 1.7% IIIPC intierim reliefl 0.2% 0.6% 0.6% 0.6% 0.6% Non Banik 13orroxving -0.2% 0.1% 0.6% 0.8% 0.0% 0.0% 0.0% 0.0% 0.0% Bank Borrowing 1.8% -1.1% -0.5% 0.1% 0.1% 0.0% 0.0% 0.0% 0.0% Adjustment to cash -2.5% 0.0% -0.5% -0.8% 0.0% 0.0% 0.0% 0.0% 0.0% P'rivatisation Funds 0.5% 0.6% 0.0% 0.1% 0.0% 0.3% 1.1% 0.0% 0.0% 11. TOTAL EXPENDITURE 14.7% 17.3% 17.4% 16.8% 18.3% 19.4% 18.6% 16.7% 16.0% Rl,ClJRRi.N1I .Xl'l.NDIIURE 13.8% 14.3% 13.6% 14.3% 13.1% 15.3% 14.6% 13.8% 13.8% CFS 4.6% 4.4% 5.1% 4.1% 3.9% 4.0% 3.6% 2.8% 2.6% Debt service 3.5% 3.9% 4.6% 3.6% 3.1% 3.3% 2.9% 2.1% 1.9% interest 1.9% 2.6% 2.1% 1.6% 1.9% 1.5% 1.4% 1.2% i.2% amortization 1.6% 1.2% 2.6% 2.0% 1.2% 1.8% 1.5% 0.9% 0.8% Othcrs 1.0% 0.6% 0.5% 0.4% 0.8% 0.8% 0.7% 0.7% 0.7% RccurrentExp.(excl.CFS) 9.3% 9.1% 8.5% 10.2% 9.2% 10.4% 11.0% 10.9% 11.2% o/Iv Salaries & wages 4.6% 4.7%X6 4.4% 4.2% 4.1% 4.0% 4.2% 4.4% 4.4% Other Charges 4.5% 2.0% 1.7% 3.1% 3.1% 4.4% 4.4% 4.5% 4.8% D)esignated ItIms (0.0% 2.4% 2.4% 3.0% 2.0% 2.1% 2.4% 2.1% 2.0% PIaymiient ol-Arrears 0.(% 0.8% 0.0% ().tIO 0.0% 0.8% (.0% (.0% 0.0% DEVELOPMENT EXPENDI'I URE 0.9% 2.9% 3.8% 2.5% 5.2% 4.1% 3.8% 2.7% 2.1% Proicets 0.9% 2.0% 3.8% 2.5% 5.2% 4.1% 3.8% 2.7% 2.1% ILocal 0.2% 0.5% 0.5% 0.3% 0.3% 0.4% 0.4% 0.4% 0.4% Foreign 0.7% 1.5% 3.3% 2.1% 4.9% 3.6% 3.3% 2.3% 1.7% Othcr Programmc Assistanice (.0% 1.0% 0.0% 0.0% 0.0% 0.0% (.0% 0.0% 0.0% Source: 1'abIe 4a 92 Table 5a: BU'DGET FRAME FOR 1995/96 -2003/04 (ANALVTICAL). (TShs. Billion) (in millions of Tanzania shillings) 1959 19/7 1997/981 1998/9 19920 000 01 02 12002120013 '200312004 Acul Ata Actual natual A=ua Liey ttr r. Pro Proj. Domestic resenue 448 572 619 689 778 897 989 1098 1225 0O'W Road Toll 34 37 38 38 44 51 55 67 Total Expenditure 446 644 730 817 1180 1289 1425 1453 1568 Recurrent espendit.re 416 519 544 680 020 976 1.106 1203 1347 Interest on external debt 39 38 48 60 47 42 43 4 1 44 Interest on domestic debt 27 74 53 30 81 75 73 71 75 %Vaves salaries 156 199 219 232 285 306 5 6 408 452 Goods scr, ices'transfers 194 209 224 355 406 553 634 682 7 76 ssRoad Sector 1 6 20 1 7 38 38 SI 56 63 72 Special cxp 35 29 44 42 37 108 84 8S CFS lOthers) 35 24 23 23 56 58 60 68 70 TRA 16 17 20 22 ra,islatat \\ ages 24 28 34 52 Retention Scheme I I 14 29) 23 36 4 1 45 s0 Election Costs 3 Census Other Char.es 153 85 82 171 216 335 370 423 499 IThclpm1...ent ex,penditure 30 125 187 1 37 360 314 319 251 221 .Pnic~ts 30 84 187 137 360 309 314 246 216 ' cal 5 20 24 19 IQ 33 33 35 40 t orcian 25 64 163 118 340 276 281 211 176 (tthet Pirogranmme *'%ssistance 40 Encrg) Fund Sonco song0 S S 5 5 0jseralIl deficit (checks issued) - before grants 2 -72 -III -1 27 -402 .392 -43,6 -355 -343 Grants 47 115 119 170 292 417 394 325 287 Batance of Paymient Suppon 30 67 2 619 73 1713 142 tiS 11 os , NDF!PRBS 70 71 132 126 97 74 project grants 1 7 49 158 lOt 208 197 198 151 116 H-IPC interim relief-?slultilaleral I I 47 54 58 59 Oserall deficit (checks issncdl - after grants 49 43 5 43 -111 25 -42 -30 -57 Adrustnmcnt -85 7 -45 - Os es-all di,ficit (chseck. cleared) -36 43 15 -2 .113 25 -42 -30 -57 Finanicinig 36 -43 1 5 2 113 -251 42 30 57 Foreign -35 9 64 -19 lOS 1 6 5 5 5 2 57 Pro6ramosni loans 5 47 50 37 55 73, tOt 77 77 project loan 8 IS5 57 1 7 1 33 79 83 59 60 anmortization -50 .53 -72 .73 -82 -135 -1310 -84 -80 Local (nell 7 1 -53 -50 2 1 8 Bank (ret) 63 -48 -24 8 8 Non-bank 8 -5 -26 1 3 borrosVinc -6 4 27 45 1mottication -4 -53 -38 -234 Pris atsation Funds 1 8 25 7 23 7 Change in Arre-r -34 .63 Bank Recapitalisation 89 CuntiLenc., -l -19)-2 Financing Gap (;/)Pw ~~~~ ~~~~~~3394 4236 49291 5532 6885 7664 8482 9352 10311I O3C fur diitr-ibotlio 299 494 574 615 0 Pyiir- I kf,/ ium h/, ,c -,wudI /1 411 -It) -2-4 2 -321) -243 -22-4 6-,,vriiiii S~iih i,-s iu-b 32 53 -5 9 -42 - ~ -//-- 104 -122 Note Recurrent expencrditure under the Poserr, Reduction Growth Facilitv I PRGFI in Tshs 933 1 billion schile the budgeted figure is Tshs 952 5 billton This divergence is explained by Tshs 5 0 billon ncr, included ini the road sector. Tshs 4 5 billion have been re-allocated from development expetiditure to recuirrent expsettditure under the road sector TIte remaining diff'erence of Tshs 10 billion is due to o,f r,rears - tile INIF presents this amount belo%% the line ushile the budget presents this amount under recuirrent expendtture Has ing taken the aforementioned into consideration the budget figurs are ci Nus-ii IlP PRBS for 2000 2001 includ, contribution from forser %IDF US$S88Omill 1shs7O.400 mill ) and EU. USS 11 c ntiIll slhs Q5O0mill l and US S 12 4 millI(shs lOOOO mill I 93 Table 5b: BU'DGET FRANME FOR 1995/1996 - 2003/04 (ANALYTICAL) % OF GDP 1995/96 ~~1996/7 197/8 19989 '999200 00/1 01/00 002/2-003 2003/2004 Acul A"ta A.tua AlIa I Actual LikelyOa-trn Pro.Ij Proi. Pro.4 Domestic r-nese 13.2% 13.5% 12.6% 12.5% 11.3% 11.7% 11.7% 11.1% 11.9% OW, Road Fund 000/0 08%' 07% 07% 0 5% 0 6%/ 0 6% 0 6% 0 7% Retention Fund Total Expenditure 131% 15.2% 14.8% 14.8% 17.1% 16.5% 16.8h Is.5% 15.2% Recurrent espenclture 12 3% 12 3% 11 0%/ 12 3% 11 00/. 