Report No. 12601-TA Tanzania Role of Government Public Expenditure Review (In Two Volumes) Volume I Main Report June 17, 1994 Country Operations Division Eastern Africa Department Africa Region FOR OFFICIAL USE ONLY MI CROGRAPH I CS Report No: 12601 TA Type: ECO Document of the Wordo Dak This document has a restricted distributionf and may be used by recipients onlty In the perfbonance, of their official duties. Its content may not otherwise be.disclosed without World Bank authorization GOVERNMENT ISCAL YEAR July I - June 30 CURRENCY EQUIVALENTS ,urrency Unit: Tanzanian Shilling (T Sh) Official Rate: US$1.00 = T Sh 499 (March 1994) ACRONYMS ANI) ABBREVIATIONS ADB African Development Bank AIDS Acquired Immune Deficiency Syndrome C&CL Contractual and Contingent Liabilities CFS Consolidated Fund Services CG Consultative Group CSR Civil Service Reform EEC European Economic Community EEB European Investment Bank ERP Economic Recovery Program GDP Gross Domestic Product GNP Gross National Product HESAWA Health, Sanitation and Water HIV Human Inmunodeficiency Virus IDA International Development Association IMF International Monetary Fund IPC Investment Promotion Center LART Loans and Advances Realization Trust LIC Low Income Countries MALD Ministry of Agriculture and Livestock Development MIC Middle Income Countries MIT Ministry of Industries and Trade MOCD Ministry of Community Development, Women's Affairs and Children MOF Ministry of Finance MOL Ministry of Labour and Youth Development MS Ministerial Supply NACP National AIDS Control Program NAFCO National Agriculture and Food Corporation NaTCAP National Technical Cooperation and Assistance Program NBC National Bank of Commerce NGO Non-Government Organization NMC National Milling Corporation NORAD Norwegian Agency for Development Cooperation OC Other Charges ODA Official Development Assistance O&M Operation and Maintenance OTTU Organization of Tanznian Trade Unions PC Planning Commission PE Personal Emoluments FOR OFFICIAL USE ONLY PER Public Expenditure Review PIP Public Investment Program PMO Office of the Prime Minister and First Vice President PSRC Parastatal Sector Reform Commission RPFB Rolling Plan and Forward Budget RS Regional Supply S&G Subventions and Grants SGR Strategic Grain Reserve SIDA Swedish International Development Authority SPA Special Program of Assistan^e for Africa SS Social Sectors SSA Sub-Saharan Africa STD Sexually-Transmitted Disease TA Technical Assistance TFC Tanzania Fertilizer Corporation TFR Total Fertility Rate UK-ODA UK Overseas Development Administration UNDP United Nations Development Program UPE Universal Primary Education URT United Republic of Tanzania VAT Value-Added Tax WDR World Development Report Tbis document has a sticted distnbution and may be used by rocipients only in the prfoman of their official dutes. Its content may not otherwise be disclosed without World Bank authorization. Volume I Contents Preface ................. Executive Summary .... ........................................... ii. Chapter 1: Introduction ............................................ 1 Background .......................................... 1 Progress Since the 1989 PER .......................... I Nature of the Problem ............. ...................... 3 Chapter 2: Role of Government ....................................... 6 Macroeconomic Foundation ........................... 7 Composition of Public Expenditures ..................... 10 Chapter 3: Economic Composition of Public Expenditures ..................... 11 Recurrent Versus Development Expenditures .................... 12 B ,akdown of Recurrent Expenditures ........................ 14 Subventions and Grants ............................ 15 Contractual and Contingent Liabilities ................... 16 Personal Emoluments and Other Charges .................. 19 Civil Service Reform ............. ...................... 20 Civil Service Employment ........................... 20 Civil Service Pay Reform ........................... a Summary of Recommendations ........ ..................... 25 Chapter 4: Spending Priorities in Selected Sectors .......................... 27 A. Social Sector Spending .......... ...................... 27 Education .... .......................... 28 Role of Govermnent in Education .28 Trends in Education Sector Spending .29 Intrasectoral Issues in Education .30 Role of Donors in Education .34 Need for Reforn in Education .35 Healdth .. 36 Role of Government in Health .36 Trends in Health Sector Spending .37 Intrasectoral Issues in Health .38 Role of Donors in Health .41 Need for Reform in Health .42 Other Social Sector Spending ............................. 43 Recommendations for Reform ........................ 43 B. Spending in the Economic Sectors ......................... 44 Agriculture ............ ............................. 45 Recommendations for Reform in Agriculture ............... 47 Industry and Trade .................................... 48 Recommendations for Reform in Industry ................. 48 Infrastructure ........................................ 48 C. Expenditures on Public Administration ...................... 49 Recommendations for Reform in Administration .... ......... 51 Chapter 5: Efficiency and Transparency of Aid-Funded Expenditures .............. 54 Public Investment Programn ............................... 54 Recommendations for Reform ................... ... 55 Budgeting of Aid-Funded Expenditures ........................ 59 Recommendations for Reform of the Aid Budgeting System ... ... 62 Accounting of Aid-Funded Expenditures ....................... 63 Recommendations for Reform of the Aid Accounting System .... . 64 Selected Bibliography ............................................ 66 Text Tables Table 1.1: Basic Economnic and Social Indicators, 1970-1990 .................... 4 Table 1.2: International Comparison of Basic Indicators, 1990 ................... 5 Table 2.1: Size of Central Govermnent Operations, 1991-94 .................... 8 Table 3.1: Breakdown of Total Expenditures, 1992-94 ........................ 11 Table 3.2: Recurrent Cost Implications of Development Expenditures .............. 13 Table 3.3: Economic Composition of Ministerial Supply Expenditures, 1992-94 .15 Table 3.4: Major Subventions and Grants, 1992-94 .15 Table 3.5: Major Contractual and Contingent Liabilities, 1992-94 .17 Table 3.6: Personal Emoluments and Other Charges, 1992-94 .20 Table 3.7: Civil Service Employment, 1961-1988 .21 Table 3.8: Personnel-Related Expenditures, 1992-94 .23 Table 4.1: Trends in Social Sector Spending, 1990-93 .28 Table 4.2: Trends in Education Sector Spending, 1990-93 .29 Table 4.3: Summa / of Total Education Sector Spending, 1993/94 .30 Table 4.4: Expenditure Allocation Across Inputs in Education, 1991/92 .33 Table 4.5: Spending On Education by Program, 1993/94 .35 Table 4.6: Trends in Health Sector Spending, 1990-93 .37 Table 4.7: Summary of Total Health Sector Spending, 1993/94 .38 Table 4.8: Recommended and Actual Per Capita Health Expenditures, 1993/94 .40 Table 4.9: Spending On Health by Program, 1993/94 .42 Table 4.10: Breakdown of Ministerial Expenditures in the Economic Sectors .... ...... 44 Table 4.11: Selected Agricultural Projects, 1993/94 ......................... 47 Table 4.12: Breakdown of Ministerial Expenditures on Public Administration, 1993/94 . .. 50 Table 5.1: Composition of External Assistance by Type, 1991 .................. 60 Table 5.2: Budgeting for Aid-Funded Project Expenditures, 1993/94 ............... 61 Text Figures Figure 4.1: Shares of Different Levels of Education, 1993/94 ................... 31 Figure 4.2: Public Expenditure per Pupil, 1993/94 .......................... 32 Figure 4.3: Shares of Different Levels of Health Services, 1993/94 ................ 39 Figure 4.4: International Cornparison of Ministerial Expenditures ................. 51 Preface This Public Expenditure Reiew (PER) for Tanzania reviews progress made in public expenditure nanagement since the last PER in 1989 and focuses on examining the implications for public expenditures of the Government's decision to redefine its role in the economy. The Govermnent aims to focus its ^carce resources on its core functions of providing law and order, and social and physical infrastructure, with the objective of improving the efficiency and equity of public ex.penditures in Tanzania. The Report is organized in two volumes. Volume I contains the Executive Summary and the main report consisting of five chapters: Chapter 1 provides an introduction to the PER; Chapter 2 discusses the role of goverment; Chapter 3 analyzes the economic composition of public expenditures; Chapter 4 examines sectoral issues; and Chapter 5 looks at issues relating to aid-funded expenditures. Volume II contains the Statistical Annex. ThJs report is a joint effort of the Government of Tanzana and a team f multiateral agencies and bilateral donors consisting of the V,;orld Bank, EC, NORAD, SIDA and UK-ODA, that vsited Tanzania from September 6-22, 1993. The joint-donor mission consisted oft Nisha Agrawal (mussion leader and princinal author, World Bank); Aeran Lee (World Bank); Jens Claussen (NORAD); Torsten Wetterblad (SIDA); David Pedley (UK-ODA); P. Ahren (consultant, NORAD); H. Bierman (consultant, EC); K Kiragu (consultant, UK-ODA); J. Nioroge (consultant, SIDA); and M. Ravicz (consultant, World Bank). Te mission was assisted by an inter-ministerial team from the Government that was coordinated by P.B. Rweyemamu (Planning Commission), and consisted of M.R. Rubunga, J. Lema, A. Mwaisumo, AM. Mwabeza, F.S. Kiongosya, 0. Mdeme, M.P. Kessy, A.M. Manyama, J. Mwilima, and C. Sonyi. Ihe Report was written under the supervision of Michael Carter, Division Chief, AF2CO. In addtion, several people from the World Bank provided guidance, includig: Peter Miovic (Lead Economist, Eastern Africa Department); Wlltiam Shaw (Country Economist, Tanzania); Charles Griffin (Econonist, Social Sectors); Mike Stevens (Lead Advisor); and Vinaya Swaroop (Peer Reviewer). Vedasto Rwechungura (Resident Mission, Tanzania) also provided assistance with putting together the report. Finaly, Kathryn Rivera (AF2CO) was responsiblefor the word processing and the physical production of the report. This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authoriation. Executive Summary Introduction 1. This Public Expenditure Review (PER) has found that the main public expenditure problem in Tanzania is that the effectiveness of Government expenditures is extremely low. The Government is involved in a wide array of activities and spreads its scarce resources too thinly over these activities. As a result, it underfunds all of them. This spreading thin of Government resources results in expenditures that have low productivity and quality, and have little impact on poverty reduction. The Govermnent has recognized this problem and has announced its decision to withdraw from a mnmber of activities and to focus on its core functions, such as providing law and order, basic health