Document of The World Bank Report No.: 78188 PROJECT PERFORMANCE ASSESSMENT REPORT TANZANIA POVERTY REDUCTION SUPPORT CREDITS 1-8 (IDA-37720 IDA-37721 IDA-H0400, IDA-41700, TF-90200, IDA-42940, IDA-45260, IDA-46350, IDA-46830, TF-94420, IDA-48130) June 14, 2013 IEG Public Sector Evaluation Independent Evaluation Group Currency Equivalents (annual averages) Currency Unit = Tanzania Shilling 2002 US$1.00 TSh 972.00 2003 US$1.00 TSh 1062.81 2004 US$1.00 TSh 1,041.01 2005 US$1.00 TSh 1,165.50 2006 US$1.00 TSh 1,261.00 2007 US$1.00 TSh 1,122.00 2008 US$1.00 TSh 1,320.00 2009 US$1.00 TSh 1,339.50 2010 US$1.00 TSh 1,505.00 2011 US$1.00 TSh 1,583.00 Abbreviations and Acronyms CAS Country Assistance Strategy MKUKUTA Swahili acronym for National CPIA Country Policy and Institutional Strategy for Growth and Reduction of Assessment Poverty DFID Department for International MTEF Medium-Term Expenditure Development (United Kingdom) Framework EU European Union NGOs Non-Governmental Organizations GBS General Budget Support PAF Performance Assessment Framework GDP Gross Domestic Product PER Public Expenditure Review HIPC Heavily Indebted Poor Countries PFM Public Financial Management HIV/AIDS Human Immunodeficiency PMO-RALG Prime Minister‘s Office—Regional Virus/Acquired Immune Deficiency Administration and Local Syndrome Government ICR Implementation Completion and PPAR Project Performance Assessment Results Report Report ICRG International Country Risk Guide PPP Public Private Partnership IDA International Development PPRA Public Procurement Regulatory Association Authority IEG Independent Evaluation Group PRSC Poverty Reduction Support Credit IMF International Monetary Fund PRSP Poverty Reduction Strategy Paper KfW Kreditanstalt fuer Wiederaufbau PSAC Programmatic Structural Adjustment (KfW Development Bank) Credit LGAs Local Government Authorities TANESCO Tanzania Electric Supply Company M&E Monitoring and Evaluation USD United States Dollars MDAs Ministries, Departments and WGI Worldwide Governance Indicator Agencies Fiscal Year Government: July 1 –June 30 Director-General, Independent Evaluation : Ms. Caroline Heider Director, IEG Public Sector Evaluation : Mr. Emmanuel Jimenez Manager, IEG Public Sector Evaluation : Mr. Mark Sundberg Task Manager : Mr. Brett Libresco Contents Preface ............................................................................................................................................ ix Summary ........................................................................................................................................ xi Series 2: PRSCs 4-8 (2006-2011, USD 1,025 million) ............................................................ xiii 1. Background and Context ............................................................................................................. 1 2. Series 1: PRSCs 1-3 .................................................................................................................... 8 Objectives, Design, and Relevance ............................................................................................. 8 Implementation ......................................................................................................................... 11 Achievement of the Objectives ................................................................................................. 12 Objective 1: Sustaining and Accelerating Economic Growth and Broadening its Impact. .. 12 Objective 2: Supporting a Results Orientation of Public Service Delivery .......................... 19 Objective 3. Enhancing Public Sector Performance ............................................................. 23 Objective 4: Strengthening Environmental Management ..................................................... 26 Ratings ...................................................................................................................................... 28 Outcome................................................................................................................................ 28 Risk to Development Outcome ............................................................................................. 30 Bank Performance ................................................................................................................ 31 Borrower Performance.......................................................................................................... 33 Monitoring and Evaluation ................................................................................................... 33 3. Series 2: PRSC 4-8 .................................................................................................................... 34 Objectives, Design, and Relevance ........................................................................................... 34 Implementation ......................................................................................................................... 37 Achievement of the Objectives ................................................................................................. 40 Objective 1: Growth and Reduction of Income Poverty ....................................................... 40 Objective 2: Improvements in Quality of Life and Social Well-being ................................. 46 Objective 3. Improved State of Governance and Accountability ........................................ 52 Ratings ...................................................................................................................................... 57 Outcome................................................................................................................................ 57 Risk to Development Outcome ............................................................................................. 58 Bank Performance ................................................................................................................ 59 Borrower Performance.......................................................................................................... 61 Monitoring and Evaluation ................................................................................................... 62 4. Lessons ...................................................................................................................................... 63 This report was prepared by Brett Libresco, with major inputs from James Sackey, Steven Webb, Roger Grawe, and Aghassi Mkrtchyan. The team visited Tanzania in March 2012. The report was peer reviewed by Brendan Horton and panel reviewed by Robert Mark Lacey. Aline Dukuze provided administrative support. i References ..................................................................................................................................... 67 Annex A. Basic Data Sheet ........................................................................................................... 73 Annex B. Complementarity of PRSCs with Other Bank Support ................................................. 89 Annex C. PRSCs 1-3: Detailed Assessment of Progress by Objective. ........................................ 94 Annex D. PRSCs 4-8: Detailed Assessment of Progress by Objective ....................................... 107 Annex E. List of Persons Consulted ............................................................................................ 134 Annex F. Borrower Comments.................................................................................................... 138 Tables Table 1-1. Time Line of Events Including the Implementation of PRSCs 1-3 ............................... 3 Table 1-2. Tanzania – Selected Economic and Social Indicators, 2000-2010 ................................ 5 Table 1-3.Time Line of Events Including the Implementation of PRSCs 4-8 ................................ 7 Table 2-1.Timing and Scale of PRSCs 1-3.................................................................................... 11 Table 2-2. Selected 2006 PEFA Ratings on Public Sector Performance ...................................... 25 Table 2-3. Summary of Outcome Ratings for PRSCs 1-3 ............................................................. 29 Table 3-1. Timing and Scale of PRSCs 4-8 .................................................................................. 38 Table 3-2. Progress on the Triggers for PRSC Program 2 (in numbers) ....................................... 38 Table 3-3. Key Outcome Indicators for PRSC Development Objectives ..................................... 40 Table 3-4. Tanzania-ICRG and CPIA Ratings ............................................................................. 53 Table 3-5. Summary of outcome ratings for PRSCs 4-8 ............................................................... 58 Figures Figure 1-1. Trends in Government Expenditure and Development Assistance. ............................. 6 Figure 2-1. Real GDP Growth in Tanzania and Comparator Countries ........................................ 13 Figure 2-2. Real Growth and Poverty in Non-Fragile Low Income Countries during 2000s ....... 14 Figure 2-3.Sector growth in Agriculture and Mining and Quarrying ............................................ 18 Figure 2-4. Total Absolute Deviation between Approved Budget Allocations and Actual Expenditures Declined Under the Operation ................................................................................. 22 Figure 2-5. Trends in Worldwide Governance Indicators for Aspects of Public Sector Performance................................................................................................................................... 24 Figure 3-1. Trend in Under-five Mortality in Tanzania ................................................................ 48 Figure 3-2. Indicators of Progress in Education. ........................................................................... 49 Figure 3-3. Total Absolute Deviation between Approved Budget Allocations and Actual Expenditures Declined Under the Operation ................................................................................. 