12 7% 130/% 12 9%1 13% Interest on external debt I1% 0 9%/ 1 0% I 1% 0 7% 0 5%1 0 5% 0 4% 0 4% Intrest on donmestic debt 0 8% 1 7% 1% 0 5% I 2 I W9.00. 0 9% /. 0 7% Wages salaries 4 6% 4 7% 4 4% 4 2% 4 I9' 409'.1 4 2% 4 4% 4 4% Gloods SC, ices'transfers S 7% 4 9%' 4 5% 6 5%, 500' . 7 2% 7 5% 7 3%1 7 5%; ciRood Fond 0 5% 0 5% 0 3% 0 7% 0 5% 0 7% 0 7% 0 7%' 0 7% Spccial e\p I 0/6 0 7% 000/". 0 8% 009'l. 05 il 139' 0 9%' 08% CFS OlOhets) I 0%1 0 6% 035% 0 4% 0 8% 0 8% 0 7% 0 7%. 07% T'RA 0 094. 0 4% 0 3% 0 4% 0 3% 0 0%1 00O% 009".6 00%i/ I'aritstalal Wa'.es 0 00/. 0 6% 06%" 0 69'. 089'l. 00% 009'.I.000O/, 00%1 Retention Scheme 009'. 0 3% 03% 05% 03%' 0 5%' 0 5% 0 59% 05%I Other CharLes 4 5% 200/% 17% 3 1% 31% 4 4% 4 4% 459'l. 4 8% De, elopment expenditure 0 9%" 200'. 38% 2 5% 52% 4 1% 3.9% 2 7% 2 1% 1"J'rijCS 0 9% 2 00/. 38% 235% 5 2%' 40/% 3 7% 2 69'. 21% Lo'cal 0 2% a05% 0 5% 0 3% 039'. 0 4% 0 4% 049'. 0 4% Fore,0. 0 7% 1 5% 3 3% 2 1% 4 9%, 3 6% 3 3% 2 3%' 1 714 Other Procramme Assistance 0 00/. I19/. 0 09/. 0 00/ 0 09' 009' 00/. 0 0% 0 0%1 0. arlI deficit (checks issued).- before grants 019'l. .1 79.'. -2 3%, -2 39'. .58 S. 5 1% .31 -38% -3 3% Grants 14% 2 7% 24% 3 1% 4 2% 5 4% 4 6% 3 5%' 2 8% import support OGL 0 00/. 1 6% 0 00/. I 3% I 1%l 2 3% I 7% I 2% 1 1% project irrants 0 59'. I 2%' 2 4% I 8% 300'. . 2 6% 2 3% 1 6% 1 1% H-IPC interim relief 0 2% 0 6% 0 6%4 06%/ 0 6% Osetrail deficit i checks issued) - after grarnts 14% 10%i 0 2% 0 8% .1 tR6' 0 3. -035% .0 39. -059'. Adjustment .2 59'. 0 09'. 01% .089'.i 009'. 00%k 00O% 009-i 0 0% 0Oreroll deficit (chtecks cleared) .1.1% 1.0% 0.3% 0.0% -1.6% 0.3% -0.5% .0.3% -0.5% Financing 1.1% .1.0% 0.3% 0.0% 1.6% -0.3% 8. 5%- 0.3% 0.5% fo'irei -I 010% 0 29' 13% -0 3% 1 5%' 0 29'. 0 6%/ 0 6% 0 6/. inmportsupportloans 0 2% 1 1% 1 6%4 0 7%' 08'. 0 99'. 12% 0 8% 0 7% projectoona 0 2% 0 4% I12% 0 3% 1 9% 09. 1 %I09/. 0 69'. 0 6% anirirization -l 59'. -l 2%1 .1 59' .139' .1 2%1 -l 89% -135% .009% O .08% Local (net) 21% -1 2% -1 00/. 0 49' 01% 0 0'.6 0 00/. 009'. 0 09'. Bank lnetl I8% -1 1% .0 5% 01% 01 009'.OM 0 00/ 0/. 00%00%6 Non-batnk 02% .0(9'i. .0 5% 029'l. 00 009' 11. 009/. 009'. 0 0% bo,rni, in. -0 2% 0.19' 0 6%/ 084 009'. 0 09'. 00/. 0 0% 0 0% amortizal-e .01% 00O. -1 1% .0 79'. .3 4%' 009O' 00O% 00% 009/. Prmisatiaio Funds 0 59' 06% 0 09.'. 0 1%' 009'. 0 39' 01% 0 0% 0 0% Change in A rroars .0 8% 009/. 009.. 009'. *0 8".6 00%l 000'. 0 0% (;I3Im,v 3.394.100 4.235.551 4.929,469 5.532.096 6,884.940 7.664.187 8.482.1911 9-351.997 10.310.999 /ivsi, / fijktrih,,i, /''ii. ....): 20%' 0 9% .0 2% .0 7%' .4 096 .3 c0, -3 8% -Z ci'. -22 (ue"iiiiiiii wwio/I d) i/itc/ I 00'. I 29'. I 5%' 0 29'. .0 '.9'. .1 0%' .1 4'. - *I . - I 2% So-rre Table 5., 94 Table 6a: SECTORAL - RECUIRRENT EXPENDITURE, 1995/1996-2000101 (TSh. Billion) ;OTE \OTE HOLDER 1995/96 | 1996/97 | 1997/98 | 1998/99 | 1999/00 | 2000/01 BUTDGET BUDGET BUDGET BUDGET BUDGET BUDGET ADMINISTRATION 23 Accountant General 22.5 27 Registrar of Political Parties .0 .0 3.4 2.6 2.7 2.8 30) Presidents Office & The Cabinet 6.7 9.5 10.5 15.7 19.6 24.6 26 ', I Vice President .2 .5 .8 .9 1.0 1.4 32 Ci% il Ser\ ice Dept. 1.6 2.0 2.4 3.7 3.4 43 33 Ethics Secretariat .0 .1 .2 .2 .2 .2 34 Foreign Aflairs 11.8 10.5 13.3 18.3 17.1 20.5 35 Perm. Comm. Enq. .1 .1 .2 .3 .2 .3 36 Cis il Ser ice Comm. .I .I .1 .2 .2 .5 25 & 37 Prim.eMinisters Office 4.9 3.9 9.1 14.1 4.7 4.9 40 Judiciarv 3.0 3.6 4.6 5.4 7.9 8.8 4 1 Justice &Constitutional Affairs .6 .8 .8 2.2 1.1 1.5 42 Olice Ot The Speaker 2.3 3.2 3.6 5.3 5.9 6.6 45 Exchequer and Audit .4 .5 .7 .8 1.1 1.8 50 Finance 22.1 20.5 33.5 51.1 49.7 51.3 5i Home Affairs 1.4 1.6 2.0 4.6 5.8 5.6 54 Radio Tanzania .3 .3 .3 1.2 1.2 1.6 55 In\ estment Promotion .2 .1 .7 .6 .2 .0 56 NORALG .. .. .. 6.2 5.3 17.5 57 Defence & National. 1.2 1.6 2.4 4.5 2.3 4.1 59 Iass Reform. Commi. .I .I .1 . I . I .2 60 Industrial Court Of Tanzania. . .1 . I . 1 .2 61 Electoral Commision 8.6 .9 1.4 3.0 16.2 30.9 63 Local Govt. Ser\ i. Comm. .2 .1 .1 2 2 .3 64 Information and Broad. .6 .. ,. 1.4 66 Planning Commission .7 .6 .9 2.7 1.9 2.3 Sub Total 67.2 60.6 91.4 145.2 148.3 192.0 DEFENCE AND SECU:RITV- 28 Police Force 17.8 23.8 28.8 29.7 34.9 42.0 29 Prison Ser\ices 11.0 12.2 17.2 16.4 19.4 20.9 38 Defence 46.9 63.8 72.5 73.2 81.7 89.- 39 National Service 5.7 7.5 9.5 11.4 12.1 12.5 Sub Total 81.5 107.3 127.9 130.8 148.1 164.6 SOCIAL SERN'ICES 46 Education 10.5 12.1 17.1 19.6 21.0 28.6 49 Water. 2.0 1.5 2.8 6.4 3.4 5.9 52 Health 13.3 18.8 25.8 37.2 33.8 40.1 53 Comm. Dev.Wome. Aff. .7 .9 1.4 1.8 2.3 20 65 Labour Youth Develop. 1.3 1.3 1.5 2.0 3.0 3.2 67 Teachers Sers ice Comm. .1 .1 .1 .1 .2 1.5 68 Science. Tech.& H Ed 17.3 21.3 22.9 32.8 32.3 44.9 70-89 Regions 85.3 99.8 118.3 120.9 168.6 206.5 Sub Total 130.5 155.8 189.8 220.8 264.6 332.8 ECONOMIIC SERN'ICES 47 Works 21.7 2.3 24.4 4.9 38.(1 56.0 48 ILands. Hous..