and basic education, and infrastructure. The focus of this PER is therefore on examining the implications for public expenditures of the Government's decision to redefine its role. The Goverrnent's objective is to increase the effectiveness of public expenditures in improving the living standards of the Tanzanian people, especially of the very poor. While the PER looked at all the sectors of the economy, it has a special emphasis on the social seciors, since these sectors were found to suffer most from a scarcity of resources and to be in need of urgent attention. The PER identifies areas of expenditures from which a substantial amount of resources could be saved during FY95 and FY96. Given that the needs of the infrastructure sector have largely been addressed by the US$ 2 billion invested in this sector over the last five years, and the further US$ one billion of investments that is planned over the next two-three years, the PER argues that the entire amount of potential savings identified should be allocated to the social sectors in the course of the next three Budget (FY95-FY97). Strategy to Improve the Allocation and Management of Public Expenditures X. Role of Government 2. The objective of the Tanzanian Government is to accelerate growth with a view to reducing poverty. There is ample evidence available from recent empirical research which indicates that what the Government needs to do to achieve tlese two goals is to provide the right macroeconomic environment for private sector development and to focus public expenditures on economic and social infrastructure. The rest it should leave to the private sector, where it should only play a supportive role of building the required social, physical, administrative, regulatory, and legal infrastructure of good quality. 3. These broadly stated functions of Government need to be translated into public expenditure priorities. The 1989 PER had found that the inability to define priorities and focus expenditures had resulted in chronic underfunding of most programs, with program managers typically receiving only one-third of the resources they needed to operate effectively. This problem is just as prevalent today and needs to be addressed urgently. The process of defining priorities for public expenditures in Tanzania should be facilitated by the adoption of the new three-year planning and budgeting framework, the "Rolling Plan and Forward Budgetd (RPFB) from 1993/94 onwards. The first RPFB has broadly articulated the role and functions of Government in each sector. As part of the preparations of the second RPFB, these need to be prioritizedfurther as planned so that they can faciltate the focussing of Govermnent expenditures on priority areas, beginning with the 1994/95 Buget. iii 4. Once the core Covernment functions have been defined more precisely through the RPFB, the process of pulling the Govermment out of non-core activities will need to be implemented. This is necessitated by the severe mismatch between government revenues and expenditures that is causing the current instability in the macroeconomic enviromnent in Tanzania (see Section 11 below). Given an already high aid dependency, and the large burden of servicing debt (both internal and external), the Govermment has no option but to increase revenues and to severely curtail expenditures. The latter will be difficult given the pressing needs for additional funds, especially in the social sectors. Given also the severe underfunding of most activities, the only way for expenditures to be reduced further is by reducing the number of activ-ties the Government is involved in, instead of only trying to cut the costs of existing activities. 5. This process has already been initiated in the Ministry of Agriculture and Livestock Development (MALD). MALD is now in the process of being reorganized in line with its new role and functions. The Government plans to undertake similar functional and organizational reviews of all its ministries in the next three years, starting with six ministries in 1994. The functions and structure of the local government administration also needs to be reviewed in parallel. Given the degree of underfunding of core activities, through these reviews the Government will need to pull back from the non-core activities that it is currently involved in, both at the Central and Local govermment levels. These functional and organizational reviews for central and local government strctures therefore need to be initiated with some urgency so that they can be completed within the three year perod of the second Roling Plan and Forward Budget, i.e., during the period 1994/95-1996/97. 6. There are two main objectives of these reviews: firstly, to improve the efficiency of govermnent organizations and consequently, the management of the economy; and secondly, to identify potential areas of savings, including from the retrenchment of personnel (see Section IV below on civil service reform). Wim the objective of improving the management of the economy, the three key central ministries, i.e., the Ministry of Finance, the Planning Commission and the Civil Service Department, should be included in the first year's program. To attain the second objective, the other three organizations to be reviewed in the first year should include those that have a large number of employees such as, at the central level, the ministries of Works, Home Affairs, Education, and Health. In addition, the staffing levels and management of local governments, in particular of the education and health employees that fall under their jurisdiction, needs to be reviewed urgently. II. Macroeconomic Stability 7. The key to laying the foundation for macroeconomic stability is fiscal prudence. While considerable progress was made in this area in the first six years following the launch of the Economic Recovery Program (ERP) in 1986, the fiscal situation has deteriorated precipitously in the last two years. The fiscal deficit (after grants) swung from a surplus of 2.3 percent of GDP in 1991/92 to a deficit of 8.1 percent of GDP a year later. During 1992/93, revenues fell by almost 7 percent of GDP compared to the previous year. The 1989 PER had found that there was widespread tax evasion, particularly in the payment of import duties, and extensive tax avoidance through the granting of excessive exemptions and "deferments" to taxpayers. These tax administration problems persist today and have combined with poor tax policies to create the fragile macro environment. 8. The Government is concerned about its inability to attain macroeconomic stability since the launch of the ERP, and to bring the rate of inflation below 20 percent per annum, let alone to single digit levels. It recognizes that such a reduction in inflation is essential for getting an adequate supply iv response from the private sector. Therefore, in response to the continued deterioration of the fiscal situation during the first half of 1993/94, in January 1994 it adopted a number of measures to improve the situation. In addition, the Government has recently requested technical assistance from the IMF in the area of strengthening overall tax administration, with special emphasis on customs administration. The recommended measures to strengthen ,ax and customs administrc tion are expected to be introduced, together with improved tax policy measures, in the 1994/95 Budget. 9. In frami.g future budgets, the objective of the Government is to fund all recurrent expenditures and an increasing proportion of development expenditures with its own revenues. This is critical, both in order to raise the domestic savings rate in the economy and also to reduce the Governmeit's dependency on foreign aid. To achieve this objective, however, expenditures will need to be curtailed severely and tax revenues wvill need to be raised considerably. m. Economic Composition of Expenditures 10. An important reason for the low effectiveness of Government expenditures in Tanzania is the inadequate funding of critical expenditures on Other Charges (OC), in particular those required for operations and mairtmnance (O&M). The 1989 PER had found that neglect of maintenance had led to a severe deterio1 .on of Government assets, in particular infrastructure, which was leading to crippling bottlenec.. Maintenance had been neglected for so long that the capital invested in infrastructure was being lost. While this problem has been corrected somewhat in the infrastructure sector, it persists uncorrected in other sectors, particularly in the social sectors. The neglect of maintenance is also reflected in the Development Budget where the bulk of expenditures consist of rehabilitation of assets necessitated by the lack of routine maintenance. 11. There are three main reasons for the inadequate funding of critical expenditures on Other Charges: the predominant one is the general underfunding of all activities due to an over-extended expenditure base; the second one is the squeezing out of non-personnel expenditures by personnel- related expenditures; and the final one is the squeezing out of critical expenditures in the Other Charges category (such as on textbooks and drugs) by non-critical expenditures (such as on student welfare and Government vehicles). 12. The first of the three reasons has been discussed above and needs no further elaboration. On the second point, data indicate that expenditures on personal emoluments grew by 50 percent in nominal terms between FY92 and FY93 and are budgeted to grow by another 40 percent between FY93 and FY94. With an inflation rate of about 23 percent in FY93 and with a similar outcome likely in FY94, this would lead to a real increase in wages and salaries of about 27 percent in FY93 and 17 percent in FY94. In contrast, the nomiin-l amount budgeted for Other Charges stagnated over the two year period, implying a real reduction of 23 percent per annum in both FY93 and FY94. Given how acute the degree of underfunding is for critical materials and O&M, this is likely to lead to a further sharp deterioration in the quality of services provided by the Government. There is thus a strong need to control fuure increases in the wage bill so that expendires on selected materials and O&M can be increased in real terms over the next few years. Finally, the third area, which deals with the misallocation of resources within the category of Other Charges, has been the subject of detailed investigation recently by a Presidential "Cost-Cutting Commission " that was set up in 1993 to identify areas of potential savings in expenditures. The Commission's Report is due shortly and will recommend, inter alia, ways of curbing wasteful expenditures on Other Charges. It is hoped that the recommendations of Presidential Commission's Report will have an impact on the formulation of the 1994/95 Budget. v IV. Civil Service Refonn 13. While there is a need to control the size of the wage bill in order to release resources for critical materials and O&M, there is also a recognized need to improve the remuneration of the civil service. The 1989 PER had found that the civil service was grossly overstaffed and underpaid. In addition, the ci-il servants were poorly equipped to do their jobs because of the lack of complementary inputs. This had resulted in a civil service that had poor morale, weak motivation, widespread absenteeism, and growing corruption. All of these problems persist today, and if anything, have become more pronounced. 