56 ii Principal Ratings TANZANIA POVERTY REDUCTION SUPPORT CREDITS 1-3 ICR* ICR Review* PPAR Outcome Satisfactory Satisfactory Moderately satisfactory Risk to Development Moderate Moderate Significant Outcome Bank Performance Satisfactory Satisfactory Moderately satisfactory Borrower Satisfactory Satisfactory Moderately satisfactory Performance TANZANIA POVERTY REDUCTION SUPPORT CREDITS 4-8 ICR* ICR Review* PPAR Outcome Moderately satisfactory Moderately unsatisfactory Moderately unsatisfactory Risk to Development Moderate Significant Significant Outcome Bank Performance Satisfactory Moderately satisfactory Moderately satisfactory Borrower Moderately satisfactory Moderately unsatisfactory Moderately unsatisfactory Performance * The Implementation Completion and Results Report (ICR) is a self-evaluation by the responsible Bank department. The ICR Review is an intermediate IEG product that seeks to independently verify the findings of the ICR. Key Staff Responsible TANZANIA 1ST POVERTY REDUCTION SUPPORT CREDIT (LOAN 37721 & GRANT H0400) Division Chief/ Project Task Manager/Leader Sector Director/Manager Country Director Appraisal Benno Ndulu Frederic Kilby Judy M. O‘ Connor Completion Robert Utz Kathie L Krumm Judy M. O‘ Connor TANZANIA 2ND POVERTY REDUCTION SUPPORT CREDIT (LOAN 3965-TA & GRANT H118- TA) Division Chief/ Project Task Manager/Leader Sector Director/Manager Country Director Appraisal Robert Utz Kathie L Krumm Judy M. O‘ Connor Completion Robert Utz Kathie L Krumm Judy M. O‘ Connor TANZANIA 3RD POVERTY REDUCTION SUPPORT CREDIT (LOAN 4110-TA) Division Chief/ Project Task Manager/Leader Country Director Sector Director/Manager Appraisal Robert Utz Kathie L Krumm Judy M. O‘ Connor Completion Robert Utz Kathie L Krumm Judy M. O‘ Connor v TANZANIA 4TH POVERTY REDUCTION SUPPORT CREDIT (LOAN 4170-TA & TF-90200) Division Chief/ Project Task Manager/Leader Sector Director/Manager Country Director Appraisal Robert Utz Kathie L Krumm Judy M. O‘ Connor Completion Paolo B. Zachia Kathie L Krumm John McIntire TANZANIA 5TH POVERTY REDUCTION SUPPORT CREDIT (LOAN 4294-TA) Division Chief/ Project Task Manager/Leader Sector Director/Manager Country Director Appraisal Robert Utz Kathie L Krumm Judy M. O‘ Connor Completion Paolo B. Zachia Kathie L Krumm John McIntire TANZANIA 6TH POVERTY REDUCTION SUPPORT CREDIT (LOAN 4526-TA) Division Chief/ Project Task Manager/Leader Sector Director/Manager Country Director Appraisal Paolo B. Zacchia Kathie L Krumm John M. McIntire Completion Paolo B. Zacchia Kathie L Krumm John M. McIntire TANZANIA 7TH POVERTY REDUCTION SUPPORT CREDIT (LOAN 4635-TA) Division Chief/ Project Task Manager/Leader Sector Director/Manager Country Director Appraisal Paolo B. Zacchia Kathie L Krumm John M. McIntire Completion Paolo B. Zacchia Kathie L Krumm John M. McIntire TANZANIA 7TH ADDITIONAL POVERTY REDUCTION SUPPORT CREDIT (LOAN 4635-1-TA) Division Chief/ Project Task Manager/Leader Sector Director/Manager Country Director Appraisal Javier Suarez Kathie L Krumm John M. McIntire Completion Paolo B. Zacchia Kathie L Krumm John M. McIntire TANZANIA 8TH POVERTY REDUCTION SUPPORT CREDIT (LOAN 4813-TZ) Division Chief/ Project Task Manager/Leader Sector Director/Manager Country Director Paolo B. Zacchia & Appraisal Kathie L Krumm John M. McIntire Yutaka Yoshino Completion Yutaka Yoshino J. Humberto Lopez Philippe Dongier vi IEG Mission: Improving World Bank Group development results through excellence in independent evaluation. About this Report The Independent Evaluation Group assesses the programs and activities of the World Bank for two purposes: first, to ensure the integrity of the Bank’s self-evaluation process and to verify that the Bank’s work is producing the expected results, and second, to help develop improved directions, policies, and procedures through the dissemination of lessons drawn from experience. As part of this work, IEG annually assesses 20-25 percent of the Bank’s lending operations through field work. In selecting operations for assessment, preference is given to those that are innovative, large, or complex; those that are relevant to upcoming studies or country evaluations; those for which Executive Directors or Bank management have requested assessments; and those that are likely to generate important lessons. To prepare a Project Performance Assessment Report (PPAR), IEG staff examine project files and other documents, visit the borrowing country to discuss the operation with the government, and other in-country stakeholders, and interview Bank staff and other donor agency staff both at headquarters and in local offices as appropriate. Each PPAR is subject to internal IEG peer review, Panel review, and management approval. Once cleared internally, the PPAR is commented on by the responsible Bank department. The PPAR is also sent to the borrower for review. IEG incorporates both Bank and borrower comments as appropriate, and the borrowers' comments are attached to the document that is sent to the Bank's Board of Executive Directors. After an assessment report has been sent to the Board, it is disclosed to the public. About the IEG Rating System for Public Sector Evaluations IEG’s use of multiple evaluation methods offers both rigor and a necessary level of flexibility to adapt to lending instrument, project design, or sectoral approach. IEG evaluators all apply the same basic method to arrive at their project ratings. Following is the definition and rating scale used for each evaluation criterion (additional information is available on the IEG website: http://worldbank.org/ieg). Outcome: The extent to which the operation’s major relevant objectives were achieved, or are expected to be achieved, efficiently. The rating has three dimensions: relevance, efficacy, and efficiency. Relevance includes relevance of objectives and relevance of design. Relevance of objectives is t he extent to which the project’s objectives are consistent with the country’s current development priorities and with current Bank country and sectoral assistance strategies and corporate goals (expressed in Poverty Reduction Strategy Papers, Country Assistance Strategies, Sector Strategy Papers, Operational Policies). Relevance of design is the extent to which the project’s design is consistent with the stat ed objectives. Efficacy is the extent to which the project’s objectives were achieved, or are expected to be achieved, taking into account their relative importance. Efficiency is the extent to which the project achieved, or is expected to achieve, a return higher than the opportunity cost of capital and benefits at least cost compared to alternatives. The efficiency dimension generally is not applied to adjustment operations. Possible ratings for Outcome: Highly Satisfactory, Satisfactory, Moderately Satisfactory, Moderately Unsatisfactory, Unsatisfactory, Highly Unsatisfactory. Risk to Development Outcome: The risk, at the time of evaluation, that development outcomes (or expected outcomes) will not be maintained (or realized). Possible ratings for Risk to Development Outcome: High, Significant, Moderate, Negligible to Low, Not Evaluable. Bank Performance: The extent to which services provided by the Bank ensured quality at entry of the operation and supported effective implementation through appropriate supervision (including ensuring adequate transition arrangements for regular operation of supported activities after loan/credit closing, toward the achievement of development outcomes. The rating has two dimensions: quality at entry and quality of supervision. Possible ratings for Bank Performance: Highly Satisfactory, Satisfactory, Moderately Satisfactory, Moderately Unsatisfactory, Unsatisfactory, Highly Unsatisfactory. Borrower Performance: The extent to which the borrower (including the government and implementing agency or agencies) ensured quality of preparation and implementation, and complied with covenants and agreements, toward the achievement of development outcomes. The rating has two dimensions: government performance and implementing agency(ies) performance. Possible ratings for Borrower Performance: Highly Satisfactory, Satisfactory, Moderately Satisfactory, Moderately Unsatisfactory, Unsatisfactory, Highly Unsatisfactory. vii Preface This is the Project Performance Assessment Report (PPAR) for eight closed Tanzania Poverty Reduction Support Credits (PRSCs), implemented as two programmatic series. Program-1 included the PRSC 1-3 series implemented from 2003-2006. Program-2 included the PRSCs 4-8 implemented from 2006-2011. World Bank disbursements for Program-1 amounted to a total of USD 432 million, while those of Program-2 amounted to a total of USD 1,025 million (including a supplemental credit of USD 170 million to offset the negative implications of the 2008 global financial crisis on Tanzania). The two programs were co-financed by Kreditanstalt für Wiederaufbau (KfW, Germany) and complemented by the General Budget Support partners (African Development Bank, European Union, Canada, Denmark, Finland, Japan, the Netherlands, Norway, Sweden, Switzerland, Ireland and the U.K. Department for International Development). The report was prepared by the Independent Evaluation Group (IEG). It is based on interviews and documents and data collected in the course of a visit to Tanzania in March, 2012, during which a number of stakeholders, including government officials, external development partners, business groups, academics, non-governmental organizations, as well as other civil society groups were consulted. The evaluation also draws upon in-depth interviews of staffs of the World Bank and the International Monetary Fund, including current and former members of the Tanzania country teams in Washington and Dar es Salaam; published and internal documents from the two institutions; and recently published and unpublished studies on Poverty Reduction Strategy Paper-related processes in Tanzania. The cooperation and assistance of all stakeholders and government officials are gratefully acknowledged as is the support of the World Bank Country Office in Tanzania. The two Tanzania PRSC series were among the earliest sustained Bank-financed operations to support a country-owned reform process conducted in the context of the Paris Declaration for Aid Effectiveness. Because their implementation spanned a long period of time, the assessment provides an opportunity to derive lessons on the long-term implications of sustained support through the PRSC instrument of development policy lending. Following standard IEG procedure, copies of the draft PPAR were sent to the government for review. Comments were received and are attached as Annex F. ix Summary Context This report assesses the outcome of World Bank support for eight Poverty Reduction Support Credits (PRSCs) in Tanzania- totaling nearly USD 1.5 billion in finance- implemented from 2003 to 2011 as two program series. After deterioration in relations between Tanzania and external partners in the early 1990s, a renewed relationship evolved around the principle of 'mutual accountability.' A government-initiated economic recovery plan in 1996 demonstrated the government's commitment to improved fiscal performance and structural reform, and introduced a new dynamic of donor engagement. The World Bank (hereinafter referred to as the Bank) financed a 1997-2000 structural adjustment credit and a 2000-2003 programmatic structural adjustment credit, which conditioned disbursement on the implementation of pre- identified policy changes. To support greater country ownership and flexibility, the two PRSC series (2003-2005, 2006-2011) continued the programmatic approach around a sequence of annual single-tranche operations, but used a framework