& LUrb. Dcv. 1.3 .6 1.8 3.4 3.6 4.3 58 Energ\ and Minerals .2 .5 .9 2.1) 3.0 62 Comm & Transport .7 2.3 5.7 13 4 10.7 10.5 Sub Total 23.9 5.7 32.8 23.7 55.3 73.8 PRODUCTIVE 24 & 4 3 Agriculture & Li estock. Coop. & Marketing 1 1.8 11.9 13.9 19.3 12.3 12.1 44 Industries and Trade 1.6 1.4 2.9 4.6 3.6 4.1 69 Tour.. Nat. Res. & Ens. 5.5 2.3 2.8 7.9 8.7 13 5 Sub Total 19.0 15.5 19.6 31.9 24.7 29.6 CFS 20 State house .4 .9 .8 .9 1.0 1.1 22 Public Debt 171.7 217.7 207.1 238.0 294.4 287.5 Suh Total 172.1 218.6 207.9 238.9 295.3 288.6 GRAND TOTAL 494.1 563.5 669.5 791.2 936.3 1081.5 Note Data for FYOI-all \otes are from the Flash Reports Saurce: NMoF. Appropriations Accounts and Expenditure Flash Report 95 Table 6b: SECTORAL-DEVELOPMENT EXPENDITUTRE (TShs. Billion) V'OTE _\'OTE HIOLDER J 1995/96 1996/97 1997/98 1998/99 | 1999/00 2000/01 | BUDGE1BUDGE1BUDGE1jBUDGE1BLUDGElBUDGET ADMINISTRATION 23 Accountant General .0 .0 .0 .0 .0 1.1 27 Registrar of Political Parties .0 .0 .0 .0 .0 .0 30 Presidents Office & The Cabinet .2 .3 .4 .0 3.6 10.4 26& 31 Vice President .0 .0 1.8 2.4 5.9 8.3 32 Civil Ser% icc .2 .I 1.4 .4 3.9 6.4 33 Ethics Secretariat .0 .0 .0 .0 3 Frcien Aflairs .0 .I ,I .0 0 .0 35 Permanent Comtn. of Enquiry .0 .0 .0 .0 .0 .0 36 Cis il Serxvice Committee .0 .0 .0 .0 .0 .0 37 Prime's Ofice 5.0 3.5 17.4 .1 34 44 40 Judiciars .0 .0 .0 .0 .0 .8 4 1 Justice .0 .2 .3 .2 .0 .0 42 Speakers Office .0 .1 .5 .0 .0 .5 45 Exchequer & Audit Depanment .0 .0 .0 .0 .0 .6 50t Finance 35.4 56.2 83.3 17.4 6.4 9.8 Sl Hone AffIairs .7 .0 .3 .0 .0 .0 34 Radio Tanzania .5 .0 .8 .0 .4 .3 55 Investment Promotion .0 .0 .0 .0 .0 .0 56 Regional Adminisstration and Local Gov. .0 .0 .0 .0 17.4 27.1 57 Defence & National 2.7 1.9 1.0 1.0 1.0 1.0 sg Law Refortn Comrn. .0 .0 .0 .0 .0 60 Industries Court of Tanzania .0 .0 .0 .0 .0 61 Electoral Commission .0 .0 .0 .8 .8 63 Local Go% t Servece Cotmm .0 .0 .0 .0 .0 64 Information .3 .0 .0 .0 .0 66 Planning Conmm. 1.7 1.4 7.4 1.5 1.3 2.5 Sub Total 46.8 63.8 114.7 23.8 44.2 72.0 DEFENCE AND SECt'RITY 29 Police Force .8 .9 4 .0 .0 .0 29 Prison Services .4 .5 .2 .0 .0 .0 38 Defence .0 .0 .0 .0 .0 .0 39 National Service .0 .0 .0 .0 .0 .0 Sub Total 1.3 1.3 .6 .0 .11 .0 SOCIAL SERV'ICES 46 Education 9.0 4.1 7.8 10.9 16.7 22.1 49 Water 10.3 1.3 7.6 30.7 30.8 29.6 52 1-lealth 8.7 4.0 26.0 21.5 23.2 34.2 53 Comm. Dev & W 1.0 2.4 3.3 2.4 2.6 1.6 65 Labor Youth 2.4 2.4 4.2 1.7 2.1 3.1 67 Teachers' Service Comm .0 .0 .0 .0 0 0 68 Science. Tech 5A4 2.8 5.2 4.7 5.1 4.2 70 & 89 Regions 19.4 10.9 21.7 29.2 29.2 37.0 Sub Total 56.2 27.9 75.8 101.0 109.7 131,8 ECONOMIC SERV'ICES 47 \Works & Trans 14.8 8.4 45.6 51.1 35.8 31.7 48 Lands. Hiousing .9 .7 1.1 .0 .5 .9 58 Energv and Minerals .1 5.6 24.1 23.8 21.2 35.4 62 Coi11. & Transport 5.4 4.0 27.0 9.2 10.0 22.4 Sub Total 21.2 18.6 97.7 84.2 67.5 90.4 PRODUlCTIV'E 24 & 43 Agr. & Livestock 1.1 3.1 5.3 23.3 28.1 16.7 44 Industries & Trade 1.3 .7 .4 .3 1.2 .4 69 Touristn. Nat. 2.2 1.9 6.1 7.7 3.7 5.5 Sub Total 4.6 5.7 11.8 31.2 33.0 22.6 CONSOLIDATED FUND SERN'ICE 20 State House .0 .0 .0 .0 .0 .0 22 Public Service .0 .0 .0 .0 .0 .0 Sub Total .0 .0 .0 .0 .0 .0 GRAND TOTAL 130.1 117.4 300.5 240.2 254.3 316. Note: Data for FYI0a11 v otes are from the Flash Reports Source: MoF. Appropriations Accounts and Expenditure Flash Report 96 Table 6c: SECTORAL - TOTAL EXPENDITURE (TShs. Billion) VOVOT OTE HOLDER 1995/96 1996/97 1997/98 | 1998/99 t 1999/00 | 2000/01 BUDGET BUDGET BUDGET BUDGET BUDGET BUDGET ADMINISTRATION 23 Accountant General .0 .0 .0 .0 .0 23.6 27 Registrar of Political Parties .0 .0 3.4 2.6 2.7 2,8 30 Presidents Office 6.9 9.8 10.9 15.7 23.2 35.0 26 & 31 Vice President .2 .5 2.7 3.3 7.0 9.7 32 Civil Service Dept. \.8 2.1 3.8 4.0 7.3 10.7 33 Ethics Secretariat .0 .1 .2 .2 .2 .2 34 Foreign Affairs 11.8 10.6 i3.3 18.3 17.1 20.5 35 Perm. Comm. Enq. .1 .1 .2 .3 2 36 Civil Scr% ice Comm. .1 .1 .1 .2 .2 .5 37 Prime Ministers Office 9.9 7.3 26.5 14.2 8.1 9.3 40 Judiciarn 3.0 3.6 4.6 5.4 7.9 9.6 41 Justice .6 1.0 1.2 2.4 1.1 1.5 42 Office Of The Speaker 2.4 3.3 4.0 5.3 5.9 7.1 45 Exchequer and Audit .4 .5 .7 .8 1.1 2.3 50 Finance 57.6 76.7 116.8 68.5 56.1 61.1 5 I Home Affairs 2.1 1.6 2.3 4.6 5.8 5.6 54 Radio Tanzania .8 .3 1.1 1.2 1.6 1.9 55 Investment Promotion .2 .1 .7 .6 .2 .0 56 Regional Adminisstration and Local .. .. ,. 6.2 22.7 44.6 57 Defence & National. 3.9 3.5 3.4 5.5 3.3 5.1 59 Lawv Reform. Commi. .I .I .1 .1 .1 .2 60 Industrial Court Of Tz. .1 .1 .1 .1 .2 .3 61 Electoral Commision 8.6 .9 1.4 3.8 17.0 30.9 63 Local Govt. Servi. Comm. .2 1 .1 .2 .2 .3 64 Information and Broad. .9 .. .. 1.4 .0 .0 66 Planning Commission 2.4 2.0 8.3 4.1 3.2 4.8 Sub Total 114.0 124.4 206.1 169.0 192.5 264.0 DEFENCE AND SECUiRITV' 28 Police Force 18.7 24.6 29.1 29.7 34.9 42.0 29 Prison Services 11.5 12.7 174 16.4 19.4 20.9 38 Defence 46.9 63.8 72.5 73.2 81.7 89.2 39 National Service 5.7 7.5 9.5 11.4 12.1 12.5 Sub Total 82.7 108.7 128.5 130.8 148.1 164.6 SOCIAL SERVICES 46 Education 19.4 16.2 24.8 30.4 37.7 50.8 49 Water. 12.3 2.8 10.4 37.1 34.2 35.5 52 Health 22.1 22.8 51.7 58.7 57.0 74.3 53 Comm. Dev.Wome. Aff. 1.8 3.3 4.7 4.1 4.9 3.6 65 LabourYouthDevelop. 3.7 3.6 5.7 3.7 5.1 6.2 67 Teachers Service Comm. .1 .1 .1 .1 .2 1.5 68 Science. Tech.