14. With 355,000 civil servants, the civil service is too large to be paid and equipped ad.,-.ttely. The Government has recognized this problem and has launched a civil service reform program, a key element of which is the retrenchmnent of 50,000 civil servants over a three year period, starting with 10,000 in 1992/93 (which was accomplished) and with 20,000 each in the subsequent two years. This retrencl';nent prograrn, however, will need to be intensified and accelerated in order to shrirk the size of the e.vil service to one that Tatzania can afford to pay adequately. Under the current program, while some caeegories of civil servants (identified to be redundant) are being retrenched, other categories continue to be hired. The Government has exempted from retrenchment personnel in education, health, and security since they are thought to be in short supply. These categories of personnel account for almost 50 percent of the civil service and continue to be hired at an estimated rate of about 10,000 per annum (the number that is being turned out from Government training institutions for these personnel). Thus, at the end of three years, there will only be a net retrenchment of about 20,000 civil servants and the civil service will still have about 335,000 civil servants. 15. A civil service consisting of 335,000 civil servants will still be too large for the Government to meet its objective of attaining a more efficient and adequately compensated civil service. Given that at the most the Govermnent can only fund about two-thirds of all activities that it is currently involved in, two-thirds of the existing civil service (about 225,000 personnel) would be the largest feasible size to target for. This is also the size of the civil service that would exist today if in the three decades since independence in 1961, the civil service (of 90,000) had grown only at the rate of growth of population (about 3 percent per annum) instead of its actual growth of 4.5 percent per annum. The retrenchment exercise to be carried out in the future would need to be carefully linked to the functional reviews that are being planned for all ministries and for the local governments as described above. To complete the retrenchment exercise within a reasonable time frame, the Government should revise its targets upwards to retrench at least 40,000 instead of 20,000 civil servants on a net basis per year for the next three years. It also needs to include in the retrenchment exercise all groups that have so far been exempt from it. While the fiscal costs of doing so will be substantial, they are affordable (at about T Sh 6 billion per annum). The exercise could be completed within the next few years while Tanzania is still receiving substantial budgetary support from donors (in the form of counterpart funds from balance of payments support) to complete its adjustment agenda. 16. The Government also needs to initiate the process of pay reform for civil servants. The 1989 PER found that during the 1970s and 1980s, there had been a severe erosion in the real wages and salaries of civil servants as a result of which average salaries in the late 1980s provided only one-fifth the purchasing power of the 1970s. The Government has tried to arrest further declines in civil service pay by the recent increases in basic salaries and by the use of various monetary and non- monetary allowances to supplement basic salaries. The pledtora of allowances, however, has led to a remuneration structure that is non-transparent, inequitable, and unmanageable. It is dear hat a vi substana enhancement of saaries wiUl need to await a drastic reduction in the :ize of the civil service. However, the Government needs to take Imnmediate action to design a transparent and eqitable remnneration package that could be introduced in the 1995/96 Rudget. Such a package would eliminate most of the existing monetay and non-monetary allowances and incorporate the monetzed value of any that remain. V. Public Investment Program 17. The Government needs to develop a public investment program that aims to enhance the efficiency of its project portfolio. One of the reasons why this efficiency is low is because of the large number of projects, which makes it impossible tor the Government to either fund them or manage them adequately. As a result, there are long delays in project completion and huge cost overruns. The large size of capital expenditures relative to recutrent ones also leads t(, a chronic insufficiency of resources to provide for the adequate operation and maintenance of completed investmerts. 18. The PER estimated that at present about one-third of all expenditures in the Deve!opment budget are in areas that are no longer a priority for Government, given its proposed new and more limited role and functions. In the context of the preparation of the first RPFB, an exercise to rationalize Development expenditures has begun but needs to be accelerated. As a first step, all projects that are no longer in priority areas need to be identified and phased out. It is recommended hat only projects inpnority areas would remain in a 'core investment program " that would receive the bulk of the Government's counterpartfunds in the 1994/95 Budget; that the number and cost of such projects be lindted so that the necessary counterpart funds are fully budgeted; and that in the event of a shortfall In the avaiabty of counterpat funds, absolute priory will be given to projects in the "core" PIP. As a second step, as additional information on individual projects becomes available, projects within the priority sectors will need to be screened to weed out the non- performing ones. Since most of these projects are funded by donors, the Government will need considerable cooperation from donors in undertaing this two-step exercise. VI. Social Sector Strtea 19. The social sectors in Tanzania are in a state of crisis. The quality and coverage of primary education has been falling and, as a result, the illiteracy rate is on the upswing. Secondary school education is grossly inadequate and is available to only a tiny 4 percent of the population. The public health situation is alarming: 50 percent of all children suffer from malnutrition and 80 percent of pregnant women are anemic. The recent spread of AIDS is further stretching to the limit the Govermment's financial and managerial resources. Added to all this are the pressures for additional services created by the high (and growing) fertility rate. 20. The 1989 PER had found that basic social services were under tremendous pressure and had warned that earlier achievements in this area (made during the mid-1970s to the mid-1980s) were in danger of being lost. Despite the pressing needs of these sectors for additional resources, the share of the social sectors in total public resources has been declining in recent years. To make matters worse, a large proportion of the limited resources that are available to these sectors is wasted on non- priority expenditures like the funding of costly and inequitable expenses for sending students and patients overseas to receive higher education and medical treatment. What these sectors require, therefore, is not just additional resources, but the more effective use of existing resources. To arrest and mverse the decline in the sociad sectors, what is required is the formulaon and adoption of vii a comprehensive "Sodal Sector Stategy" as soon as possible. This work has been delayed considerably but is now undzrway and needs to be completed with urgency. VII. Allocadon of rublic Resources to the Social Sectors 21. Tanzania needs to increase the amount of public expenditures allocated towards building human capital and to decrease other expenditures. Greater investment in people requires more, a id more effective, expenditures in education, health, nutrition, and family planning. Given the low levels of provision of these services that we are starting from, the demands of these sectors for additional resource are tremendous. The PER reveals that just to provide quality Universal Primary Education (UPE) and basic health care-both stated objectives of the Government-would require additional resources to the tune of T Sh 210 billion (US$ 500 million) annually, amounting to 50 percenw of total Government expenditures or 18 percent of GDP in 1993/94 . If we add this to the existing expenditures on health and education (T Sh 90 billion), the total resources required just for these two areas would amount to over 85 percent of net Government expenditures (net of debt servicing), which is clearly unfeasible. 22. Given the large and pressing needs of the social sectors, several types of measures will need to be taken to attain the Government's objectives in these sectors, even by the year 2000. Firstly, as part of the Social Sector Strategy (SSS), the Government will need to prioitize further even within the critical sectors of basic health and education. It will need to make difficult choices, for example, by asking itself whether it wants to focus first on improving the quality of primary education for those currently enrolled in the system, or whether it wants to spend resources on extending the coverage to those who are currently being denied any education whatsoever, albeit of a poor quality. Given that part of the reason why enrollment rates are low and declining has to do with the poor quality of education being provided, the PER recommends that the first priority in education should be to improve the quality of education. Within health, the Government faces similar difficult choices. Should it first focus only on fuli; funding a preventive health care program and delay the much- needed improvements to the curative health care system, or should it move simultaneously, but more slowly, on both fronts? Again, given the tight budgetary situation, and the higher social returns from preventive care, the PER recommends that the Government first focus on improving the preventive health care program. 