of indicative triggers and ex-post prior actions derived from the government's own poverty reduction strategies rather than formal conditions. The new PRSC series also contributed to a joint multi-donor dialogue comprising a harmonized approach to general budget support. Series 1: PRSCs 1-3 (2003-2005, USD 432 million) The first series lacked a functional results framework. Objectives for the series were not settled until the second operation, and program documents did not clearly articulate how progress in adopting the proposed policy measures would help to achieve the program ' s objectives. This evaluation reconstructs a plausible route by which PRSC-supported activities could have been expected to contri.bute to achieving the stated objectives. Progress was modest on the first objective of sustaining and accelerating economic growth and broadening its impact. Growth of gross domestic product (GDP) was robust during the period, ranging between 7 and 8 percent per annum (a significant portion of this was driven by government spending financed through increased aid flows). Income poverty reduction, however, was significantly less than envisioned. It fell just two percentage points at the national poverty line between 2001 and 2007, yielding a very low growth elasticity of poverty by international standards. The PRSC's contribution to these outcomes was limited to a supporting role in the International Monetary Fund-led program to maintain macroeconomic stability. The growth rate of industry was strong at 10 percent annually, but this was likely driven by external factors as the business environment did not improve by available measures despite PRSC support for several reforms. Growth performance in agriculture met its five percent target, but implementation of the PRSC-supported reform program in the sector was slow and weak. On the second objective of supporting results orientation ofpublic service delivery, the Bank brought greater policy content and a hi gher level of rigor to the performance assessment framework and helped to set up a system for monitoring progress in the sectors. Data were widely available on outcomes in health and education and these sectors showed the greatest improvements during the period--expanded access to education, reduced infant mortality, and declining AIDS incidence. But the quality, frequency, and availability of data on the outcomes of other govenunent services such as water and electricity were weak, and evidence on outcomes from new service delivery surveys were not linked to the strategic planning process. The focus of the government and of development partners was initially on demonstrating that adequate resources were devoted education, health, I-INI AIDS, agriculture, roads, water and the judiciary. In practice, the share of budget allocated to these priority sectors was flat as a percentage of GDP during the period, growing in real terms. The program directly contributed to improving public sector performance, the third objective. The procurement process was enhanced, and corruption appears to have subsided. Public Financial Management exceeded the minimally acceptable level for general budget support, and important aspects-including cash management and policy- based budgeting- were strengthened. With regard to the fourth objective--the strengthening ofenvironmental management through policy dialogue-the series included neither triggers nor prior actions. The Performance Assessment Framework to which the PRSC was aligned included environmental issues and the govenunent, with limited emphasis, took some action on this front. But the PRSC put little if any emphasis on the topic, without defining the nature of the expected outcome of its support or tracking the "monitorable actions" identified in the program documents. Overall, the PRSC program yielded steady progress during its implementation, achieving important outcomes in macroeconomic management and public sector governance, and laying the groundwork for improvements in agricultural productivity and the business environment. The dialogue centered on annual public expenditure reviews authored jointly by the government and the Bank in which key govenunent officials and Bank staff met every two weeks. While the intensity of the process had high transaction costs, it was considered valuable by government officials and Bank staff because it increased ownership at higher levels of govcnunent as well as in the Bank, and it ensured that all partners were up to date on the program's progress. The outcome of the first series (PRSCs 1-3}-an aggregate rating of the relevance of objectives, relevance of design, and achievement of the objectives- was moderately satisfactory. Although broadly formulated, relevance of objectives was substantial and well-aligned with the govenunent Bank strategies, with an appropriate focus on growth, governance, and income poverty reduction to complement the programs of other development partners. Relevance of program design was modest because while it appropriately focused on improving public sector performance, a neglected cross-cutting issue, it lacked a functional results framework thalt articulated how its activities would lead to desired outcomes. The objectives of supporting results orientation to public sector delivery and of enhancing public sector performance were substantially achieved, while the objectives of sustaining and accelerating economic growth and broadening its impact and of strengthening environmental management were modestly achieved. Risk to development outcomes was significant because many of the program achievements were one-time actions that would need to be repeated, or the development of frameworks- such a.;; better budget execution and a new system for procurement-that had not yet been fully implemented. Bank performance was moderately satisfactory because of strong analytical underpinnings and close supervision and dialogue conducted as part of the process of preparing annual public expenditure reviews, though the results framework was weak. Borrower performance was also moderately satisfactory because a strong network of highly-committed reformers were able to maintain reform momentum particularly on public sector issues, even as monitoring instruments were weak and ownership was lacking in some key sectors, notably agriculture. Series 2: PRSCs 4-8 (2006-2011, USD 1,025 million) The objectives of the second series were not clearly stated at the outset, but the program put emphasis on the harmonization and ownership goals of the Paris Declaration, and so it was organized around the government's National Strategy for Growth and Reduction of Poverty (known by its Swahili acronym as MKUKUTA). With regard to the first objective, growth and reducing income poverty, GDP growth remained strong throughout, never dipping below 6 percent per annum even in the face of the global fmancial crisis. Reliable data are not available regarding the trend in income poverty. The 2007 household survey showed a very low elasticity of poverty with respect to GDP growth, a circumstance that has likely persisted. While three-quarters of Tanzania's poor derive their income from agriculture, agricultural productivity improvements were marginal. Attribution of the strong growth performance to the PRSC program is difficult, because implementation of relevant portions of the program were generally limited. The agriculture sector development strategy was severely delayed. Slow progress on business environment reforms, even the risk of their reversal, suggests that few new income- generating opportunities for the poor were created in rural areas. And progress in· improving reliability, quality, and access to electricity and improving transportation networks--considered critical steps to accelerating private sector growth- was modest. There was modest progress in improving quality oflife and social well-being. Earlier gains in access to education at both the primary and secondary levels were maintained. However, evidence on quality-as measured by literacy and numeracy of schoolleavers- is mixed and suggests learning achievements have lagged. Key outcomes in health-including infant and under-5 mortality, AIDS prevalence, and child nutrition--continued to improve. Survey data suggest an improvement in access to clean water in rural areas from 45 percent to 58 percent from 2007-2010 (still short of the 65 percent target), but backsliding in access to basic sanitation facilities. The program's contribution to these outcomes was tenuous. It made important contributions to improving the process-helping to set standards for assessing the sector review process in the first year or two-but thereafter added little value beyond what was already occurring at the sector level. Governance and accountability reforms covered wide-ranging topics with modest results. Little progress was made with respect to budget and public financial management issues according to available indicators and counterpart feedback. In civil service reform, the adopted strategy relied on across-the-board pay enhancements, despite evidence suggesting that payments targeted to areas of need are more likely to enhance performance. Although more money was disbursed to local goverrunents, little authority was granted for human resource management. Compliance with national procurement regulations rose from 39 to 70 percent and the Public Expenditure and Financial Accountability score rose substantially on this dim ens~on, but external assessments questioned the integrity of the procurement selection process at national and local levels. Corruption perception indicators suggest the government's anti-corruption measures had little traction. Overall, the PRSC program stalled abruptly around 2006/07 as the stage of the reform process moved from policy design to implementation. This change in outcomes coincided with three interrelated dynamics. First, changes in personnel across development partners resulted in multiple, and hence less coherent, perspectives of the reform agenda. Second, some development partners deemed the Public Expenditure Review process too Bank- centric, while new leadership in the Bank and government viewed it as too time intensive. A joint annual review of general budget support, comprising each development partner participating on an equal footing irrespective of their financial contribution, became the primary vehicle for dialogue, diluting trust and communication with government partners especially around public financial and public sector management issues. Third, the series covered many more sectors, greatly expanding the potential number of government counterparts, many of whom had limited interest in, and commitment to, the PRSC agenda. The outcome of the second series (PRSC 4-8) was moderately unsatisfactory. Objectives were aligned with the CAS, MKUKUTA, and the programs of other donors, and their relevance