& H Ed 22.7 24.2 28.1 37.5 37.4 49.1 70 & 89 Regions 104.7 110.7 140.0 150.1 197.8 243.5 Sub Total 186.7 183.7 265.6 321.8 374.3 464.6 ECONOMIIC SERV'ICES 47 Works. 36.5 10.7 70.0 56.1 73.8 87.7 48 Lands. Hous..& Urb. Dev. 2.2 1.3 2.8 3.4 4.1 5.2 58 Energy and Minerals. .3 6.1 25.1 25.8 24.2 38,5 62 Comm.& Transport 6.0 6.3 32.7 22.6 20.7 32.8 Sub Total 45.0 24.3 130.6 107.9 122.8 164.2 PRODUCTIV'E 24 & 43 Agriculture & Livestock 12.9 15.0 19.2 42.6 40.4 28.8 44 Industries and Trade 2.9 2.0 3.3 5.0 4 8 4.5 69 Tour.. Nat. Res. & Env. 7.8 4.2 8.9 15.6 12.5 19.0 Sub Total 23.6 21.3 31.4 63.1 57.7 52.3 CONSOLIDATED FUND SERVICE 20 State house .4 .9 .8 .9 1.0 1.1 22 Public Debt 171.7 217.7 207.1 238.0 294.4 287.5 Sub Total 172.1 218.6 207.9 238.9 295.3 288.6 GRAND TOTAL 624.2 680.9 970.1 1031.5 1190.7 1398.3 Note: Data for FYO1-all votes are from the Flash Reports Source: MoF. Appropriations Accounts and Expenditure Flash Report 97 TABLE 7a: ACTUAL RECURRENT EXPENDITURE BY VOTE AS THE SHARE OF TOTAL RECURRENT EXPENDITURE, FY96-FYOI VOTE V'OTE HOLDER 1995196 | 1996/97 | 1997/98 | 1998/99 | 1999/00 | 2000/01 ADMINISTRATION 27.0 Registrar of Political Parties 0.0 0.0 0.5 0.3 0.2 0.3 30.0 Presidents Office 1.6 1.7 1.5 2.0 2.2 2.4 31.0 2nd Vice President 0.0 0.1 0.1 0.1 0.1 0.1 32.0 Civil Senrice Dept. 0.3 0.3 0.4 0.5 0.4 0.4 33.0 Ethics Secretariat 0.0 0.0 0.0 0.0 0.0 0.0 34.0 Forcien Affairs 3.1 2.1 2.2 2.5 1.9 2.0 35.0 Perm. Comm. Enq. 0.0 0.0 0.0 0.0 0.0 0.0 36.0 Civil Service Comil. 0.0 0.0 0.0 0.0 0.0 0.1 37.0 Prime Ministers Office 1.1 0.7 0.8 0.8 0.5 0.5 40.0 Judiciarv 0.6 0.6 0.7 0.7 0.7 0.9 41.0 Justice 0.1 0.1 0.1 0.2 0.1 0.1 42.0 Office Of The Speaker 0.6 0.5 0.5 0.7 0.7 0.6 45.0 Exchequer and Audit 0.1 0.1 0.1 0.1 0.1 0.2 i0.0 Finance 3.3 3.4 4.1 6.6 5.5 4.0 51.0 Home Affairs 0.3 0.3 0.3 0.6 0.5 0.5 54.0 Radio Tanzaniia 0.1 0.0 0.0 0.1 0.1 0.1 55 ( Investmcnt Promotion 0.0 0.0 0.1 0.1 0.0 0.0 56.0 Reeional Adminisstration and Local Gov. .. .. .. 0.4 1.9 0.4 57.0 Def'ence & National. 0.3 0.3 0.2 0.4 0.3 0.4 59.( Law Reform. Comimii. 0.0 0.0 0.0 0.0 0.0 0.0 60.0 Industrial Court Of Tz. 0.0 0.0 0.0 0.0 0.0 0.0 61 0 Electoral Coimmisioni 4.9 0.2 0.2 0.4 0.7 3.0 63.0 ILocal Gov t. Servi. Comm. 0.0 0.0 0.0 0.0 0.0 0.0 64.0 Inlforilationi and Broad. 0.1 ., .. .. ., 0.0 66.0 Plannhig Commission 0.1 0.1 0.1 0.3 0.2 0.2 Sub Total 16.8 10.7 12.3 16.9 16.1 16.3 DEFENCE AND SECItRITY 28.0 Police Force 4.3 4.2 3.7 3.8 3.7 4.1 29.0 Prison Services 2.5 2.1 2.2 2.1 1.8 2.0 38.0 Defence 11.1 11.1 9.8 .. 8.9 8.7 39.0 National Service 1.3 1.3 1.3 1.5 1.2 1.2 Sub Total 19.1 18.7 17.0 16.9 15.6 16.1 SOCIAL SERVICES 46.0 Education 2.5 2.1 2.7 2.5 2.3 2.8 49.0 Water. 0.4 0.3 0.4 0.8 0.4 0.5 52.0 I-lealthi 2.7 2.8 3.8 4.8 3.6 3.9 53.0 Comm. Dev. Worme. Af'f 0.2 0.2 0.2 0.2 0.2 0.2 65.0 Labour Y'outh Develop. 0.3 0.2 0.2 0.3 (1.2 0.3 67.0 Teachers Scrv ice Comm. 0.0 0.0 0.0 0.0 0.0 0.1 68.0 Science. Tech.& H Ed 3.9 3.4 3.1 4.2 3.5 4.4 70 & 89 Regions 18.4 17.3 17.2 16.5 17.9 19.9 Sub Total 28.4 26.2 27.6 29.3 28.1 32.1 ECON'OMIIC SERVICES 47.0 WN'orks. 1.2 4.1 2.6 0.8 3.3 4R.0 Lands. Illous..& Urb. Dcv. 0.3 0.1 0.1 0.4 0.4 0.4 58.0 Enerv and Nlinerals. 0.( 0.1 0.1 0.2 0.3 0.4 62.0 Comnin.& Transport 0.1 0.4 0.5 1.7 1.3 1.0 Sub Total 1.6 4.6 3.3 3.1 5.4 7.3 PRODI'CTIV'E 43.(0 Agriculture & Livestock 2.4 2.1 1.8 2.3 0.8 1.1 44.0 Industries and Trade 0.5 0.2 0.4 0.6 0.4 0.4 (9.( Tour.. Nat. Res. & Fitn. 1.3 0.4 0.3 1.0 0.8 1.1 Sub Total 4.1 2.7 2.6 4.0 2.1 2.7 CONSOLIDATED FlND SERV'ICE 20.0 State house 0.1 0.1 0.1 0.1 0.1 0.1 22.0 Public Debt 29.8 36.9 37.1 29.7 32.5 25.5 Sub Total 29.9 37.0 37.2 29.9 32.6 25.6 GR.AND TOTAL 100.0 100.0 100.0 100.0 100.0 100. Souree: Table 6a. 6b. and 6c 98 TABLE 7b: ACTUAL DEVELOPMENT EXPENDITURE BY VOTE AS THE SHARE OF TOTAL DEVELOPMENT EXPENDITURE, FY96-FYOI V'OTE VOTE HOLDER 1995/96 11996/97 11997/98 11998/99 119990 20/0 DEV'. DEV. DEV. DEV. DEV. DEV. ADNIINISTRATION' 27 0 Registrar ofPolitical Parties 0 0 0.0 0.0 0 0 0 0 0 0 30 0 Presidents Office 0.8 0.5 0 1 0.0 3 4 8 2 31 0 2nid V'ice Plresident 0 0 0 0 0.3 0.7 1 5 0 7 32 0 Civil Service Dept 0 2 0 2 0.2 0 0 7.8 9 5 330( Ethics Secretariat 0.0 0000 0.0 0 0 0 0 34 0 Foreign Affairs 0 0 0 0 0 0 0.0 0 6 0 00 35 0 Penn. Commi. Enq 0 0 0.0 0 0 0 0 0.0 00( 36 0 Civ-il Service Comm 0 0 0 0 0.0 0 0 0.0 0.0 3 70 Primne Mlinisters Off-ice 4 3 8 5 7 3 0.1 2.8 0 6 400( Judiciary 0 0 0 0 0.0 0 0 0 0 1.3 41.0 Justice 0.0 0.5 0.0 0.0 0.0 0 0 42 0 Office Of The Speaker 0 1 0 2 0 2 0.0 0 0 0.0 45 0 Exchequer and Audit 0 0 0 0 0 0 0 0 0.0 0.1 10 0 Finance 3 5 37.7 90 00 I S 1.4 51I0 lomne Af'fairs 0 6 0 0 0 0 0.0 0.0 0 0 54 0 Radio Tanizaniia 2 7 0.0 0 9 0 0 0.3 0 5 55 0 Ilnxestmenti Promnotion 0 0 0 0 0 0 0 0 0 0 0 0 -56 0 Reizional Adininisstration and Local Gov 0 0 00 00 0.0 00 04 57 0 Defence & National 6 4 0.2 4 3 0 0 00 1.6 59 0 La%\ Reform. Commi 00 0.0 00 00 00 0.0 60 0 lIndustrial Court Of Tz. 0 0 0.0 0 0 0 0 0 0 0.0 61 0 Electoral Comninision 0 0 0 0 0.0 0.0 0.0 63 0 Local Gov t. Serv'i. Commii 0.0 0 0 0.0 0.0 0 0 0.0 64 0 Inifon-nation and Broad. 