23. A second set of measures will be to reallocate expenditures from other sectors to the social sectors. The PER indicates that there is considerable scope for reallocating from other expenditures (including special expenditures, ministerial expenditures and regional expenditures) to the social sectors. The savings in special expenditures, as proposed in the first RPFB, arise from reducing the costs of maintaining the Strategic Grain Reserve, eliminating the fertilizer subsidy, limiting salary increases so that the wage bill is kept constant in real terms, and by reducing financial assistance to parastatals (see Chapter 3 for details). The savings in ministerial expenditures arise from reducing expenditures on public administration by two percent of the projected GDP for FY95 ( i.e., one percent or about T Sh 13.8 billion each from general administration and defence), and by reducing by half expenditures on high cost social welfare programs (see Chapter 4 for details). The savings in regional expenditures arise by reducing the costs of running the regional layer of government in line with their reduced responsibilities (see Chapter 4 for details). 24. While some of these savings cannot be achieved in time for the 1994/95 Budget, substantial amounts of the proposed savings in special expenditures can be undertaken immediately and can be effected in the 1994/95 Budget. It is recommended, therefore, that an addonal amount of at least viii T Sh 20 billion be allocated pnmarily to materials and O&M expenditures for basic social services in the 1994/95 Budget, of which half would be for improWng the quality of primary education and the other half for preventive health. To the extent possible, these funds should be allocated and spent through localgovernments. In addition, the 3udgets for FY96 and FY97 should be restructured so that further savings that can potentially be realized from ministerial and regional expenditures are reallocated to basic social services, with the target being to allocate an additional one percent of GDP each to basic health and basic education in both years. 25. A third set of measures to build human capital entails reforms within the social sectors. In the social sectors, the Government's role is to provide basic health and basic education. As alluded to above, there is ample scope in Tanzania to improve the efficiency and equity of resource allocation within the social sectors. Within education, the Government spends only 50 percent of its budget on primary education; the rest is divided between secondary (13 percent), tertiary (19 percent) and other education. Despite the substantial amount of resources spent on post-primary education, access to such education is limited to a privileged few. This is because these institutions have a high cost structure (teachers typically work half the norm in private institutions) and spend almost half of their budgets on student boarding and welfare costs. Thus, those who are fortunate enough to have access to such education receive not only free education but also have their entire cost of living paid for by the Government. As a result, while the Government spends only US$ 15 per primary school student per annum, it spends more than 10 times as much on each secondary school student (US$ 167), and more than 200 times as much on each university student (US$ 3,500). 26. Reforms are needed urgently in the way that the education sector's resources are allocated across different levels of education and across different inputs. Given that the benefits to society are greatest from investing in primary education, and that these benefits fall with increases in the level of education, Tanzanian policy makers need to examine the tradeoffs inherent in the provision of different levels of education in Tanzania. It needs to be recognized that the cost of sending one student to university in Tanzania is not sending 238 students to primary school. Secondly, policy makers need to recognize that if inputs were combined in a more efficient way, then access to education could be increased substantially even within the existing resource envelope. For example, within the existing resource envelope for seconday educaton, the enrollment rate could be quadrupledfrom the current 4percent to 16perceni (the average for Sub-Saharan Africa) if student welfare expenses were eliminated and if teachers in public schools were required to work additional hours. In university education as well, where a third of the budget is spent on student welfare, and where there are only 3.5 students per teacher, there is clearly ample scope for improvement. This issue of the appropriate intra-sectoral allocation of resources in education needs to be addressed more fully in the SSS. 27. In the health sector as well resources are allocated in an inefficient and inequitable manner. Preventive health care, which is the mr-ost cost-effective use of public resources, receives only 14 percent of the health budget, while the bulk of the health sectors' resources are spent on curative health care provided through hospitals (over 40 percent). If we compare the pattern of health care spending in Tanzania with that recommended by the recent World Development Report (WDR, World Bank, 1993d) on health, we find that about two-thirds of the total health budget in Tanzania is spent on what the WDR calls "essential health care". The WDR has put together a limited package of public health measures and essential clinical interventions that it recommends as a top priority for Government expenditures. The public health package for preventive care includes: immunizations; school-based health services; information and selected services for family planning and nutrition; programs to reduce tobacco and alcohol consumption; regulatory action, information, and limited ix public investments to iinprove the household enviromnent; and AIDS prevention. As compared with the recommended annual spending in these areas of about US$ 4.2 per capita for low income countries, Tanzania spends 0.50 cents per capita, which amounts to only 12 percent of WDR recommnended standards. If the resources that were currently being spent on non-essential health services were redirected towards this public health package, it would be possible to fund preventive health services at up to 40 percent of the recommended level rather than at the current 12 percent. 28. In terms of curative care, the 1993 WDR recommends a cost-effective package of essential clinical services that the Govermnent should focus on providing. This package consists of five groups of interventions that are likely to have the largest impact on reducing morbidity and mortality and include: services to ensure pregnancy-related care; family planning services; tuberculosis control; control of sexually-transmitted diseases; and care for the common serious illnesses for young children, which include diarrheal disease, acute respiratory infection, measles, malaria, and acute malnutrition. In a low income country like Tanzania, this recommended package is estimated to cost US$ 7.8 per capita per annum. Annual expenditures in Tanzania for these five areas, by contrast, are only US$ 1.7 (about 20 percent) per capita. Instead, budgetary expenditures are allocated to services, such as treatment abroad, spending on food for patients in hospitals, and training of medical staff, that should be paid for by the recipients of these services. Again, the issue of the appropriate intra- sectoral allocation of resources in health will need to be addressed more fully in the SSS. 29. This leads us to the fourth set of measures, namely, cost recovery, which will need to be adopted in the social sectors in order to achieve the Government's objective of improving the delivery of basic social services. Given the magnitude of the needs of these sectors, a reallocation of public resources, either from other sectors to the social sectors, or from low priority expenditures to higher priority ones within the social sectors, will not suffice. In order to achieve the goals for the social sectors, scarce public resources will necessarily need to be supplemented with private ones. There is ample scope for doing this. The collection rate of the nominal Universal Primary Education (UPE) levy (T Sh 200 per annum) is only 36 percent; the fee for students in secondary public schools (T Sh 2,500 per annum) is less than half of that charged in private schools; and, university students pay no fee at all. In health as well, the recent introduction of user fees for certain categories of expenditures could be accelerated, while at the same time the Government could promote a health insurance system and the private provision of health care. The issue of cost recovery will also need to be addressed more fully in the Government's SSS. VIH. Strengthing Local Governents 30. After being abolished in 1972, Local govermnents were reintroduced in Tanzania in 1983 and charged with the responsibility of delivering, inter iQa, basic health and education services. However, while their responsibilities increased vis-a-vis those of Regional governments, their resources did not, and as a result their finances remained precarious. The 1989 PER found that Local governments received inadequate transfers from the Central government for effectively carrying out their enhanced responsibilities and that their own revenue base was not well-developed. Administratively as well, Local governments remained weak due to the shortage of qualified staff, compounded by the loss of experience and institutional continuity during the period of their abolition. 31. This continues to be a problem today. During 1992, the Government began preparations of a Cabinet paper proposing a strategy for strengthening Local governments. This work needs to be completed with some urgency. A strategy for strengthening Local govemnments should be adopted during 1994 and should be implemented during 1995. As part of tis strategy, financial resources x and skilled personnel should be transferred from Regional governments to Local governments, where they are in extremely short supply. This would serve to improve the delivery of basic social services. IX. Role of Donors 32. Tanzania receives a large amount of foreign assistance from a large number of donors. In 1991, it received net official development assistance (ODA) to the tune of US$ 1076 million, which amounted to 39 percent of its GDP and to US$ 43 on a per capita basis (World Bank, 1993d). This measure excludes the aid that Tanzania received through the Paris Club in the form of a net reduction in debt service of US$ 243 million in 1991. Together, this amount makes Tanzania one of the largest recipients of aid in the developing world, relative to the size of its economy or its per capita income. 33. The 1989 PER had found that Tanzania's dependence on foreign aid had increased substantially and that the Recurrent budget, in addition to the Development budget, had become critically dependent on aid flows. Donors contributed heavily to Recurrent budget programs, both by directly supplying consumables such as school textbooks and essential drugs, as well as through their balance of payments support and commodity import support programs. The problem of aid dependency has increased since the last PER. In 1993/94, donors will fund over 80 percent of Development expenditures, a quarter of which are actually recurrent in nature and consist of directly funded recurrent items such as textbooks and essential drugs. In addition, they are projected to fund almost one-third of Recurrent expenditures (excluding debt amortization) through loans and grants for balance of payments support. 