is rated substantial. Program design adhered to Paris Declaration principles but had shortcomings, including a lack of selectivity and a delay in making the results framework fully operational until the third operation, and was modestly relevant. All three objectives were modestly achieved. Risk to development outcome is significant because of weakening government commitment to both public sector r~fonns and sector-specific policies (although the Bank reports that the intense process for dialogue around core public sector issues has been revamped). Bank performance was moderately satisfactory due to solid grounding in analytical work and adherence to the Paris Declaration-inspired approach of the day. But the results framework linking support to expected outcomes, largely shared by the government and development partners, was weak. Borrower performance was moderately unsatisfactory because of a weakening of government commitment, and because policies agreed to by central ministries were not pursued with vigor by relevant line ministries. Lessons The implementation ofPRSCs 1-8 spanned nearly a decade during a period of sustained macroeconomic stability and fast economic growth, offering lessons both for Tanzania and other countries with similar development challenges. • Direct and sustained focus and clarity of vision over the entire course of implementation of PRSC programs that cover many years is required to facilitate lasting reforms, particularly with regard to public sector governance. • PRSCs can make little progress without a foundation of trust between the Bank and the government, and among development partners at the institutional and individual levels. • PRSC programs risk becoming irrelevant if the design, scale, and scope are not adjusted as conditions change. • The lack of well-defined and realisti.c objectives and a fully developed results framework makes adjustment to changing circumstances difficult. • Regular reporting on key outcome indicators is essential so that data can drive adjustments to the PRSC program. • Transitioning to the use of PRSCs as the primary instrument for resource transfer is only likely to be successful in rare circumstances, when strong central ministries can fully pursue a sector-inclusive dialogue representing the perspectives of key players throughout government. • Elevating to the national dialogue sector-specific structural reforms that stalled at the sector level does not guarantee their implementation. ~~£~ Caroline Heider Director-General Evaluation 1. Background and Context 1.1 The United Republic of Tanzania (Tanzania) has a per capita income estimated at USD509 per year (2010, Atlas Method), land area of 886,100 square kilometers, and projected population of about 43.2 million (2010), growing at about 2.5 percent a year. The economy is heavily dependent on agriculture, which accounts for 50 percent of gross domestic product (GDP) and provides 85 percent of exports and 80 percent of employment. Agricultural products are primarily maize, paddy, coffee, cotton, tea, cashew nuts, sisal fiber, and tobacco. Industry accounts for some 15 percent of GDP and is mainly limited to processing agricultural products and light consumer goods. The mining sector has great potential, yet to be fully and carefully developed. Tourism (wildlife and coastal) also has tremendous potential and has become one of Tanzania's most dynamic sectors, showing significant growth in recent years. The service and informal sectors are growing rapidly and are increasingly important sources of employment. 1.2 Economic and Political Background for PRSCs 1-3 (2003-2005). Economic policies following independence in 1961 proved damaging as the country‘s rural development strategy for self-reliant villages resulted in drastic decline in food production. Beginning in 1967, the government adopted the ‗Arusha Declaration‘ as a blueprint for the country‘s development. It involved the nationalization of most of the economy, as well as the resettlement of villages into larger communities in the countryside in order to improve their proximity to services. This policy stance, based on Ujamaa (or African socialism) was supported by many donor countries as well as multilateral institutions, resulting in considerable inflows of aid. However, the Ujamaa experiment failed to deliver the promised rapid social advancement. This was compounded by an inefficient public sector: overstaffed state agencies in the banking, industrial and farming sectors were allocated a significant share of budget resources at the expense of the social services. In addition, external shocks throughout the 1970s led the country into economic crisis by the early 1980s. The war with Uganda and the second oil shock, both at the end of the 1970s, led to a severe resource shortage. 1.3 The economic policies pursued under the Arusha Declaration were issues of much debate in the early 1980s, with policymakers divided between those supporting and opposing market reforms. In the political sphere, Julius Nyerere, Tanzania‘s leader since independence, retired as President in 1985 but continued to be the head of the ruling party, Chama Cha Mapinduzi. Ali Hassan Mwinyi, the second president, initiated an economic recovery program during 1986–89, which was accompanied by political changes in the late 1980s and early 1990s. During his second term of office, however, President Mwinyi‘s government demonstrated a much lower commitment to economic reform. As a result, overall macroeconomic performance deteriorated and jeopardized the economic recovery plan‘s achievements of the late 1980s and early 1990s. 1.4 In addition, a confluence of factors—poor economic performance and policies, and deteriorating governance—led to a breakdown in the relations between Tanzania and its external partners in the mid-1990s. In mid-1994, a group of ―Independent Advisors on Development Cooperation Issues between Tanzania and its Aid Donors‖ chaired by 1 Professor Gerald Helleiner, was formed to assess whether development cooperation could be strengthened. The Helleiner Report (Helleiner and others, 1995) led to a major revamping of relations with development partners, organized around the principle of ‗mutual accountability.‘ With donor rapprochement in the wake of the Helleiner Report and the election in 1995 of Benjamin Mkapa, as Tanzania‘s third president, the government initiated another economic recovery plan in 1996. There was a strong commitment to improving fiscal performance and instituting structural reforms to enable both accelerated growth and poverty alleviation, and a new dynamic of donor engagement including IMF and World Bank support. During this period the government placed reformers in key leadership positions in the Ministry of Finance and elsewhere. In April 2000, the International Development Association (IDA) and the International Monetary Fund (IMF) supported a comprehensive debt reduction package for Tanzania under the enhanced Heavily Indebted Poor Countries (HIPC) Initiative (IDA/IMF, 2000) that cancelled outstanding debts to the two institutions1, as noted in Table 1-1. 1.5 The government‘s program was laid out in a number of new strategy documents, including the Development Vision 2025, National Poverty Eradication Strategy, Tanzanian Assistance Strategy, and the Public Expenditure Review/Medium Term Expenditure Framework, jointly produced by the government and Bank.2 The Development Vision 2025 document updated Tanzania‘s national vision, first expressed in the Arusha Declaration. It projected a future free of poverty, with high levels of human development, and Tanzania graduating to a middle income status by 2025. The National Poverty Eradication Strategy was more specific in defining goals and strategies in the areas of economic growth, income levels, primary education, literacy, access to water and sanitation, disease burden, mortality and health, employment, housing and infrastructure, some with quantitative indicators and targets specified. The Tanzania Assistance Strategy set out to ―restore local ownership and leadership by promoting partnership in the design and execution of development programs‖ following the rapprochement with development partners in 1995. The process for preparing annual public expenditure reviews became the main avenue for policy dialogue and strategy formulation among domestic stakeholders and with development partners before the poverty reduction strategy paper (PRSP) process was launched.3 1.6 The economy appeared to respond swiftly to the new reforms, with an average real GDP growth rate of 5 percent from 1999 onwards. This was a significant improvement compared to the less than 3 percent growth realized in the early 1990s which had resulted from the constraints posed by falling commodity prices, the slowdown in the world economy, and declining world tourism. Simultaneously, the government dealt with many of the 1 Assistance depended on several conditions, including the adoption and implementation of a participatory poverty reduction strategy (IMF, 1996). 2 In addition, an Anti-Corruption Strategy which emphasized increased transparency, accountability and governance, and an External Debt Strategy aimed at preventing the country from reverting to serious external debt problems post-HIPC, were also introduced. 3 The PRSP became a required input for accessing financial support from both the Bank (IDA) and the IMF, beginning1999 (Independent Evaluation Office and Operations Evaluation Department, 2004). 2 distortions that existed in the 1990s—markets became freer and the public sector smaller. Inflation also declined from over 30 percent in 1995 to 5 percent in 2000. Prospects for improved real GDP growth for the medium term increased, with agriculture, mining and the tourism sectors becoming the driving forces as their shares of the growth rate increased. Table 1-1. Time Line of Events Including the Implementation of PRSCs 1-3 Year Principal Government of Tanzania Events Key Development Partner Events 1995 First Multi-party Election 1996 Economic Recovery Program (FY96-00) IMF Enhanced Structural Adjustment Facility 1997 Public Expenditure Report (PER) carried out IDA First Structural Adjustment Credit (SAC-10) annually since then 1998 Multilateral Debt Relief Fund established (July) 1999 Launching of participatory PRS by Tanzania/ Tanzania declared eligible for relief under Interim-PRSP prepared enhanced HIPC Initiative (November) 2000 Full PRSP (FY02-04) IDA Programmatic Structural Adjustment Credit HIPC decision point reached (March) (PSAC) World Bank CAS (2000-03) endorsed by Board 2001 First PRS Progress Report Harmonized Donor Support mechanism (PRBS) HIPC completion point reached (November) IDA Primary Education Development Program Household Budget Survey as basis for Poverty (PEDP) Report of 2002 Country Financial Accountability Assessment carried out. Agriculture Sector Development Strategy 2003 Country Procurement Assessment Report IDA First Poverty Reduction Support Credit undertaken. (PRSC-1) –May IMF Poverty Reduction and Growth Facility-July 2004 IDA Second Poverty Reduction Support Credit (PRSC-2) 2005 IDA Third Poverty Reduction Support Credit (PRSC-3) 2005 Paris Declaration Sources: Various Bank, IMF and government documents. 