0 6 0.0 0 0 0.0 0 0 66 0 Planning Commnission 2 1 1.2 1 7 0.6 1 4 3 4 Sub Total 21.4 49.0 24.2 1.4 18.8 27.6 DEFENCE AND SECURITY 28 0 Police Force 1.0 0 7 0 2 0 0 0.0 0 0 29 0 Prison Services 0.3 0 2 0 0 0 0 00( 0 0 38 0 Defence 0.0 0 0 0 0 0 0 0 0 0 0 39.0 National Serv ice 0 0 0 0 0 0 0 0 0 0 0 0 Sub Total 1.3 0.9 0.2 0.0 0.0 0.0 SOCIAL SERV'ICES 46 0 Educationi 4 6 8 7 6 9 9 3 10 8 3 2 49.0 Water, 0.6 0.5 0 6 17.3 8.l 3.2 52 0 Hlealthi 0.6 0 9 6 3 lOS 9 6 39.9 53 0 Commi. Dev.\Vome Aff 0 5 0 0 0.0 1 4 1.3 2.3 65.0 Labour Youith Develop. 15.6 6.5 I 4 0.3 0 6 0 1 67(1 Teaclhers Service Commi 00 00 0 0 00 0 0 6 6 6S 0 Sciencc. Techi& H Ed 2 5 4 2 0 9 1 9 IS5 10 3 70 & 8.9 Reczions 13.0 SI1 6 5 8 9 0 0 10 3 Sub Total 37.5i 26.0 2 2. 5 49.6 31. 67.2 ECONOMIIC SERVICES 47 0 Works 22 9 8 3 0 0 1 73 19 4 4.1 48 0 L.ands. IlIOUs.& Urb De\ 0 2 00 0.0 0 0 0 0 00 58 0 Enerp\ anid Mlinerals 0 0 3.2 7 7 14.6 9 3 0 2 62 0 Commn &- Transport 0 0 10.2 32.4 8 3 9 4 0.0 Sub Total 23.1 21.7 40.1 40.2 38.1 4.3 PRODUCTIVE .130 Agriculture &Livestock 5903 6 1 8 6 II I 0 44 ( Industries and Trade 2 9 0.1 0 3 0 1 0.3 00 69)0 T'our.Nat Res & Eniv 7 9 2 I 6 5 0.1 0 0 0.0 Sub Total 16.7 2.5 13.0 8.8 11.3 0.9 CON'SOLIDATED FLUND SERV'ICE 20 0 State house o.o 0 0 0 0 0.0, 0 0 0 0 22 0 Public Debt oo0 0.0 0 0 0.0 0 0 0(1 Sub Total o.o 0.0 0.0 0.0 0.0 0.(I GRAND TOTAL 100.0 100.0 100.0 100.0 100.0 100.01 Source: Table 6a. 6b. and 6c 99 TABLE 7b: ACTUAL TOTAL EXPENDITURE BY VOTE AS THE SHARE OF TOTAL EXPENDITURE, FY96-FYOI VOTE VOTE HOLDER 1995/96 1996/97 1997/98 1998/99 1999/00 2000/01 TOTAL TOTAL TOTAL TOTAL TOTAL TOTAL ADMINISTRATION 27 0 Registrar of Political Parties 0.0 0.0 04 0 3 0 1 0.3 30 0 Presidents Office 1.5 16 1.4 1.7 2.3 28 310 2nd Vice President 00 0 1 0 1 02 0 3 0 2 32 0 Civil Service Dept 0 3 0.3 0 3 0 4 1.2 1.0 33 0 Ethics Secretariat 0 0 0.0 0 0 0.0 0 0 00 34 0 Foreign Aflairs 30 2.0 20 2 2 1 7 1.8 350 Perm Comm, Enq 00 00 00 00 0.0 0.0 36 0 Ci% il Service Comm 0 0 0 0 0 0 0 0 0 0 00 37 0 Prime Ministers Office 1.3 1 2 I 5 0 7 0 8 0.5 40 0 Judiciary 0.6 0 6 0 6 0 6 0.6 0.9 410 Justice 0.1 0.2 0.1 02 0 1 0 1 42 0 Office OJ'TheSpeaker 0 5 0.5 05 0.6 0 6 06 450 Exchequer and Audit 0.1 0.1 01 01 0.1 0.2 50 0 Finance 3 3 5 6 4 7 5 8 5.1 3.8 51 1 Home Affairs 0.3 03 02 05 0.5 05 54 0 Radio Tanzania 0 2 0 0 0 1 0.1 0.1 0.1 55 0 Investment Promotion 0 0 0 0 0 1 0.1 0 0 0.0 56 0 Regional Adminisstration and Local Gov. . .. 0.3 1.7 0.4 57 0 Defence & National 0 5 0 3 0 7 0.3 0 2 0 5 59.0 Law Reform. Commi 0.0 00 00 0.0 00 0.0 60 0 Industrial Cour Of Tz 0 0 0 0 0 0 0.0 0.0 0.0 61 0 Electoral Commision 4.7 02 02 .. 06 2.8 63.0 Local Govt. Servi. Comm 0 0 00 00 00 00 0.0 64 0 Information and Broad. 0.1 66 0 Planning Commission 0 2 0 2 0 3 0.4 0.3 0.4 Sub Total 17.0 13.1 13.6 15.0 16.4 16.9 DEFENCE AND SECURITN' 280 Police Force 4 1 3 9 3 3 3 3 3 3 3 9 29.0 Prison Services 24 2.0 20 1 8 1.6 1.9 38 0 Defence 10.7 10.4 8 7 7.9 8.2 390 National Service 1 2 1 2 1 2 1.3 1 1 1.2 Sub Total 18.4 17.6 15.1 14.8 14.0 15.1 SOCIAL SERV'ICES 46 0 Education 2 6 2 5 3 2 3.3 3 3 2 8 49 0 Water. 0 4 0.3 0 4 2 9 1 2 0.8 i2 0 llealth 2 7 2.7 41 5.5 4 3 6 0 53 0 Comm Dev.Wome. Aff 0.2 0.1 0 2 0 4 0.3 0.3 65 0 Lahour Youth Develop. 0 9 0 6 0 3 0.3 0 2 0 3 67.0 Teachers Service Comm 0 0 0 0 0 0 0.0 0 0 0.1 68.0 Science. Tech.& H Ed 3 9 3 4 2 9 3 9 3 3 4 5 70 & 89 Regions 18.2 165 160 156 160 193 Sub Total 28.8 26.2 27.1 31.8 28.5 34.1 ECONOMIC SERVICES 470 Works 2 1 43 23 29 50 54 481) Lands. llous.& Urb De% 03 01 01 0.3 04 04 58 0 Encrev' and Minerals 00 03 09 2.0 1 3 0 3 62 0 Comm & Transport 0 1 1 0 4 0 2.5 2 2 1.0 Sub Total 2.5 5.7 7.3 7.7 8.9 7.1 PRODI'CTIV'E 43 0 Agriculture & Livestock 25 2 0 23 3 1 1.9 i 1 44 0 Industrics and Trade 0 6 0 2 0 4 0 5 0 4 0 4 69 0 Tour. Nat Res & En% 1 6 0 5 1 0 0.9 0 8 I I Sub Total 4.6 2.7 3.7 4.6 3.1 2.6 CONSOLIDATED FUND SERV'ICE 200 State house 0 1 0 1 0 1 0 1 0 1 0 1 22 0 Public Debt 28 6 34 6 33 1 26 1 29 0 240 Sub Total 28.7 34.7 33.2 26.2 29.1 24.1 GRAND TOTAL 100.0 100.0 100.0 100.0 100. 100.0 Source: Table 6a. 6b. and 6c 100 TABLE 8a: ACTUAL RECURRENT EXPENDITURE AS A PERCENTAGE OF BUDGETED EXPENDITURE BY VOTE, FY95-FYO1 \OTE VOTE HOLDER 1995(961 1996(97 | 1997(98 | 1998/997 1999/00 | 2000101 ADNIINIISTRATION 270 Registrar of Political Parties 100.1 100.4 49.6 988 30 0 Presidents Ol'fice 99 6 100.0 100.0 99 0 98.7 100.4 .10 2nd Vice President 90.8 99.9 90.5 99.8 1000 1000 320 Civil Service Dept 880 1006 101.3 100.0 996 99.9 33 0 Ethics Secretariat 100.3 98.6 97.8 82 8 99 8 34 0 Foreign Affairs 111 9 114.1 114 6 105.7 97 0 97 4 3;5 0 Pernn Comm. Enq 93.1 100.0 100.1 99.3 95 2 118.1 360 Civil Service Comm 83.4 98.4 111.4 99.5 999 1000 37 0 Prime Nlinisters Office 99 9 100.2 63 0 42.7 100.0 103 1 40 0 Judiciary 89.2 100 1 98 2 98.8 75 3 996 41 0 Justice 76.9 999 105.7 89.7 858 103 1 42.0 Office OfTlle Speaker 101 8 99.4 990 1008 998 99.9 450 Exchequer and Audit 971 993 98.6 97 1 100.0 988 50 ( Finance 63 8 96 1 84.3 101.5 97.6 79 5 :1 0 liome Aftairs 85.9 983 85 1 101.4 82.7 93.1 540 Radio Tanzania 79.1 100.2 998 933 95.6 75.1 55 0 Investmenit Promotion 66.0 96.2 99.9 98 1 92 7 56 0 Regional Adminisstration and Local Gov. 44.4 314.9 22 1 57 0 Defence & National 96.3 99 9 70.8 68.2 100 0 101.5 59 0 Law Reform. Commi. 