34. As stated above, the Government of Tanzania aims to reduce its aid dependency in the medium-term and to fund all of its recurrent expenditures and an increasing proportion of its capital expenditures through its own revenues. In the short-term, however, Government efforts to improve revenue collection will need to be supplemented with efforts to improve the efficiency of aid utilization. Towards this end, donors will need to continue to provide support that is recurrent in nature, albeit within a framework that explicitly recognizes the implications of this for the sustainability of Tanzania's expenditures. Currently, Tanzania's capital expenditures are too large relative to its recurrent expenditures for the investments that are taking place to be adequately maintained and operated in the future. As part of its public investment strategy, the Government aims to reduce the size of capital expenditures by closing down non-priority and non-performing projects. Given the needfor additonal resources in the social sectors, and given that the bulk of these needs are recurrent in nature, it is important that donors not continue with their preference for funding capital expenditures only and, in the short run at least, are willng to adopt a more progrmn- oriented approach and fuid slices of "development programs" based on sector strategies, irrespecdve of whether the expenditures required are recurrent or capital in nature. In additon, to improve the efficiency of this aid, the aid should be provided, to the extent possible, in the form of balance of payments support. While the funding of recurrent expenditures raises difficult issues of sustainability of Government expenditures, which will need to be dealt with in the future, it at least begins to tackle the current problem of the low effectiveness of capital expenditures in Tanzania. X. Budgeting and Accoumting of Aid-Fimded Expenditures 35. Tanzania's large aid program is administered by a large number of donors-in 1992/93, 34 donors (15 multilateral and 19 bilateral ones) provided aid to Tanzania. The large aid program, combined with the large number of donors, creates enormous difficulties in aid coordination and xi management for the Government. It also makes it extremely difficult to budget and account for this aid. The 1989 PER had noted that the proliferation of programs financed through aid had been accompanied by an increasing inability to manage and utilize the aid flows efficiently. This problem has worsened with the growth in the size of the aid program. 36. Donor cooperation and assistance will be required for achieving success in the Government's efforts to improve the budgeting and accounting of aid-funded expenditures. Because of a variety of reasons, almost 85 percnt of donor-provided development aid is controlled and disbursed dfrety by donors. This weakens the Government's ability to account for these expenditures. It also makes the Govermnent reluctant to budget for expenditures that it knows it will later be unable to account for. In turn, the lack of budgeting makes it difficult to record these expenditures. The weaknesses in the budgeting and accounting systems feed on each other and create a vicious circle of unbudgeted and unaccounted expenditures that weaken the Government's expenditure management system and reduce its accountability to Parliament. 37. As a result of these problems, until 1992/93 a substantial proportion of the aid provided by donors completely bypassed the Government's budgeting system. Since then, the Governnent has made a substantial effort to improve the coverage of the Budget and it is estimated that about 60 percent of all project-related aid is now on budget. The Government should now budget at least for al project-related aid in the 1994/95 Budget and at the same time it should adopt measures to improve the accounting of such aid. Donors are urged to cooperate in the Government's efforts to improve the transparency and effectiveness of aid-funded expenditures. Condusion 38. The Tanzanian Government faces a severe challenge to improve the living standards of its population. One of the key instruments through which it will attain this goal is to improve the efficiency and equity of public expenditures. The PER reveals the severity of the problems that confront the Government in this area, and the difficult choices that need to be confronted urgently by policymiakers. The PER also lays out a comprehensive strategy for improving the allocation and management of public expenditures. While some of the elements of this strategy can be adopted immediately, others can only be implemented in the medium term. However, Tanzanian policymakers need to recognize that there is a high cost to further delays in restructuring public expenditures, given that Tanzania is falling further behind the rest of the world in human capital investment, and because failure to act risks losing the support of weary donors. 39. The PER highlights the need for a more radical restructuring of public expenditures than what has been proposed so far under the Rolling Plan and Forward Budget. Over the next three years, the Budgets should be formulated keeping in mind the need to: (i) increase expenditures on essential Other Charges, primarily critical inputs and O&M expenditures, by controlling the growth in the wage bill and by reducing non- essential expenditures on Other Charges; (ii) reduce the size of the civil service sharply so that the wage levels for the remaining civil servants can be increased; (iii) reduce the size of the investment program to one that can be adequately operated and maintained in the future by closing down non-priority, non-perforning projects; and xii (iv) increase the allocation of public resources to basic education and basic health. This reallocation of public expenditures needs to be supplemented with various measures to improve their management. xiii Summary of Recommendations for Reform The key components of the Government's strategy to improve public expenditure management are: (i) Role of Government: To deal with the severe problem of underfunding of all Government activities, in the context of the second RPFB define the role of Government more precisely with a view to substantially reducing its size. To pull out of non-core activities, complete organizational reviews of all ministries within the three year period of the second RPFB, i.e., during 1994195-1996/97; (ii) Macroeconomic Stabiiy: To reduce the mismatch between revenues and expenditures, which is the key cause of macro-instability, introduce measures to strengthen tax and customs administration together with improved tax policy measures to raise the collection of taxes in the 1994/95 Budget; (iii) Economic Composition of Expenditures: To ensure the funding of critical Other Charges, such as essential materials and O&M expenditures, control the growth in the wage bill and reduce expenditures on non-essential components of Other Charges in the 1994/95 Budget; (iv) Civil Service Reform: To improve the effectiveness of the civil service and to make salary enhancement a possibility, aim to reduce the size of the civil service by one-third (to 225,000) by retrenching 40,000 civil servants per annum (net) for the next three years. On pay reform, start preparations for introducing a transparent salary structure that incorporates all monetary and non-monetary allowances in the 1995/96 Budget; (v) Social Sector Strategy: To arrest and reverse the decline in the social sectors in recent years, prepare a comprehensive Social Sector Strategy in the near future; (vi) Allocation of Publc Resources to the Social Sectors: To improve the provision of basic social services, substantially increase the allocation in the 1994/95 Budget to primary education and preventive health, primarily for expendnures on critical materials and O&M. Allocate a further one percent of GDP each to basic education and basic health in the FY96 and FY97 Budgets; (vii) Strengthening Local governments: To strengthen Local governments, who bear the main responsibility for the delivery of key social services, a strategy paper on measures to be adopted to strengthen Local governments should be prepared for presentation to Cabinet as soon as possible; (viii) Public Investment Program: To improve the efficiency of public investments, prepare a 'core investment program" whose size is limited to one that can be adequately operated and maintained in the future, and whose coi.tents are limited to high priority projects. This core program will receive the bulk of Government's budgeted counterpart funds in the 1994/95 Budget; (ix) Role of Donors: To improve the Government's capacity to manage and control externally-funded expenditures, adopt comprehensive sector strategies for donors to fund jointly with Government, instead of the current project-by-project, donor-by-donor approach; (x) Aid Budgeting and Accounting: To improve the transparency of the utilization of aid, continue to improve the coverage of the Development Budget and adopt measures to improve the accounting for aid- funded expenditures. 1 Introduction 1.1 This chapter provides the background and introduction to the study by reviewing the progress made in public expenditure management since the first Public Expenditure Review for Tanzania was undertaken in 1989. It examines the nature of the problems plaguing the policymakers in Tanzania today and indicates the urgent need for the Government of Tanzania to improve the effectiveness of its expenditures, especially those devoted to investments in human capital. 1.2 The rest of the report is organized into four chapters. Chapter 2 elaborates on the role of government in economic development and examines the implications for public expenditures. Chapter 3 looks at the allocation of resources across economic categories. Chapter 4 examines spending priorities in selected key sectors. Finally, Chapter 5 deals with issues relating to improving the transparency and effectiveness of aid-funded expenditures. Background 1.3 The first Public Expenditure Reiew (PER) for Tanzania was undertaken in 1989 by the World Bank jointly with the Government of Tanzania. It covered a wide range of issues relating to public sector management. It found that the rapid buildup of the public sector in Tanzania following the Arusha declaration of 1967 had combined with the decade of economic decline preceding the launch of the Economic Recovery Programn (ERP) in 1986, to place an enormous strain on the ability of the Tanzanian Government to effectively manage its gigantic public sector. Poor public sector management had manifested itself in a wide range of problems including poorly planned and budgeted government expenditures; increased aid dependence accompanied by a lack of transparency in aid utilization; a heavily distorted tax structure; and an ineffective and demoralized civil service. The 1989 PER made a number of recommendations on how to develop an effective public expenditure strategy that would deal with these problems. Progress Since the 1989 PER 1.4 Based on the reconmendations of the 1989 PER, the Government initiated reforms in some of the key problem areas of public sector management, as described below. So far, the strategies for reform, and the institutional framework necessary for their implementation, have been put in place. However, the actual implementation of most of these reforms is in its infancy, and the recommendations of the 1989 PER remain just as relevant today. 