1.7 The Bank‘s program assistance to Tanzania prior to the PRSC series included a sequence of structural adjustment credits, combined with sector investment and sector adjustment lending and analytical and advisory services. In the social sectors, Bank assistance included a human resources development pilot project which closed in 2005, an adjustment lending operation for primary education executed between October 2001 and October 2004 and another adjustment operation supporting secondary education implemented between June 2004 and December 2007. In the health sector, the Bank contributed to the health sector multi-donor ―basket fund‖—in which the government pools funds from multiple partners in support of a broad sector program—through two adjustable program loans (APLs), the Health Sector Development I and II with two rounds of additional financing between 2000 and 2009. The basket fund supported specific sector reforms through the line ministry‘s budget and also provided significant funding for non-wage expenditures at both national and local levels. A multi-sector HIV/AIDS project supported Tanzania‘s efforts to reduce HIV transmission and to mitigate the adverse consequences of AIDS. A project in the water sector supported technical, commercial, and financial rehabilitation of the water supply and sanitation services in Dar es Salaam. A second Tanzania Social Action Fund 3 project was approved on November 30, 2004, as a follow-up to continue with the successful outcomes in supporting social protection and inclusion at the community level. 1.8 Other sector projects complemented the Bank‘s policy dialogue initiated under the poverty reduction support process. The Agriculture Sector Development project (1998-2004) was deepened by the Participatory Agricultural Development and Empowerment Project Credit, which was approved by the Bank‘s Board in May 2003 to support investments in appropriate technologies to reduce soil fertility decline. Efforts to improve Tanzania‘s infrastructure were supported through projects in the road and railway sectors, as well as through an urban rehabilitation project. In the power sector, the Bank helped the government implement the Power Sector Restructuring Program. It encouraged the government to develop the domestic gas market and to generate lower-cost power through the Songo Songo Gas Development and Power Generation Project. The Bank also provided support under the Local Government Support Project to strengthen fiscal decentralization, improve accountability in the use of local government resources, and improve the management of intergovernmental transfer systems. 1.9 Economic and Political Background for PRSCs 4-8 (2006-2010). The second series of the PRSCs were initiated under the economic momentum built by the economic recovery plan, and continued through the first PRSC series. Economic reform in Tanzania had started to yield positive results during the late 1990s and the early 2000s. GDP growth accelerated, resulting in around 4 percent per capita per annum growth since the early 2000s (see Table 1- 2). Growth was accompanied by higher investments and a favorable external environment with strong prices for commodities exported by Tanzania. Economic performance remained strong even during the subsequent global financial crisis, beginning in 2008. But the improved growth during the decade was not translated into a significant decline in income poverty. Available poverty data4 indicate that there was only a marginal change in poverty between 2000 and 2007, from 35.6 percent households below the national poverty line to 33.4 percent.5 In addition to the weak income poverty response to growth, household consumption data also indicated that the poorest households had little gains in consumption between 2000 and 2007, while the top deciles experienced consumption growth well above the average, thus indicating rising inequality. The modest reduction in income poverty surprised government officials and development partners, some of whose attention was devoted to attacking or supporting the data. The ensuing rethinking of the programs of development partners, including the Bank, and the government‘s ongoing defense of its development strategy became a factor in eroding trust and confidence among the government and development partners according to multiple interviews. 1.10 On the other hand, the strong growth and foreign aid flow increased resources available to the government, which translated into sizable gains in access to education and 4 The most recent poverty data are based on the Household Budget Survey of 2007 (See United Republic of Tanzania, 2011). 5 The Government first poverty reduction strategy paper (PRSP) adopted in 2000 projected that GDP growth of around 4 percent could likely halve Tanzania‘s level of poverty by 2010, but the elasticity of poverty reduction to growth has been much weaker than expected (World Bank, 2008b). 4 health services. During the 2000s, Tanzania enjoyed improved health outcomes such as increased life expectancy largely due to vaccination campaigns and a reduction in malaria, and apparent reduced maternal mortality towards the end of the decade. These resulted in substantial increase in Tanzania‘s composite health index of United Nations‘ Human Development Index. Remarkable gains were also achieved in education, especially in terms of average years of schooling, although slow income growth translated into very little success in reducing malnutrition faced by the poorest households. Table 1-2. Tanzania – Selected Economic and Social Indicators, 2000-2010 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Output and Prices Real GDP growth (%) 4.9 6.0 7.2 6.9 7.8 7.4 6.7 7.1 7.4 6.0 7.0 Real per capita GDP growth (%) 2.3 3.3 4.4 4.0 5.0 4.5 3.8 4.1 4.4 3.0 3.9 Consumer prices (end of period) 5.9 5.1 5.3 5.3 4.7 5.0 7.3 7.0 10.3 12.1 6.2 External Sector Export, f.o.b ($m) 1,361 1,767 1,900 2,169 2,615 2,972 3,446 4,102 5,578 5,149 6,388 Imports, f.o.b. ($m) 2,050 2,210 2,144 2,659 3,458 4,205 5,113 6,274 8,661 7,543 8,975 Real effective exchange rate, 100.0 92.4 89.4 94.5 100.5 100.8 87.3 2004=100 Public Finance (% of GDP) Revenue (excluding grants) 12.0 12.1 12.1 12.7 13.6 12.6 14.4 15.9 16.2 15.8 16.1 Total grants 3.7 4.5 6.2 6.1 7.8 5.4 5.0 6.9 5.1 4.6 6.4 Recurrent expenditures 12.8 13.6 14.8 16.4 17.7 17.3 17.3 14.9 17.7 18.8 18.1 Development expenditures 3.7 3.4 5.0 5.6 7.3 5.7 6.2 8.0 8.4 8.6 10.9 Overall balance -1.6 -1.1 -1.6 -3.2 -3.5 -5.0 -4.1 0.0 -4.8 -7.0 -6.5 Savings and investment (% of GDP) Investment 16.8 17.4 19.1 20.3 21.7 22.9 24.2 25.5 26.7 29.8 30.6 Gross Domestic Savings 10.1 13.2 16.9 16.1 15.2 14.0 11.0 12.6 10.3 17.9 16.8 Balance of payments (% GDP) Current Account Balance -4.2 -3.8 -1.0 -1.1 -4.1 -7.8 -7.9 -11.0 -12.9 -9.0 -8.6 Foreign Direct Investments 4.5 3.7 3.7 3.1 1.8 6.6 2.8 3.5 1.9 1.9 1.9 External debt stock 70.1 62.3 65.7 62.4 66.9 59.1 28.3 29.8 28.8 34.3 37.6 Memorandum Items: CPIA rating 3.9 3.9 3.9 3.8 3.8 3.8 Bank Disbursements ($m) 142 119 148 397 343 275 416 475 394 608 653 Bank Disbursements (% GDP) 1.4 1.1 1.4 3.4 2.7 1.9 2.9 2.8 1.9 2.8 2.8 Human Development Index 0.36 0.42 0.43 0.44 0.45 0.45 0.46 Sources: IMF, World Development Indicators, Tanzania’s PRSP and MTEF, staff’s calculations 1.11 During the period 2000-2010 the Government was highly reliant on foreign aid (Figure A). Official Development Assistance averaged 13 percent of GDP, with general budget support (GBS) providing roughly one-fourth of this (4 percent of GDP) in the years for which data is available. The World Bank's PRSCs accounted for roughly one quarter of 5 GBS. In recent years, development partners, including the Bank, have begun to scale back resource transfer through general budget support. Persistent heavy reliance on GBS has provoked questions from the Bank (World Bank 2012a) and development partners about the sustainability of official development assistance as a contributor to Tanzania‘s growth. Figure 1-1. Trends in Government Expenditure and Development Assistance. 35 30 25 Percent of GDP 20 15 10 5 0 2003 2004 2005 2006 2007 2008 2009 2010 Government Expenditure Net ODA GBS PRSC Source: IMF, World Development Indicators, Tanzania Development Partners Group, World Bank Business Warehouse database 1.12 Overall, public expenditures on education and health increased from 3.6 and 1.5 percent respectively of GDP in 2000-02 to 6 and 3 percent of GDP respectively by end of the decade. The introduction of the Poverty Reduction Strategy Paper (PRSP) and the PRS process in early 2000s helped to streamline and prioritize donor assistance to Tanzania. Relationships between Tanzania and its development partners were shaped also by the principles developed under the multi-donor budget support mechanism, which resulted in the share of budget support in total donor funding to increase from 33 percent in 2003/04 to about 51 percent by 2007/08 and stabilize thereafter at that level (Ministry of Finance, 2008). 1.13 The Bank has been an important partner in Tanzania‘s economic reform process during the second half of the decade as in the first half (Table 1-3) with increased financial and technical support. By 2004 outstanding credit to the Bank reached around USD4 billion, of which about 75 percent was written off as part of the HIPC process. As of 2011 the outstanding debt to the Bank was about USD3 billion. 1.14 Overall, the effort by the Bank to transition from traditional adjustment lending to poverty reduction support was motivated by the goal of promoting country ownership of the reform process and donor harmonization (as per the emerging principles of the Paris Declaration).6 Indeed, Tanzania and development partners were on the leading edge in 6 The principles of the Monetary Consensus on Financing for Development (2002), the Rome Declaration on Aid Harmonization (2003), the Marrakech Memorandum on Managing for Results 6 implementing the harmonization agenda. While the previous Bank‘s adjustment lending relied primarily on externally monitored, ex-ante policy conditionality, the PRSC sought to reinforce government ownership and the participation of local stakeholders by promoting the use of country processes for policy design, implementation, monitoring and evaluation. The PRSCs also provided added flexibility for adjusting the direction of policy reform in response to changing circumstances. Importantly, the second PRSC series put major emphasis on the harmonization goals of the Paris Declaration. The Bank‘s country strategy served as an annex to a joint partner strategy—the Joint Assistance Strategy for Tanzania— and the Bank entered into a new arrangement for dialogue with the government that was more inclusive of other development partners. Table 1-3.Time Line of Events Including the Implementation of PRSCs 4-8 Year Principal Government of Tanzania Events Key Development Partner Events 2005 Public Expenditure Review Union Elections East African Community Custom Union Paris Declaration 2006 Public Expenditure Review PRSC-4 Procurement Act Agriculture Sector Development Program 2007 Public Expenditure Review PRSC-5 Household Budget Survey CAS (2007-2011) Country Economic Memorandum and Poverty Assessment 2008 Public Expenditure Review PRSC-6 Accra Accord 2009 Public Expenditure Review PRSC-7 PRSC-7 Supplement 2010 Public Expenditure Review PRSC-8 Sources: Various Bank, IMF and government documents. 