61.2 102 8 101 5 99 2 94 4 99.6 600 Industrial Court OfTz. 35.0 99 8 100.5 997 88 1 1000 61 0 Electoral Commision 241 9 99.5 106.1 99 5 36.5 992 63.0 Local Govt Servi. Comm 58 5 99 8 99 9 99.8 990 95 9 64 0 Information and Broad. 71.0 660 PlanningCommission 700 100.1 96.7 96.1 70.3 1000 Sub Total ' 106.0 101.1 91.5 91.0 95.9 86.8 DEFENCE AND SECURITY' 280 PoliceForce 101.0 1000 870 999 922 998 290 Prison Services 95.4 1000 87.9 99.9 83.4 99.7 3S 0 Defence 100.4 99 5 92 0 1010 96 1 999 390 National Service 94 6 100.0 93.7 103.2 90 4 1000 Sub Total 99.5 99.7 90.5 100.8 93.0 99.9 SOCIAL SERVICES 46 0 Education 99 8 99.3 109.7 99.7 98 5 99 7 490 Water. 81 5 100.0 90.1 100.1 100.0 906 520 Health 87.1 84 4 100.4 99 7 94.9 99.6 53 0 Comm Dev Wome Aff. 91.7 99.9 100.0 102 7 62 4 100 0 650 I.aboLir Youth Develop. 96 0 99 9 100.4 101 9 51 3 94 3 67 0 Teachers Service Comm. 75.9 99 8 98 0 99 8 99.7 59.2 68 0 Science. Tech.& H Ed 96.4 911 92 2 99 9 94.7 99 8 70 & S9 Regions 91 3 99 1 99.2 106.7 93 8 98.4 Sub Total 92.1 96.3 99.3 103.6 93.8 98.5 ECONO.MIC SERV'ICES 470 \'orks 23.9 10042 726 1278 764 1001 48 0 Lands. -lous .& Urb Dcv. 93.8 72.9 32 4 80.1 99 8 97 0 580 Enerev and Ninerals. 406 1000 90.5 98.3 969 1193 62 0 Comm.& Transport 61 7 100 0 63.3 97.0 III 3 100 0 Sub Total 28.8 462.4 69.4 101.1 85.8 100.7 PRODt'C'rlT'E 43 0 Agricuilture & Livestock 84 8 99 2 895 94 7 60 2 95 8 44 0 Industries and Trade 119 8 98 8 99 9 98 8 92 6 100 7 690 Tour. Nat Res & Env 97.6 99 3 82 2 100.8 850 86 2 Sub Total 91.6 99.2 90.0 96.8 73.7 92.1 CONSOLIDATED FU:ND SERV'ICE 200 Statehouse 1029 963 1000 1000 1000 999 0 Public Debt 73 4 96 9 122.3 97 5 97 4 90.8 Sub Total 73.5 96.9 122.2 97.6 97.4 90.9 GRAND TOTAL 85.7 101.5 101.9 98.7 94.1 94.6 101 TABLE 8a: ACTUAL DEVELOPMENT EXPENDITURE AS A PERCENTAGE OF BUDGETED EXPENDITURE BY VOTE, FY95-FYO N'OTE N'OTE HOLDER 1995/96 | 1996/97 | 1997/98 | 1998/99 |1999/00 ADMINISTRATION 27.0 Registrar of Political Parties 30 0 Prcsidents Office 72.9 63 8 26.3 100.0 31 0 2nd Vice President 14.8 33 6 27.0 32 0 CiN il Service Dept 23 2 85 8 12 3 0 3 211.5 33.0 Ethics Secretariat 34 0 Foreign Affairs 0 0 0 0 35 0 Pern Comm Enq. 36 0 Civil Ser% icc Comm 37 0 Prime Ministers Otlice 15 7 93 5 349 1000 87.5 40 0 Judiciary 0.0 0° 0 00 41 0 Justice 25.4 95 2 7 8 2 0 42 0 Off-ice OFlThe Speaker 100 0 68 6 36.9 45 0 Exchequer and Audit ( Rl Financc 1 8 25 6 8 9 0.0 25 5 ;1 ( Ilome Affairs 14.8 00 54.0 Radio Tanzaniia 100 0 100.0 68 3 55 0 Investment Promotioni 56 0 Regional Adminisstration and Local Gov 0.3 570 Defenice & National 42.9 42 3549 0.0 00 59 0 Lawv Reformn Commi. 60 0 Industrial Court Of Tz. 61.0 Electoral Commision 0.0 63 0 Local Gos t Servi. Comm. 64.0 Intfonnation and Broad 35 6 660 Planniig Commissioni 21 9 31 8 196 41 S 120.3 Sub Total 8.4 29.2 17.5 6.4 45.3 DEFENCE AND SECURITY 28 0 Police Force 22 8 32 8 54.7 29 0 Prisoni Services 11.3 11 7 0 0 38 0 Deftence 0 0 39 0 National Service Sub Total 18.7 25.1 34.5 SOCIAL SERVICES 46 0 Education 9 5 80 8 73.2 94.5 68.7 490 Water. 1 2 154 65 62.3 27 8 520 I-lealth 1 4 88 20.1 540 440 530 Comm Dcv Wome Atf 9 1 0.0 0.0 65.0 51.1 65 0 ILabotir Youth Develop. 121.0 103 8 26 7 20.6 30 3 67 0 Teachers Service Comm. o8 0 Science. Tech.& H Ed 8.4 55.6 14 0 43.6 30 9 70 & 89 Regions 12.3 17 8 24.8 33.7 0.0 Sub Total 12.2 35.4 24.6 54.2 30.8 ECONO\IIC SERVICES 47 0 \orks 28.4 38 0 0.0 37 4 57.6 48 (1 Lands. Hous .& Urb. Dev. 3.5 00 0.0 00 5S(1 EnerEv and Minerals. 2.0 21 7 26 5 67.5 46 5 62 0 Comm & Transport 0.0 97 3 99.6 100.0 100.0 Sub Total 20.0 44.4 34.0 52.7 60.0 PRODUCTIV'E 43 1) Agriculture & Livestock 1000 3 3 95.0 40 8 41 9 44 0 Industries and Trade 41 5 5.1 79.4 366 24 3 690 Tour.Nat.Res.&EnN 647 41 7 889 1.5 00 Sub Total 66.6 16.5 91.4 31.1 36.6 CONSOLIDATED FUND SERV'ICE 20.0 State house 22 0 Public Debt Sub Total GRAND TOTAL 14.1 32.4 27.6 45.9 41.8 102 Table 9a: RECURRENT BUDGETED EXPENDITURE - REGIONS, FY96 (TShs. Billion) VOTE rOTE HOLDEJ 1995/96 1996/97 1997/98 1998/99 1999/00| 2000/01 | I BUDGE BUDGE BUDGE BUDGET BUDGE 1BUDGET 70 Arusha 5.5 6.3 7.6 7.7 12.2 15.1 71 Coast 3.1 3.1 3.7 3.8 5.9 7.1 72 Dodoma 4.5 5.2 6.1 6.1 8.9 10.5 73 Iringa 5.2 5.6 6.6 6.8 7.3 11.4 74 Kigoma 3.4 3.7 4.4 4.7 5.1 7.9 75 Kilimanjaro 5.6 7.4 8.6 8.8 12.7 15.2 76 Lindi 2.7 3.0 3.5 3.6 5.1 6.4 77 Mara 4.6 5.0 6.1 6.1 8.6 11.0 78 Mbeya 5.5 6.3 8.0 8.2 11.5 13.4 79 Morogoro 4.9 5.6 6.7 6.8 9.9 11.5 80 Mtwara 3.3 3.7 4.3 4.4 6.4 7.7 81 Mwanza 7.1 6.9 8.4 8.4 12.3 13.9 82 Ruvuma 3.8 4.4 5.2 5.2 7.6 8.8 83 Shinyanga 5.0 5.4 6.4 6.6 7.3 11.3 84 Singida 3.1 3.6 4.1 4.2 5.9 7.0 85 Tabora 3.8 4.4 5.1 5.3 7.1 8.4 86 Tanga 5.2 5.9 7.0 7.1 10.1 12.0 87 Kagera .6 5.2 6.2 6.3 9.0 10.8 88 D'Salaam 5.5 5.6 6.4 6.7 10.0 10.6 89 Rukwa 2.9 3.2 3.8 4.0 5.6 6.3 Total 85.3 99.4 118.3 120.9 168.6 206.5 