1.5 Planning and Budgeting Reforms. In this area, one of the key reforms initiated in 1993/94 was the replacement of the old style planning techniques with a "Rolling Plan and Forward Budget" (RPFB) framework. The RPFB is a medium-term financial framework that aims to reestablish the link between Government's stated priorities and the actual resource allocation process. It is also the tool through which the Govemnment aims to introduce much-needed reforms in expenditure management such as improving the links between planning and budgeting; strengthening the links between the recurrent and development budgets; rationalizing the public investment program; and improving the transparency and efficiency of aid-funded expenditures. 2 Chapter 1 1.6 Tax Refonn. While efforts to reform the tax system in a piecemeal fashion have been underway since the launch of the ERP in 1986, the Government has only just recently started a systemic tax reform exercise. This exercise is based on the recommendations of the Presidential Commission of Enquiry into Public Revenues, Taxation and Expenditure (henceforth Tax Commission), which was set up in October 1989, and which presented its report to the Government in December 1991. With respect to both direct and indirect taxes, the Report urged, broadly, that rates be lowered and tax bases broadened. In the area of indirect taxation, one of the key recommnendations was to replace the current multiple-rate structure of sales and excise taxes by a value-added tax (VAT). The Government accepted the bulk of recommendations made by the Commission but implemented only some of them in the 1992/93 Budget (see Chapter II for details). In particular, a number of tax rates were lowered in the 1992/93 Budget but without the accompanying measures that had been recommended to broaden the tax base, contributing thus to the fiscal crisis that has since emerged. These base-broadening measures, such as introduction of the VAT and the monztization and taxation of the remuneration package for civil servants, now need to be implemented with some urgency. 1.7 Civil Service Reform. In order to attain a leaner and more efficient civil service over the next the to five years, the Government initiated a program of civil service reform (CSR) in the 1992-93 Budget. The reform program comprises the following elements: (1) improving personnel control and nmanagement to stop the ever-growing civil service; (2) retrenchment and redeployment of 50,000 redundant civil servants, to be phased over three years; (3) pay reform to establish an adequate and transparent remuneration structure; and (4) organizational and efficiency reforms to enhance the productivity of the civil service. With the assistance of several donors, including the UNDP, the World Bank, UK-ODA and SIDA, the Government has prepared an overall strategy for the CSR program linking all of these elements. The implementation of these reforms was initiated in 1992/93 with the retrenchment of the first 10,000 civil servants. Preparations have also commenced for the retrenchment of the next 20,000 targeted for retrenchment in 1993/94. In other areas, however, implementation is still in the preliminary stages. 1.8 Parastatl Reform. In January 1992, the Government initiated reforms in the parastatal sector by declaring in a Policy Statement that it had "decided to carry out a systematic reform of the parastatal sector, by attracting participation from within and outside and by revitalizing the management of institutions and enterprises within the state sector". The basic objectives of these reforms are to: (i) improve the efficiency of the sector; (ii) reduce the fiscal burden of loss-making enterprises; and (iii) expand the role of the private sector. In order to guide and implement the process of parastatal restructuring, the Government appointed a Presidential Parastatal Sector Reform Commission (PSRC). The Commission has prepared a five-year plan to privatize all commercial parastatals and to improve the efficiency of those that are to remain in the parastatal sector for the time being (such as the utilities). Implemnentation of this plan has begun and du.ing 1993, 12 commercial parastatals were either sold or placed in receivership and another six receiverships were initiated; an action program for the enforcement of a hard budget constraint on Government funding of parastatals was prepared; performance contracts were drafted for five major non-commercial parastatals; and the Govermment began the preparation of a program of support for retrenched parastatal employees. 1.9 Strengthening Local Governments. Progress has been slow in this area. There have been a number of studies that have recommended measures and proposed strategies for the strengthening of local governments. While these measures and strategies have been discussed and debated widely within Govermment, they have yet to be adopted and implemented. Chapter 1 3 Nature of the Problem 1.10 Tanzania is one of the poorest countries in the world, with the standard dollar estimate of its per capita income in 1991 being only US$ 100 (World Development Report, World Bank, 1993d). I While alternative methods of estimation may change the absolute value of the estimate of Tanzania's per capita income, they do not much alter Tanzania's ranking amongst the poorest countries of the world. 2 According to recent estimates, about one-third of Tanzanians live in absolute poverty, i.e., they are unable to afford the basic needs for survival. 3 1.11 Table 1.1 below shows the progress (or the lack thereof) that has occurred in Tanzania over the period 197-'990 in some basic indicators of economic and social development. It reveals that despite the gains made since the launch of the Economic Recovery Program in 1986, nominal per capita incomes in 1990 are not significantly higher than those in 1970. 4 The ERP has led to an average growth rate of about 4 percent per annum and has halted the decade of economic decline that preceded it. However, with the population growing at over 3 percent per annum during the period, this has resulted in only very modest gains in per capita incomes. The story on the social indicators is mixed: while life expectancy at birth and infant mortality rates have improved over the last 20 years, the primary school enrollment rate has declined during the 1980s and there has been no progress in the numbers attending secondary school. The decrease in primary school enrollment rates, combined with the deterioration in the quality of the education, is already beginning to show effect: the illiteracy rate has increased from 10 percent in 1986 to 16 percent in 1992, and at current trends, is projected to rise to 24 percent by the year 2000. Thus, it is evident that some of the social gains made earlier are now being eroded. The problem is compounded by a high fertility rate (6.6) that creates tremendous demands for additional resources and makes it difficult to provide either the desirable quantity or quality of social services. I This standard dollar estimate of Tanzania's per capita income is likely to be an underestimate and needs to be treated with caution. For example, the recent Poverty Profile (World Bank, 1993c) of Tanzania based on a sample of 1,046 households estimates the nominal figure for per capita expenditures to be US$ 280. 2 For example, the United Nations Intemational Comparison Programme (ICP) estimates Tanzania's per capita income based on the purchasing power of currency (PPC) method to be US$ 570 in 1991. However, even using this method, Tanzania remains one of the poorest countries in the world. 3 See 'Tanzania: A Poverty Profile". World Bank (1993). 4 Intertemporal comparisons of dollar per capita incomes can be misleading due to differences in the degree of overvaluation of the Tanzanian Shilling in the three periods under compaison. 4 Chapter 1 Table 1.1: Basic Economic and Social Indicators, 1970-90 Indicator 1970 1980 1990 | Per Capita Income (US$) I 100 280 110 1 Life Expectancy at Birth (years) 45 47 48 Infant Mortality Rate (per 1000 live births) 132 119 115 Primary School Gross Enrollment Rate 34 93 63 Secondary School Gross Enrollment Rate 3 3 4 Total Fertility Rate 6.4 6.8 6.6 1 May be substantially distorted due to large fluctuations in the official exchange rate during this period. Source: World Bank: World Development Reports and African Development Indicators. 1.12 Table 1.2 compares the performance of Tanzania in 1990 with that of two groups of developing countries: Sub-Saharan African (SSA) countries and low income countries (LIC). sIt reveals the fact that Tanzania lags far behind other countries in human capital investments and the living standards of its population. The Tanzanian per capita income is only a third of the average for SSA or LICs; its life expectancy at birth at 48 years is about the same as the average for SSA but substantially lower than the 62 years average achieved in LICs; its infant mortality rate (115) is higher than the average for SSA (107) and much higher than the average for LICs (69). Even in primary school enrollment, where considerable gains had been made during the 1970s, the enrollment rate in Tanzania (63) is now lower than the average for SSA (68) and substantially lower than in UCs, which on average have achieved the goal of Universal Primary Education (UPE). Tanzania's secondary school enrollment rate (4) is the lowest in the world (matched only by Malawi) and only one-fourth of the average for SSA and one-tenth of the average for LICs. Finally, the total fertility rate (TFR) in Tanzania (6.6) is comparable to that in SSA, but is high compared to other LICs (3.8). A high fertility rate implies the need to incur ever-increasing expenditures even to maintain the existing level of social service delivery, which is of poor quality by international standards. With the demands created by the high fertility rate-leading to a population growth rate of over 3 percent and a population structure where over 45 percent of the population is below 15 years of age-a heroic effort will be needed to actually attain an improvement in the quantity and quality of human capital investment. 5 Low income countries are those with a GNP per capita of US$ 635 or less in 1991. chapter 1 5 Table 1.2: International Comparison of Basic Indicators, 1990 Indicator Tanzania SSA' LIC2 Per Capita Income (US$) 110 340 350 Life Expectancy at Birth (years) 48 51 62 Infant Mortality Rate (per 1000 live births) 115 107 69 Primary School Gross Enrollment Rate 63 105 68 Secondary School Gross Enrollment Rate 4 17 41 Total Fertility Rate 6.6 6.5 3.8 1 Sub-Saharan Africa. 