1.15 The PRSC was the means by which the Bank coordinated with other donors in including its policy measures in the existing general budget support policy matrix, the performance assessment framework (PAF). GBS was initially an outgrowth of the poverty reduction strategy process and dialogue surrounding the HIPC initiative. The PRSC began outside of GBS, which was at the time focused on priority sectors and the allocation of budget resources. While the Bank gave this aspect of GBS some degree of attention, it viewed budget support primarily as an instrument to support growth and poverty reduction. Other development partners were not in universal agreement regarding this adjustment, but agreed to modify the PAF so the Bank could participate in joint GBS. This tension between the Bank and several other development partners persisted, and the results of the 2007 household budget survey that showed a weak growth-poverty linkage became a source of contention, with the Bank left in somewhat of a defensive position. Moreover, different partners preferred different strategies for support—with varied focus on processes, inputs, (2004) leading to the Paris Declaration on Aid Effectiveness (2005) laid down a roadmap to improving the effectiveness of aid by emphasizing harmonization, alignment, country ownership and mutual accountability. 7 and outputs/outcomes—and so general budget support retained these multiple approaches in spite of a common PAF. 1.16 Under GBS, the dialogue between the government and development partners was heavily and increasingly driven by process—with the creation of an elaborate system of working groups and annual reviews of sector progress. At the outset, this emphasis was useful and according to multiple interviews it helped to re-establish trust among partners, keeping all parties informed and allowing broad participation. The government was able to use the GBS process to coordinate disparate agendas among donor partners; donors were able to ensure access to top policymakers. But starting around 2007, process came to dominate content as development partners cut back on technical staff in the field and a less reform- minded government came to power. The increased focus on process led skeptics, including on the Bank‘s country team, to become disenchanted with GBS and the PRSC instrument. Some viewed the government as intent on doing only the minimum required, while other development partners saw the Bank as trying to undermine the process. This disillusionment with GBS led partners to abandon their early enthusiasm to use the instrument as the primary vehicle for development support, with plans to use it in a more modest fashion for the foreseeable future. 2. Series 1: PRSCs 1-3 Objectives, Design, and Relevance 2.1 Objectives. The program objectives for PRSCs 1-3 were first articulated in the program documents for PRSC-2, and repeated in PRSC-3. This evaluation considers these objectives, specified below, as those of the full series:  Sustaining and accelerating economic growth and broadening its impact;  Supporting results orientation of public service delivery;  Enhancing public sector performance; and  Strengthening environmental management. 2.2 The objectives of the series were not settled until the second operation. In the program document for PRSC-1, objectives were stated as: (i) promoting accelerated growth; (ii) broadening the impact of growth; (iii) aligning public investment with [Poverty Reduction Strategy goals; and (iv) increasing the efficiency of public service delivery. Program documents for PRSCs 2-3 adjusted and condensed these objectives, and added an objective on environmental management. 2.3 Complementing PRSC-1, which focused on private sector development and public sector management, PRSC-2 aimed at reducing poverty, improving M&E, sustaining macroeconomic stability, improving the effectiveness of service delivery, and fostering environmental sustainability. The last operation of the series, PRSC-3 focused on enhancing the reforms targeted by the first two operations, leveraging progress in the implementation of sectoral programs to reduce poverty (especially primary education, basic health, water, rural roads, agricultural research and extension, the judiciary and HIV/AIDS), facilitating actions 8 to strengthen public sector performance and enhance efficiency and effectiveness of the use of resources, supporting government's efforts to promote environmental sustainability of Tanzania's development program, and strengthening the alignment of the PRSC to government strategy.7 2.4 Relevance of Objectives. The Bank‘s PRSC objectives were built on the country assistance strategy (CAS) of 2001-03, which was extended to 2006, and had anticipated development policy lending in 2003. This strategy focused on creating higher growth and economic opportunities for the poor as well as helping build country capacity, empowerment and accountability. It was fully consistent with Tanzania‘s own poverty reduction strategy (United Republic of Tanzania 2000), whose two pillar approach to tackling income poverty and supporting human capabilities, survival and well-being included action on maintaining macroeconomic stability, rural sector development and export growth, private sector development, education, health, social well-being, vulnerability, and the environment. The PRSC focus on governance, growth, and income poverty reduction supplemented the focus on human development services in the programs of other development partners. 2.5 The series also built on a prior operation—the Programmatic Structural Adjustment Credit (PSAC), which supported the government‘s efforts to sustain macroeconomic stability, strengthen financial accountability, deepen the implementation of the privatization program, improve competitiveness and transparency of markets, and further reduce barriers to internal and international trade (World Bank, 2000). The program document for PRSC-1 explicitly noted that the operation would focus on private sector development and public sector management to complete and deepen the reform agenda implemented under the PSAC, and lay the ground for attacking income poverty more aggressively. The scope of the PRSC series was country-wide and intrinsically aligned to the government‘s PRSP and the Bank‘s Country Assistance Strategy (CAS). In this respect, the objectives of the PRSC were relevant in helping deal with the constraints to growth and poverty reduction which were the concern of the government‘s program. In addition, the focus of the program reflected the lessons from previous reforms initiated under the Tanzania poverty reduction strategy and supported by the Bank under the PSAC. While the range of the objectives of the supporting government program was very broad, the focus of the PRSC concentrated on public sector issues, an area in which other development partners were less involved. Relevance of the objectives of PRSCs 1-3 is therefore rated substantial. 2.6 Design. The PRSCs identified important policy areas to achieve their objectives. For the first objective on growth and broadening its impact, the program focused on macroeconomic stability, rural development and export growth, and private sector development. For the second objective on results orientation of public service delivery, the PRSCs were to monitor resource allocation to, and outcomes in, the poverty reduction strategy pillar covering enhancing human capabilities, survival, and well-being. The series supported reforms in public expenditure management, which could help the country to better monitor resource allocation. For the third objective on enhancing public sector performance, 7 In addition, over time the PRSC was increasingly integrated into the Poverty Reduction Budget Support (PRBS) framework of other development partners (Refer to the Partnership Framework Memorandum, Annex 7 of World Bank, 2005). 9 the PRSCs sought to improve the effectiveness of public services, minimize resource leakage, and strengthen accountability. The program focused on most of the important policy areas, including public financial management, civil service pay, procurement, and to some extent decentralization. For the fourth objective on strengthening environmental management, the PRSCs were to support the mainstreaming of environmental concerns into the poverty reduction strategy, budget process, and sectoral policies, improve the understanding of poverty-environment linkages, and strengthen institutional capacity to integrate environmental assessments into sectoral strategies, policies, and activities as the district and local level. 2.7 Relevance of Design. The policy areas selected for inclusion in the PRSCs were largely appropriate. On accelerating growth and broadening its impact, a focus on macroeconomic stability, rural development, and private sector development made sense: more than 80 percent of Tanzania‘s poor live in rural areas and three-quarters derive their income from agriculture (United Republic of Tanzania 2009); private sector development particularly in rural areas could enhance non-agriculture employment opportunities for the poor; and macroeconomic stability was a precondition for strong growth. The policy areas under the third objective could also contribute directly to enhancing public sector performance. 2.8 Linkages of the policy areas supporting the second and fourth objectives were less clear. There were no specific policy actions related to improving data collection or analysis. Support for the fourth objective was not anticipated until the second of three operations and only delivered through dialogue. The intention appeared to be to introduce the topic to the national dialogue rather than influence policies directly, but the program documentation does not make the case that the dialogue would, or could, lead to the outcome desired. 