Note:Data for FY 01 are from the Expenditure Flash Report. Source: MoF, Appropriations Accounts and Expenditure Flash Reports 103 Table 9b: BUDGETED DEVELOPMENT EXPENDITURE - REGIONS, FY (TShs. Billion) VOTE IVOTE HOLDERI 1995/96 1 1996/97 1997/98 j 1998/99 | 1999/00 | 2000/01 70 Arusha 1.0 1.6 2.9 3.5 3.5 3.4 71 Coast .4 .5 .8 .5 .5 .6 72 Dodoma 1.2 .8 .5 2.4 2.4 2.6 73 Iringa 1.7 .9 1.8 2.8 2.8 5.0 74 Kigoma 1.1 .5 .3 1.4 1.4 .2 75 Kilimanjaro .6 .2 .2 .3 .3 .4 76 Lindi 1.6 .3 .8 .4 .4 1.3 77 Mara .5 .5 .7 1.7 1.7 3.5 78 Mbeya .8 .4 .4 .4 .4 .7 79 Morogoro .5 .3 1.7 .6 .6 3.2 80 Mtwara .2 .5 .8 .6 .6 1.3 81 Mwanza .4 .8 1.3 1.8 1.8 2.2 82 Ruvuma .6 .2 .3 .4 .4 .5 83 Shinyanga .5 .3 1.6 .3 .3 .3 84 Singida .5 .4 .5 .4 .4 .9 85 Tabora .3 .2 .2 .2 .2 .3 86 Tanga 1.5 .8 1.4 1.5 1.5 .9 87 Kagera 2.8 .7 2.2 5.6 5.6 9.3 88 D'Salaam 1.9 .7 3.1 4.0 4.0 .2 89 Rukwa 1.5 .4 .3 .3 .3 .3 Total 19.4 10.9 21.7 29.2 29.2 37.0 Note: Data for FY 01 are from the Expenditure Flash Report. Source: MoF, Appropriations Accounts and Expenditure Flash Reports 104 Table 9c: BUDGETED TOTAL EXPENDITURE - REGIONS (TShs. Billion) VOTE |VOTE HOLDER| 1995/96 | 1996/97 | 1997/981 1998/991 1999/00 | 2000/01 70 Arusha 6.4 7.9 10.5 11.2 15.7 18.6 71 Coast 3.5 3.6 4.5 4.3 6.4 7.7 72 Dodoma 5.8 6.0 6.6 8.5 11.4 13.1 73 Iringa 6.9 6.4 8.5 9.6 10.1 16.4 74 Kigoma 4.5 4.2 4.7 6.1 6.5 8.1 75 Kilimanjaro 6.1 7.6 8.8 9.0 13.0 15.6 76 Lindi 4.3 3.3 4.3 4.0 5.5 7.7 77 Mara 5.0 5.5 6.8 7.9 10.3 14.5 78 Mbeya 6.3 6.7 8.4 8.7 11.9 14.1 79 Morogoro 5.4 5.9 8.4 7.4 10.6 14.7 80 Mtwara 3.5 4.2 5.1 5.0 7.0 9.0 81 Mwanza 7.4 7.7 9.7 10.2 14.0 16.1 82 Ruvuma 4.4 4.6 5.5 5.6 8.0 9.3 83 Shinyanga 5.5 5.7 8.0 6.9 7.6 11.6 84 Singida 3.6 4.0 4.6 4.6 6.3 7.9 85 Tabora 4.1 4.6 5.3 5.6 7.4 8.8 86 Tanga 6.7 6.7 8.4 8.5 11.6 12.9 87 Kagera 3.4 5.8 8.4 11.9 14.6 20.1 88 D'Salaam 7.4 6.3 9.5 10.7 14.0 10.7 89 Rukwa 4.4 3.7 4.2 4.3 5.9 6.5 Total 104.7 110.3 140.0 150.1 197.8 243.5 Note: Data for FY 01 are from the Expenditure Flash Report. Source: MoF, Appropriations Accounts and Expenditure Flash Reports 105 Table 10a: ACTUAL RECURRENT EXPENDITURE AS PERCENTAGE OF BUDGETED RECURRENT EXPENDITURE - REGIONS, FY96-FY01 VOTE VOTE HOLDER 1995/96 1996/97 1997/98 |i998/99 1999/00 2000/01 70.0 Arusha 91.2% 100.0% 93.3% 108.2% 95.8% 99.9% 71.0 Coast 80.3% 99.6% 97.8% 108.5% 96.6% 94A4% 72.0 Dodoma 93.1% 99.9% 99.2% 106.0% 99.7% 99.9% 73.0 Iringa 78.5% 95.,0% 98.5% 108.3% 31.9% 99.6% 74.0 Kigoma 87.4% 99.8% 98.8% 104.3% 124.4% 100.0% 75.0 Kilimanjaro 110.3% 99.9% 99.4% 105.7% 98.8% 100.0% 76.0 Lindi 45.9% 99.9% 99.1% 108.9% 98.6% 84.9% 77.0 Mara 94.1% 99.6% 97.6% 109.6% 99.7% 99.4% 78.0 Mbeya 97.5% 99.5% 99.3% 103.9% 99.0% 98.4% 79.0 _ Morogoro 93.4% 99.8% 98.9% 109.8% 99.3% 99.6% 80.0 Mtwara 91.5% 99.6% 98.7% 104.7% 99.9% 98.6% 81.0 _ Mwanza 90.4% 101.8% 69.5% 108.5% 99.9% 100.0 % 82.0 Ruvuma 96.0% 100.0% 97.5% 106.4% 99.1% 100.0% 83.0 Shinyanga 91.2% 100.3% 91.9% 111.5% 131.9% 100.0% 84.0 Singida 92.6% 99.6% 99.2% 105.9% 98.6% 99.3% 85.0 Tabora 94.8% 98.8% 99.3% 102.1% 99.2% 99.8% 86.0 Tanga 92.3% 99.9% 98.2% 106.3% 99.3% 99.6% 87.0 Kagera 105.4% 100.3% 98.8% 109.0% 98.9% 84.0% 88.0 D'Salaam 97.6% 100.0% 98.7% 100.7% 27.1% 107.9% 89.0 Rukwa 88.4% 99.6% 92.1% 103.7% 97.8% 94.3% _Total 91.3% 99.7% 95.7%1 106.7%/ 93.8% 98.4% Note: Data for FY 01 are from the lxpcnditure Flash Report. Source: MoF, Appropriations Accounts and Expenditure Flash Reports 106 Table 1OB: ACTUAL DEVELOPMENT EXPENDITURE AS PERCENTAGE OF BUDGETED DEVELOPMENT EXPENDITURE - REGIONS, FY96-FYO1 VOTE VOTE HOLDER 1995/96 1996/97 1997/98 1998/99 1999100 2000/01 70 Arusha 1.3% 0.6% 4.3% 3.0% 0.6% 14.3% 71 Coast 92.8% 17.8% 19.0% 63.8% 56.3% 49.3% 72 Dodoma 1.1% 1.3% 23.1% 4.4% 0.2% 10.5% 73 Iriniga 1.3% 2.3% 5.0% 47.1% 82.7% 7.0% 74 Kigoma 21.1% -13.9% 37.9% 0.0% _ 0.4% 100.0% 75 Kilimanjaro 0.5% 17.1%' 58.5% 0.0% 2.0% 100.0% 76 Lindi 0.8% -12.5% 13.3% 25.4% 2.4% 22.3% 77 Mara 51.8% 65.7% 191.6% 149.3% 85.1% 0.0% 78 Mbeya 13.0% 13.8% 25.3% 23.4% 9.1% 71.1% 79 Morogoro 15.3% 57.2% 17.6% 17.1% 0.0% 9.5% 80 Mtwara 6.7% 4.3% 12.2% 17.6% 0.8% 21.7% 81 Mwanza 42.1% 72.7% 56.4% 45.5% 37.0% 18.1% 82 Ruvuma 23.8% 66.9% 34.2% 58.0% 41.4% 58.6% 83 Shinyaniga 2.6% 9.7% 7.1% 0.0% 0.0% 100.0% 84 Singida 2.4% 4.9% 21.2% 25.0% 1.2% 23.3% 85 Tabora 4.3% 7.8% 40.9% 44.3% 2.1% 100.0% 86 Tanga 0.8% 2.6% 6.9% 7.2% 61.9% 100.0% 87 Kagera 8.3% 17.9% 55.5% 61.5% 66.9% 3.6% 88 D'Salaam 1.5% 2.8% 2.4% 1.8% 0.1% 100.2% 89 Rukwa 46.4% 39.8% 45.6% 36.2% 1.7% 100.0% Total 12.3% 17.8% 24.8% 33.7% 33.1% 17.8% Note: Data for FY 01 are from the Expenditure Flash Report. Source: Mol, Appropriations Accounts and Expenditure Flash Reports 107 Table IOC: ACTUAL TOTAL EXPENDITURE AS PERCENTAGE OF BUDGETED TOTAL EXPENDITURE - REGIONS, FY96-FY01 VOTE VOTE HOLDER 1995/96 1996/97 1997/98 1998/99 1999/00 2000/01 70 Arusha 77.7% 79.5% / , 69.0% 75.4% _, 74.5% 84.2% 71 Coast 81.8% 88.4% 84.2% 103.2% 93.5% 90.9% 72 Dodoma 73.7% 86.9% 93.4% 77.2% 78.6% 82.4% 73 Iringa 59.4% 82.5% 78.4% 90.4% 46.0% _ 71. 