2 Low Income-Countries (defmed in footnote 5). Source: World Bank: World Development Reports and African Development Indicators. 2 Role of Government *...the 1992/93 Budget has been evolved to address a number of areas including the redefining of the role of the Government with a vew to reducing its size. In this regard the Government will be constrained to confine uself to its traditionalfunctions which it can reasonably sustain. n Minister of Finance Budget Speech, June 1992 2.1 This chapter examines the implications for public expenditures of the Govermnent's decision to redefine its role in the economy. It indicates that there is no scope for further expansion of public expenditures without damaging the fragile macroeconomic enviromnent. In fact, given the current mismatch between revenues and expenditures, and given the general tightening of the aid climate, there is an acute need to reduce ine size of public expenditures. However, the only way that a reduction in the size of expenditures can be achieved together with an improvement in the quantity and quality of basic govermnent services is by improving the allocation of public expenditures. How this should be done, both across and within sectors, is the subject of discussion of the following chapters. 2.2 The 1991 World Development Report (WDR) on "The Challenge of Development" drew on an exhaustive review of the experience of developing economies during the past 30 years to explain the sharp contrast between the successes of some and the failures of others. A central issue in development and a principal theme of that WDR is the interaction between governments and markets. The WDR suggests a reappraisal of the respective roles of the market and the state. Put simply, governments need to do less in those areas where markets work, or can be made to work reasonably well, and do more in those areas where markets alone cannot be relied upon. Above all, as far as priorities for government actions are concerned, the implications are that governments should: (i) provide a stable macroeconomic foundation without which little can be achieved; and, (ii) invest in education, health, nutrition, family planning, and poverty alleviation; and build social, physical, administrative, regulatory and legal infrastructure of better quality. 2.3 Recent research on the rapidly growing economies of East Asia also further supports the desirability of a two-track approach to development policy emphasizing macroeconomic stability on the one hand and investments in people on the other. 6 This research also indicates that the key to ataining macroeconomic stability in these countries was fiscal prudence. This chapter deals with the issue of macroeconomic stability in Tanzania. It examines in detail the fiscal situation, and especially 6 See 'The East Asian Miracle: Economic Growth and Public Policy", World Bank, 1993a. 7 Chapter 2 the revenue issues that are the key cause of the current macroeconomic problems that are facing Tanzania. The following chapters examine the implications for Government expenditures of the newly defined role of the Goverunent in Tanzania. Macroeconomic Foundation 2.4 In Tanzania, while the macroeconomic situation has improved since the launch of the ERP, it nevertheless remains fragile. 7 The performance of investmen., savings, and exports have all improved after the launch of reforms. However, the savings-investment gap remains large, and official exports finance only about one-third of imports, making Tanzania extremely dependent on foreign aid to finance its imports, investments, and even recurren expenditures. As a result, there remains a critical need to maintain prudent fiscal and monetary policy and to continue with the efforts to bring inflation down. Furthermore, appropriate market-based incentives for raising the domestic savings rate and improving the efficiency of investment are necessary if Tanzania is to reduce its dependency on foreign aid. 2.5 Table 2.1 provides data on the size of Central Govermnent operations in Tanzania in recent years to indicate the magnitude of the problem facing the Tanzanian Government. It reveals the growing mismatch between revenues and expenditures in recent years and the fragile fiscal situation that has resulted. In FY91 and FY92, the Govermment maintained a tight fiscal stance and, as a result, in both years it generated a fiscal surplus after grants. In addition, in FY92 it also generated positive savings of 1.6 percent of GDP on its current account. In comparison, preliminary results for FY93 indicate that the fiscal position has deteriorated severely. Revenues fell from 23.5 of GDP in FY92 to 16.8 percent in FY93, a fall of 6.7 percent of GDP in one year. The resulting fiscal deficit (after grants) in FY93 was 8.1 percent of GDP, and Government dissavings were at 7.8 percent of GDP. 2.6 The FY93 fiscal outcome eroded the gains made in revenue enhancement during the reform period, when revenues went up steadily from a mere 14.9 percent of GDP in FY86 to a high of 23.5 percent of GDP in FY92. Expenditures have also grown over the period and bave risen from 23.0 percent of GDP in FY86 to 26.3 percent in FY92. The rpismatch between revenues and expenditures in FY93 contributed to the deterioration in the macroeconomic environin: -it of the economy. Measures were therefore adopted in the FY94 Budget that aimed to enhance revenue collection and to reduce the deficit back to more sustainable levels (1.3 percent of GDP after grants). 2.7 Preliminary data for the first half of FY94 indicate, however, that the revenue enhancement measures adopted in the June 1993 Budget have not succeeded in reversing the decline in revenue collection experienced in FY93 and that in fact, the macroeconomic situation has deteriorated dramatically. The decrease in revenue collection has meant that the Government has resorted to bank borrowing, from the central bank and from the commercial banks, to finance its activities and is thus contributing both to continuing high levels of inflation and to a credit squeeze on the private sector. Hence, the weak financial performance of the Tanzanian economy during FY93 and FY94 can be attributed principally to the weak fiscal position, which has become a cause of great concern. 7 For an evaluation of the macroeconomic performance of Tanzania during the five yea; period following the launch of the ERP, see "Strucwtal Adjustmnent, Economic Performance, and Aid Dependency in Tanzania'. by Agrawal, Ahmed, Mered and Nord (World Bank Working Paper No. 1204, October 1993). Chapter 2 8 Table 2.1: Size of Central Government Operations (selected years between 1986 and 1994) (perceat of GDP) Actual Prelim Bud,eted Actual FY86 FY92 FY93 FY94 Total Revenue 14.9 23.5 16.8 20.5 Total Expenditure 23.1 26.3 30.8 33.4 Recurrent Exp 19.2 21.9 24.6 21.9 Development Exp 3.9 4.4 6.2 11.5 Deficit before Grants -8.2 -2.8 -14.0 -13.0 Grants ,, 5.2 6.0 11.6 Deficit after Grants .. 2.3 -8.1 -1.3 (checks issued) Financing 2 6.0 -2.0 7.8 1.3 Foreign (net) 1.6 2.3 3.0 4.5 Domestic (net) 4.4 -4.4 4.8 -3.2 Bank 3.7 -4.9 4.4 -3.4 Non-Bank 0.7 0.6 0.4 0.3 Memo item Government SavingS 3 -4.3 1.6 -7.8 -1.4 Inflation Rate 32.4 22.1 23.2 12.04 1 The lsrge increase in the size of budgeted Development expenditues between FY93 and FY94 reflects the Government's efforts to record more fiuly in the Budget the totality of the aid that it receives from donors. 2 The difference between the financing amount and the deficit after grants is due to the "float, which is the difference between checks issued and checks cleared. 3 Defined as the difference between revenues and recurrent expenditures (on a checks issued basis). 4 Targeted rate. 2.8 Weaknesses in the fiscal situation arose in 1992, when the Government made a partial attempt to reform the tax system following the presentation of the Tax Commission's Report in December 1991. In June 1992, consistent with the recommendations of the Tax Commission, various taxes were simplified, and a number of rates were lowered in the FY93 Budget. These were accompanied by some measures aimed at widening the tax base. It was expected that these measures would yield additional revenue that would more than offset the anticipated decline in revenue due to the tax rate reductions. The revenue enhancing measures included the elimination of exemptions on customs duties and the improvement of duty collection with the introduction of preshipment inspection arrangements. In the event, however, there were major shortcomings in the implementation of these measures to widen the tax base, and indeed, tax exemptions, far from being eliminated, have increased even further. This is reflected in the fact that in September 1993, for example, less than 10 percent of import taxes payable under the preshipment inspection/tax assessment program had actually been paid. In these circumstances, and exacerbated by the introduction of a remission order at the end of August 1993 exempting most raw materials from import duties, Government borrowing from the banking system has continued to increase sharply during the first half of FY94. 9 Chapter 2 2.9 In addition to poor implementation of the announced measures, another major factor contributing to the current fiscal problems was the inordinate delay in the adoption of important base- broadening measures recommended by the Tax Comnmission. The two major recommendations in this area were the introduction of a value-added tax (VAT) and the monetization and subsequent incorporation of all monetary and non-monetary allowances paid to formal sector employees into the income tax base. In the June 1992 Budget Speech, the Government announced it intention to replace the existing Sales Tax with a VAT in January 1994. Given the substantial amount of lead time (ranging from 18-24 months) that is necessary for the preparation for a VAT, its implementation by January 1994 would have required that preparations commence immediately following that announcement. Instead, to this day, very little progress has been made in this area, and even if intensified efforts are made to launch the preparations immediately, the earliest that it will be possible to introduce the VAT would be by mid 1995. 2.10 The adoption of the second major base-broadening measure, i.e., the widening of the income tax base, has also suffered inordinate delays. One of the key elements of the Govermnent's overall program to improve the efficiency of the civil service is pay reforn (see Chapter 3 for details). The urrent remuneration system for civil servants consists of a meager basic salary supplemented by a host of non-taxable monetary and non-monetary allowances. This severely erodes the tax base, especially since the Government's practice of paying non-taxable allowances to its employees has also been adopted by the private sector. The Tax Commission strongly recommended the taxation of all monetary allowances and fringe benefits. However, no progress has been made as yet in this area. Given the tight fiscal situation, there is an urgent need to adopt such base-broadening measures. 