2.9 Many of the policy actions in the first series were ‗low hanging fruit‘, actions that were relatively easy to complete due to political consensus, but which would require a number of follow-on actions before development outcomes could be improved. Only nine of the 28 prior actions in the first series involved implementation of specific reforms (which could plausibly contribute to achieving the program‘s objectives), while 13 related to drafting and approving legislation or executing the budget in line with poverty reduction strategy priorities, and six involved developing frameworks for future reforms or reporting on progress. 2.10 The series lacked a functional results framework. Objectives for the series were not settled until the second operation, and program documents did not clearly articulate how progress in adopting the proposed policy measures would help to achieve the program‘s objectives. Outcome indicators were inconsistent for each operation and for the overall series. For several objectives, it was not obvious that these policy changes would contribute much to the desired outcomes, with prior actions and proposed triggers often preliminary in nature. This evaluation reconstructs a plausible route by which PRSC-supported activities could have been expected to contribute to achieving the stated objectives. 2.11 Overall, due to the relevance of the program‘s policy areas to its objectives but the weak results framework, the relevance of program design is rated modest. 10 Implementation 2.12 Table 2-1 summarizes the approval, effectiveness, closing dates and size of each of the three PRSCs. All three operations were co-financed by the Kreditanstalt für Wiederaufbau (KfW, Germany) and complemented by the GBS partners (African Development Bank, European Union (EU), Canada, Denmark, Finland, Japan, the Netherlands, Norway, Sweden, Switzerland, Ireland and the U.K. Department for International Development (DfID)). All three PRSCs were consistent with a performance assessment framework (PAF) common to all GBS partners—indeed, the PAF was adjusted and made more complex to accommodate them. Triggers were developed in close cooperation with government counterparts. None of the three PRSCs was restructured, and there was no mid-term review for an individual operation or the series as a whole. The actual amount of PRSC-3 was USD150 million rather than the USD175 million which was originally envisaged in the CAS as development policy lending. Budget support was expected to gradually become the primary vehicle for resource transfer from IDA to Tanzania. Because of delays in the implementation of agriculture sector reforms (see below), and in order to maintain consistency between IDA's PRSC financing and the pace of structural reforms, the USD25 million increase in PRSC financing was postponed until the follow-up PRSC-4 (the first credit in the second series of PRSC support). Table 2-1.Timing and Scale of PRSCs 1-3 Operation Proposed Actual Approval Effectiveness Closing Delays Implementing amount amount date date date agency Ministry of PRSC-1 $132m $132m 05/29/2003 07/18/2003 06/30/2004 None Finance Ministry of PRSC-2 $150m $150m 07/29/2004 09/15/2004 06/30/2005 None Finance Ministry of PRSC-3 $175m $150m 09/08/2005 10/20/2005 6/30/2006 None Finance 2.13 Within the context of the regular supervision by Bank staff, PRSC implementation was supported by ongoing IDA operations—for example, a FY1999 tax administration project, a FY2000 public sector reform program, and a FY2003 participatory agriculture development and empowerment project. The government also coordinated implementation of its program through sector working groups, which were set up as instruments for dialogue and monitoring of the PRSP for all stakeholders. The coherence of the government‘s policy framework in the context of the first PRSP8 facilitated the yearly selection of triggers, while the ex-post nature of financing allowed for a more flexible matching of PRSC policy areas with the government policy choices, although various program documents of the PRSC series noted that perceptions of implementation speed on some actions diverged considerably between the government and Bank staff. Slow implementation was attributed by Bank staff to capacity limitations, although some development partners consulted pointed to conflicting political commitment and personalities in government as the program evolved. Several 8 The Government‘s second PRSP, subsumed under the broader policy framework of MKUKUTA was issued in 2005 (United Republic of Tanzania, 2005a). 11 government officials noted the Bank was slow to provide no objection when necessary, but these complaints appeared to be limited to sector investment lending rather than the PRSCs. 2.14 Performance of the PRSCs was formally monitored in the context of the GBS annual review process, which involved a mid-year review around March-April and an Annual Review in October, with inputs from various parts of government, civil society and the GBS partners as organized through the technical committees and working groups. A critical underlying instrument that facilitated the monitoring of PRSCs 1-3 was the Public Expenditure Review (PER) process which pre-dated the PRSCs. This process involved bi- weekly meetings between the Bank and government officials as a means to co-author an annual PER, and associated Poverty and Human Development Reports. But the dialogue reached far beyond the purview of the reports themselves, covering many of the critical development issues Tanzania was facing. According to multiple government officials and Bank staff interviewed, this intensive process created a degree of trust on a technocratic level among Bank representatives and key technocratic reformers in the government. It was helpful that the Bank‘s Lead Economist in the country became the Governor of the Central Bank during the series, solidifying a technical client relationship. 2.15 The data generated by the two set of documents allowed for some evidence-based assessment of progress of the PRSCs, although some development partners9 have noted that the government commitment to the PER and associated MTEF seemed to have begun weakening by the end of the series. Overall, the results framework for PRSCs 1-3 was rudimentary and accounted for the major effort made during the follow-up series to strengthen the M&E system. Achievement of the Objectives 2.16 In the absence of a robust results framework that organizes policy areas—and their associated triggers and prior actions—around the series‘ four objectives, this evaluation reconstructs the most plausible route by which the activities undertaken by the operations could have been expected to contribute to achieving the stated objectives. Indicator areas were identified in the program document for PRSC-2, but without specific indicators or targets. Trends for key outcome indicators for the four objectives, as identified by this evaluation, are summarized in Table 5. Analysis on the achievement of each objective follows, with greater detail provided in Annex C. OBJECTIVE 1: SUSTAINING AND ACCELERATING ECONOMIC GROWTH AND BROADENING ITS IMPACT. 2.17 Results for sustaining and accelerating economic growth and broadening its impact were mixed.10 Economic growth was robust during the period 2003-2005, with real GDP growth ranging between 6 percent and 8 percent per annum, more stable than other strong 9 See comment by KfW on the Implementation Completion and Results report for PRSCs 1-3. 10 ―Broadening its impact‖ is taken to mean that growth is shared with the poor, and a reduction in income poverty is taken as the key associated indicator. 12 performers in the region and significantly faster than the region as a whole, as shown in Figure 2.1. From the Bank's own analysis, much of this growth (and that in subsequent years) was fuelled by fiscal expansion supported by massive aid inflows (World Bank 2012a, 'Tanzania Economic Update'). Over 2000-2005, government spending mostly financed through increased aid flows contributed 3.8 percentage points of the 6.8 percent annual growth. Growth otherwise would have remained at levels experienced during the 1990s, calling into question whether reliance on this external support was contributing to the goal of sustainable growth. Figure 2-1. Real GDP Growth in Tanzania and Comparator Countries 16.0% 14.0% 12.0% Uganda 10.0% Tanzania 8.0% Rwanda 6.0% Kenya 4.0% Sub-Saharan Africa 2.0% (developing only) 0.0% 2000 2001 2002 2003 2004 2005 2006 Source: World Development Indicators 2.18 Moreover, Tanzania's strong growth did not translate into the expected income poverty reduction, a result similar to the disappointing experience in nearby fast-growing sub-Saharan African countries (figure 2.2 shows the relationship between growth and extreme poverty for non-fragile low income countries). Data were infrequently available. But the 2007 Household Budget Survey revealed only a comparatively modest reduction in income poverty (using the government‘s basic needs poverty line) between 2001 and 2007, which fell from 35.7 percent to 33.6 percent. Among rural populations, the poverty headcount fell from 38.7 percent to 37.6 percent, while it fell slightly faster in urban areas (a substantial portion of the overall modest decline was due to urbanization).11 At the country‘s extreme poverty line, the headcount rate also fell modestly from 18.7 percent to 16.6 percent. 11 This is substantiated by a 2005 independent evaluation of general budget support to Tanzania, which concluded that the disappointing outcome on poverty was a result of the fact that ―the structure of growth is skewed towards urban rather than rural areas, towards mining and services rather than agriculture and towards the richer rather than the poorer‖ (Overseas Development Institute and Daima Associates Limited 2005). 13 2.19 When using the much higher international threshold for extreme income poverty of $1.25 per day, there was more progress, with the overall poverty headcount falling from 84.6 percent in 2001 to 67.9 percent in 2007, while the Gini measure of inequality grew from 35 to 38 (PovcalNet). The major discrepancy in measured outcomes was largely a result of a different choice of price index. But even this measure compares unfavorably to other low- income countries given Tanzania‘s fast growth, demonstrating a low growth elasticity of extreme poverty—below -0.5—compared to international benchmarks identified in the literature of -1.6 (Bourguignon 2003) and -2.6 (IEG 2005).12 The elasticity in Tanzania was, however, roughly on par with neighboring Mozambique and Uganda, and was greater than that in Rwanda, each of which experienced fast growth but comparatively low poverty reduction during similar periods.13 Figure 2-2. Real Growth and Poverty in Non-Fragile Low Income Countries during 2000s 12 Weaker poverty 10 response to growth Annual average growth rate Rwanda Mozambique 8 Uganda Tanzania 6 4 Stronger poverty response to growth 2 0 (8.0) (7.0) (6.0) (5.0) (4.0) (3.0) (2.0) (1.0) - 1.0 More poverty reduction -2 Average annual change in extreme poverty incidence, % Source: World Development Indicators Note: Observations include countries with at least two poverty data points at least three years apart between 2000 and 2010 inclusive. Countries above the trendline have experienced a relatively weak extreme poverty response to economic growth, those below a relatively strong one. 2.20 What was the PRSCs‘ contribution to these outcomes? The program document for PRSC-2 outlined three areas through which the series would contribute to the government‘s program: 12 Poverty appears to be less elastic to growth among low-income countries than across all countries, but Tanzania‘s record remains below par in this regard. 