1% 74 Kigoma 71.9% 90.1% 95.5% 80.0% 97.5% 100.0% 75 Kilimanjaro 100.3% 97.6% 98.5% 102.7% 97.0% 100.0% 76 Lindi 29.2% 91.2% 83.9% 100.3% 91.4% 74.4% 77 Mara 90.1% 96.7% 107.7% 118.3% 97.2% 75.6% 78 Mbeya 87.1% 94.8% 95.7% 99.7% 95.6% 97.0% 798 Morogoro 85.8% 97.3% - 82.6% 102.1% 93.5% 80.2% so Mtwara 86.8% 89.0% 85.3% 94.4% 91.5% 87.7% 81 Mwanza 87.8% 98.8% 67.8% 97.7% 92.0% 88.8% 82 Ruvuma 86.5% 98.5% 94.2% 103.0% 96.2% 97.9% 83 Shinyanga 82.8% 96.2% 74.9% 106.0% 126.0% 100.0% 84_ _ Singida 79.4% 89.9% 91.1% 98.6% 92.1% 91. 1% 85 Tabora 88.3% 94.0% 96.6% 99.6% 96.1% 99.8% 86 Tanga 71.6% 88.6%i 82.50%- 89.3-%i 94.6 %- , , 99.6i%e 87 Kagera 25.6% 90.8% 87.5% 86.5% 86.5% 46.6% 88 D'Salaani 73.4% 89.1% 67.7% 64.0% 19.4% 107.7% 89 Rukwa 74.4% 92.3% 88.3% 99.1% 93.0% 94.6% _____ Total 76.793% 91.6%/c 84.7% 92.5% 84.9% 86:2% Note: Data for FY 01 are from the Expenditure Flash Report. Source: MoF, Appropriations Accounts and Expenditure Flash Reports 108 Table 11: SECTORAL - RECURRENT EXPENDITURE BEFORE AND AFTER REALLOCATIONS, FY00 (TSHS. Mill BUDGET AFTER "| PROVED BUDGET FTR ACTUAL EltPENDITURE AS N'OTE \VOTE HOLDER APPROVED BUDOET REALLOCATION ET ACTUAL EXPENDITURE REALLOCATION ASA OE %OF APPROVED BUDGET IOT APPROVED BUDGET AFTER REALLOCAInON ADMINISTRATION 23 Accountant General .. .. .. 27 Registrar of Political Parties 2.68 .. 2.68 1.33 100% 50% 30 Presidents Office 14.41 5 17 19.58 19.32 136% 990 26 & 31 Vice President .55 13 .68 .68 124% 100% ,2 Civil Service Dept. 2.48 .92 3.40 3.39 137% 100% 33 Ethics Secretariat .16 .01 .17 .13 107% . 77% 34 Foreign Affairs 11.32 2 58 ' 13.89 16.61 123% 120% 35 Perm Comm. Enq 20 01 .21 .20 107% 95% 36 Civil Service Comm. 15 04 19 19 126% 100% 25 & 37 Prime Ministers 03icc 2.51 2 19 4.70 4 70 187% 100% 40 Judiciar\ 7 95 . 7.95 5.95 100% 75% 41 Justice 1 02 .05 1 07 .92 105% 86% 42 Office Of The Speaker 5.22 .68 5.90 5.89 113% 100% 45 Exchequer and Audit 1.01 06 1.07 1.07 106% 100% 50 Finance 128.41 4 54 61.00 48.54 48% 80% 51 Home Affairs 5.36 .49 5 85 4.83 109% 83% 54 Radio Tanzania 1.18 .03 1.21 1.08 102% 89% 55 iivestment Promotion .19 00 19 .18 101% 97% 56 NIORALG 1743 .. 17.43 17.62 100% 101% 57 Defence & National. 2.05 .29 233 2.33 114% 100% 59 Law Reform. Commi. .12 .. 12 .11 100% 94% 60 Industrial Court OfTz. 17 . 17 15 100% 88% 61 Electoral Commision 16 21 .03 16 24 5.93 100% 36% 63 Local Govt Servi. Comm 19 .01 20 20 103% 99% 64 Information and Broad. 66 Planning Commission 1 60 32 1.92 1 35 120% 70% Sub Total 222.56 17.55 168.16 142.71 76% 85% DEFENCE AND SECUiRIT)' 28 Police Force 34.90 .. 34.90 32.20 100% 92% 29 Prison Services 19.42 . 19.42 16.19 100% 83% 38 Defence 77.10 6.12 83.22 78.96 108% 95% 39 National Service 10.15 1.96 12.10 10 95 119% 90% Sub Total 141.57 8.08 149.65 138.30 106% 92% SOCIAL SERVICES 46 Education 18.02 2.95 20.97 20.66 116% 98% 49 Water. 2.80 .58 3.39 3.39 121% 100% 52 Health 31.58 2.19 33 78 32.01 107% 95% 53 Comm Dev Wome. AIT. 2.18 .15 2.33 1.45 107% 62% 65 Labour Youth Develop. 2.77 .21 2.98 1.53 108% 51% 67 Teachers Service Comm 20 03 .24 24 116% 100% 68 Science. Tech & H Ed 30.92 1 58 32 51 31.04 105% 96% 70 - 89 Regions 134.09 41.10 175 19 172.89 131% 99% Sub Total 222.57 48.81 271.38 263.21 122% 97% ECONOMIIC SERVICES 47 W'orks . 38 00 .. 38.00 4.62 100% 12O 48 Lands, Hous & Urb. Dev. 355 .09 3.65 3.64 103% 100% 58 Energy and Minerals. 2.23 .76 2 99 2 89 134% 97% 62 Comm &Transport 1066 1.21 11.87 11.86 111% 100% Sub Total 54.45 2.06 56.50 23.01 104% 41% PRODUCTIVE 43 Agriculture & Livestock 15.72 -3.39 12.33 7.42 78% 60% 44 Industries and Trade 2.80 84 3 64 3.37 130% 93% 69 Tour.. Nat. Res & Enm 8.31 43 8 73 7.43 105% 85% Sub Total 26.82 24.70 18.21 92% 74% CFS 20 State house .94 .05 " .99 .99 105% 100% 22 Public Debt 265.36 29.00 " 294.36 286 69 111% 97% Sub Total 266.30 .. 295.34 287.68 111% . 97% GRAND TOTAL Note: ' Retention Scheme Funds (Approved Estimates) Req. no. TYIG/86/111100 - Special Requisition Source: MoF. Appropriations Accounts and Expenditure Flash Reports for vote 38, 68, 73 & 75. 109 Table 12: IREGIONS - RECURRENT EXPENDITURE REALLOCATIONS, FY00 (TShs. Billion) BUDGET AFTER RE T A ACTUAL EXPENDITURE AS A VOTE VOTE HOLDER APPROVED BUDGET REALLOCATION UDGE AFTER ACTUAL PERCENTAGE OF BUOGET REALLOCATION PERCENTAGE OIF APPROVEELLCAIO BUDGET ATRRALCTO 70 Arusha 9.3 2.9 12.2 11.7 131% 96% 71 Coast 4.9 .9 5.9 5.7 119% 97% 72 Dodomna 6.5 2.4 8.9 8.9 137% 100% 73 Iringa .. 74 Kigoma 5.1 1.3 6.4 6.4 126% 99% 75 Kilimanjaro 9.7 3.0 12.7 12.6 131% 99% 76 Lindi 4.2 .9 5.1 5.1 123% 99% 77 Mara 6.4 2.2 8.6 8.6 135% 100% 78 Mbeya 8.8 2.6 11.5 11.4 130% 99% 79 Morogoro 7.6 2.3 9.9 9.9 131% 990/0 80 Mtwara 5.1 1.3 6.4 6.4 126% 100% 81 Mwanza 9.1 3.2 12.3 12.2 135% 100% 82 Ruvuma 6.0 1.6 7.6 7.6 126% 990/0 83 Shinyanga 7.3 2.7 10.0 9.6 137% 96% 84 Singida 4.8 1.1 5.9 5.8 123% 990% 85 Tabora 5.5 1.6 7.1 7.1 129% 990/0 86 Tanga 7.9 2.2 10.1 10.1 129% 990/0 87 Kagera 6.9 2.1 9.0 8.9 130% 99% 88 D'Salaam 7.5 2.5 10.0 2.7 134% 27% 89 Rulkwa 4.1 1.5 5.6 5.5 137% 98% Total 126.8 38.6 165.3 155.8 130% 94% Source: MoF, Appropriations Accounts 110