2.11 While progress has been slow in taking the more medium-term measures of tax reform, such as the preparation for the VAT and the widening of the income tax base, the Government has recognized the serious deterioration in the fiscal situation and in January 1994, it took several other measures to improve the situation. A package of revenue measures was adopted that included the reimposition of a 10 percent customs duty on all imported raw materials for locally manufactured geonds (thereby rescinding the exemptions order issued in August 1993), the inclusion of the excise tax in the sales tax base of locally produced goods, the strict adherence to preshipment inspection, the immediate withdrawal of the concession allowing staggered payments of taxes combined with renewed efforts to collect arrears that had resulted from this concession, the strict monitoring of bonded warehouses, and a review of tax exemptions with a view to decreasing their number. New taxes include a surtax on cars, an increase in the transport withholding tax, an increase in the customs duty on all imported commodities which are also produced locally, and an increase in a number of customs fees and penalties. The Government estimates that this package of revenue measures will generate revenues equivalent to about 2.8 percent of GDP on an annualized basis. While the main emphasis of the package adopted in January 1994 was on stemming the deterioration on the revenue side, some measures were also taken on the expenditure side, to curtail expenditures as well as to improve their monitoring and control. 2.12 A critical priority for Tanzanian policymakers is the need to attain and maintain mnacroeconomic stability. As Table 2.1 indicates, the Government's objective is to bring inflation down to 12 percent in FY94 from a level of 23 percent in FY93. However, in the first half of PY94, the annual rate of inflation continued at the level of FY93. To achieve the 12 percent target for FY94 would thus imply a marked reduction in the rate of monetary growth during the second half of FY94, to about half its current level of about 35 per cent per annum. This in turn will necessitate a significant tightening of the fiscal stance. While the budgeted fiscal deficit (after grants) for FY94 is only 1.3 percent of GDP, given the poor revenue performance in the first half of the year, it is Chapter 2 10 unlikely that this target can be met even taking into account the measures adopted in January 1994, without further efforts to enhance revenues and curtail expenditures. 2.13 A second priority for the Tanzanian policymakers is to ac4'elerate growth by stimulating the private sector. Large fiscal deficits that are financed domestically have negative consequences for the private sector, whether the deficits are financed through banks or the non-bank sector. During FY93, the Government financed the bulk of its deficit by borrowing from the banking system. This mechanism limits the access of the private sector to available credit and creates the danger that private sector development would be stifled due to lack of credit. During FY94, the Government has increased its reliance on non-bank borrowing through the use of Treasury bills (T-bills). During the first half of FY94, the stock of T-bills outstanding has increased over three-fold, from T Sh 4.7 billion at the end of June 1993 to T Sh 16.2 billion at the end of December 1993. To facilitate the sale of T-bills, weekly auctions were introduced by the Bank of Tanzania in August 1993. The interest rates determined through this mechanism have been fairly high (averaging about 40 percent currently). While resorting to non-bank financing of the deficit reduces the risk of building up furter inflationary pressures in the economy, it creates the risk of crowding out private investment. The only way for the Government to prevent destabilizing the economy further and to prevent stifling the private sector is to take all the measures necessary to reduce the fiscal deficit to a level that can be fmanced without recourse to domestic financing. Composition of Public Expenditures 2.14 In addition to stressing the role of goverunents in providing a sound macroeconomic foundation for growth, the 1991 WDR also provides direction for the focus of public expenditures. In particular, it stresses the need for investing in people. The WDR argues that the economic and social returns from public and private investments in people are often extremely high. The Government has an important role to play in adequately financing basic social services-education (especially primary education), health care, nutrition and family planning services-especially for the poor. Therefore, a priority for Government must be to spend adequately and effectively on these basic services. In addition to spending on people, the Government also needs to incur other expenditures that have proven to have high returns, such as law and order, agricultural research and development (R&D), and investments in infrastructure. The next two chapters examine the composition of Government expenditures in Tanzania in greater detail. 3 Economic Composition of Public Expenditures 3.1 The Goverunent of Tanzania faces a formidable challenge: it needs to substantially downsize expenditres in relation to GDP and it needs to improve their allocation. This chapter examines the composition of spending by "economic" categories, i.e., wages and salaries, other goods and services, suosidies and transfers, and so forth. The following chapter examnines key expenditure issues by "function" or sector, i.e., education, health, agriculture, and so forth. The main data sources for the PER are the Budget Estimates for various years. Each year's Estimates are published in four volumes: Revenues (Volume I); Consolidated Fund Services and Ministerial Supply (Volume Il); Regional Supply (Volume III); and the Development Budget (Volume IV). The most recent Budget Estimates are for FY94. Forecasts for FY95 have been obtained from the first Rolling Plan and Forward Budget (RPFB), which was for the years 1993/94-1995/96. Table 3.1: Breakdown of Total Expenditures, 1992-94 (percentage shares) Actual Expenditure Approved Budgeted Etimates FY92 Estimates FY93 FY94 Recurrent 83.9 79.5 68.3 Consolidated Fund Services 27.2 21.6 19.4 Ministerial Supply 47.3 43.3 36.8 Regional Supply 9.4 14.6 12.0 Development 16.1 20.5 31.7 Total Expenditures 100.0 100.0 100.0 Source: For Recurrent expenditures, FY94 Budget Estimates; for Development expenditures, the FY92 figure is obtained from the Report of the Controller and Auditor-General; the FY93 figure is a preliminary estimate obtained from the Tanzanian authorities; and the FY94 figure is from the FY94 Budget Estimates. 3.2 Table 3.1 above depicts the breakdown of total expenditures into broad categories for the period FY92-FY94. In FY94, it shows that Recurrent accounted for 68 percent and Development expenditures for the remaining 32 percent of total expenditures. Furthermore, Recurrent expenditures are broken down into three broad categories: Consolidated Fund Services (CFS), Ministerial Supply, and Regional Supply, which in FY94 accounted for 19 percent, 37 percent, and 12 percent, respectively, of total expenditures. CFS is essentially for debt servicing. The remainder is recurrent expenditure broken down by level of Govermnent. In Tanzania, spending is undertaken by two levels of governments: the Central govermment (including regional governments) and Local governments. Govermnent Recurrent expenditure (net of debt servicing) is allocated to 41 "ministerial supply votes" or central spending agencies and to 20 "Regional supply votes" or Regional governments. In addition, 12 Chapter 3 the Central Goverment gives block grants to Local governments, which consist of 19 urban and 82 district (rural) councils. I Recurrent Versus Development Expenditures 3.3 One reason for the low productivity of Tanzania's investment expenditures is that there are not enough recurrent funds available to operate and maintain these investments once they are completed. Before we examine the relative sizes of the Recurrent and Development budgets, the reader is cautioned that in looking at Development expenditures in Tanzania, it needs to be kept in rnind that until recently the bulk of these expenditures bypassed the budgetary system. Thus, in FY92, only about 20 percent of project-related aid was in the Budget. While the coverage of the Development budget has improved significantly in the last two years, it is estimated (see Chapter 5) that still only about 60 percent of project-related aid has been brought into the FY94 Budget. 3.4 If we look at the breakdown of total expenditures into Recurrent and Development expenditures during FY94, Table 3.1 indicates that Recurrent expenditures account for 68 percent and Development expenditures for the remaining 32 percent. However, the Development budget significantly overestimates the share of capital spending in total spending since the PER estimates that almost a quarter of the expenditures on the Development Budget are actually recurrent in nature. If we adjust for this, then the share of capital expenditures in total expenditures drops to 24 percent. However, even the adjusted figure for Tanzania is large compared with the average for Africa (20), Asia (15) or All Developing Countries (14). 9 Given the already severe underfunding of recurrent expenditures, the implications of these high capital expenditures in Tanzania for future recurrent costs need to be examined carefully. 3.5 The effect of the Development budget on future recurrent spending depends on both the iagnitude as well as the composition of the Development budget, since some investment expenditures have a relatively high recurrent impact (mainly in the social sectors), while those in the economic service areas have a relatively low impact. Little empirical work has been done on the recurrent implications of investment in Tanzania. However, using the same sectoral "r" coefficients (which estimnate the annual recurrent implication as a fraction of the investment) as were used in the 1989 PER for Tanzania, the average "r" coefficient and the impact on recurrent expenditures arising from Tanzania's Development budget for 1993/94 is shown in Table 3.2 below. '
Группа Всемирного банка · Pre-2003 Economic or Sector Report
Tanzania - Role of government : public expenditure review (Vol. 1 of 2) : Main report
Открыть оригинал документа
Полный текст размещён на сайте публикующей организации. lawenc.com индексирует метаданные и ведёт на официальный источник.
Полный текст
Основные сведения
Организация
Группа Всемирного банка
Тип документа
Pre-2003 Economic or Sector Report
Страна
Танзания
Источник
Всемирный банк