13 Mozambique and Rwanda have clearly had disappointing poverty outcomes in the early- and mid- 2000s. Uganda‘s poverty reduction has been steady, but still not as fast as expected given its very fast economic growth. 14  Sustained macro-economic stability, seen as a precondition for growth and implementation of the development strategy, by supporting a broader effort among development partners;  Enhanced opportunities in formal and informal sectors through private sector development and strengthening the business environment; and  Raised agricultural productivity and profitability to enhance incomes of the population in rural areas, where poverty was most widespread and deeply entrenched. 2.21 There was substantial progress in the first area, and modest progress in the second and third. The country was successful in maintaining macro-economic stability, with the Bank‘s contribution coming through a broader effort among development partners led by the IMF. The business environment did not change markedly according to available indicators, and reforms in private sector development were slow. Similarly, reforms in the agriculture sector were delayed and diluted, and growth in the sector does not appear to have translated into higher income among the rural population. 2.22 Monitoring indicators were inconsistent from operation to operation. Many indicators identified in PRSC-1 were dropped, without discussion, in the program documents for PRSC- 2 and PRSC-3. This was because the Bank committed to adhere to the common Performance Assessment Framework, which was developed after the first PRSC program document and utilized a second specific set of indicators. 2.23 The Bank recognized that achieving the overall growth and poverty reduction objectives of the PRSCs required the maintenance of macroeconomic stability. The IMF took the lead on these issues under its Poverty Reduction Grant Facility; its reviews under the Facility served as an input into the Bank‘s assessment of the macroeconomic situation, an implicit condition for the program. The Bank focused on ensuring sustainability of Tanzania‘s external debt following the substantial debt cancelation under HIPC, and the preservation of budget allocated to priority sectors. This was intended to maintain the accelerated economic growth that was seen to be in part a result of improved fiscal performance in the late 1990s, while directing additional resources generated through debt cancellation toward development priorities. The program thus encouraged prudent management and contracting of external debt by emphasizing the adoption of a national debt strategy, amendments to laws and ordinances related to the contracting and procurement of debt, an annual borrowing and repayment plan, stronger controls on contingent liabilities, and the establishment of a national debt management committee. 2.24 Macroeconomic management during 2003-05 overall was strong. The IMF deemed Tanzania‘s structural reform program on track, with the country meeting all but one quantitative indicator for the Poverty Reduction Grant Facility in December 2005. The country‘s growth, inflation, and external position were all considered consistent with IMF program assumptions (IMF 2006). While growth was strong, inflation remained low, hovering around 5 percent per annum. The fiscal deficit grew somewhat 5 percent of GDP in 2005, but even this was in part due to a drop in grant finance. As discussed under the second objective, the budget resources allocated to priority sectors were flat as a share of GDP. Investment continued an upward trend rising from 19.1 percent of GDP in 2002 to 22.9 15 percent in 2005, while gross domestic savings fell modestly. The current account deficit, after hitting a low of -1.0 percent in 2002, rose to 7.8 percent in 2005. 2.25 Reforms in debt management were successful, in part because the PRSCs built on the dialogue around, and overlapped with the final stages of, the HIPC initiative,14 as well as with the multilateral debt relief initiative process which created strong incentives for Tanzania to reform its debt management system and introduce institutions for prudent macro- policies. All envisaged debt targets were met, and the country had significant scope to expand its fiscal space at the end of the period. With debt cancellation under HIPC and the Multilateral Debt Relief Initiative, the debt to GDP declined to 60 percent in 2005, while debt service declined from the peak of almost 40 percent of exports to about 5 percent. The net present value of debt-to-export ratio was estimated to be 64 percent by 2006. These figures were considered well below debt sustainability thresholds (World Bank, 2007e: 14). 2.26 Private sector development was perceived as essential for growth and poverty reduction. The PRSC built on tangible steps taken with support from the PSAC, which included the advancement of the privatization agenda, with government disengagement from commercial activities almost completed and significant progress toward greater private sector involvement in the provision of utilities; progress in strengthening the business environment and reform of land legislation; and the establishment of an improved regulatory environment. 2.27 By the available indicators, the business environment did not improve, suggesting that this was not a major contributor to strong industrial growth. The results from the two investment climate assessments roughly corresponding to the beginning and end of the period (2003, 2006) suggest no major change.15 The PRSCs focused on three primary constraints to private sector development: the lack of a private sector development strategy and small and medium enterprise policy, rigid labor laws, and a need to improve the business environment. The lack of a comprehensive strategic framework to help with coordination and targeting of the various initiatives in support of private sector development was seen as major problem, as earlier efforts had been piecemeal. 2.28 The industrial growth rate for the period 2003-05 was indeed strong, at 10 percent per annum, meeting the targeted range of 8 to 10 percent. This rapid growth, however, was driven by capital-intensive sectors, including for example a strong supply-side response in the mining sector to high commodity prices, and later a slowdown as global aggregate demand subsided in 2008 and 2009 (see Figure 2.2). It was not related to broad economy- wide reforms or growth across sectors targeted by Bank-supported government reforms. Average annual growth in mining and quarrying over 2000-2005 was 15.2 percent and accounted for roughly one-third of industrial growth, despite its one-sixth share of industrial GDP (World Bank 2007e). Moreover, the impact of industrial growth on poverty is limited 14 The dialogue surrounding the HIPC initiative was in itself a trust-building exercise which laid the groundwork for GBS and the PRSC series. 15 There were eight crop boards covering tobacco, cotton, pyrethrum, sisal, coffee, cashew nut, tea, and sugar. 16 because the labor intensity of mining is low, and the manufacturing sector remains small and concentrated in Dar es Salaam. 2.29 Reforms supported by PRSCs 1-3 do not appear to have had a significant effect on private sector development. Implementation remained in its early stages. Under PRSC-1, the Ministry of Industry and Trade prepared a small and medium enterprise policy and launched an associated public awareness campaign, but its implementation was much slower than expected due to insufficient funding from the budget. The private sector strategy that was expected to be completed under PRSC-3 was delayed because the government wanted to first complete the Diagnostic Trade Integration Study, which was not finalized until June 2005, and later dropped during the next series. 2.30 Fragmentation of labor policies in six different, outdated pieces of legislation impeded market-oriented deployment of labor. The government was expected to prepare a new labor policy and approve it under PRSC-2, while in the context of PRSC-3 a new Labor Act was expected to be prepared and presented to parliament for approval. Phase-1 labor legislation covering employment relations, collective labor relations, dispute resolution, and labor market institutions was submitted to the Parliament in November 2003, and approved in April 2004. A task force assigned to do the preparatory works for phase-2 legislation (covering occupational safety, health workers compensation, and employment and skill development) was expected to be finalized under PRSC-3 (2005), but was postponed for 2006 (and was not followed up by the second PRSC series). 2.31 There was a significant unfinished agenda in improving the investment climate and the business environment. In July 2002, the government approved the Business Environment Strengthening for Tanzania program, partly supported by the PSAC. The PRSC focused on the implementation plan for the business environment program that targeted eight priority areas of action. Following approval of the program under PRSC-2, the PRSC program shifted focus to replacing the business licensing regime by a registration system. A draft bill on Business Activities Registration that outlined the direction of the business environment reforms was submitted to Parliament in July 2005. 2.32 Agriculture and rural development were considered critical in broadening the impact of growth to generate poverty reduction given the preponderance of poverty in rural areas, and the reliance of the vast majority of the poor on the sector. The envisaged outcomes for the sector were: an increase in growth of real agricultural GDP from 3.6 percent in 2000/01 to an average of 5 percent by 2010. This was expected to lead to: (i) an increased share of income or consumption accruing to the poorest 20 percent, (ii) a reduced share of population living on less than USD1 per day, and (iii) an increased ratio of trade to GDP. The targeted growth performance in agriculture was roughly achieved. The share of income accruing to the poorest 20 percent fell from 7.3 to 6.8 percent (PovcalNet). A poverty assessment in 2008 suggests that poverty reduction due to agricultural growth was marginal. Overall, the rural poverty headcount barely budged between 2001 and 2007, falling from 38.6 percent to 37.4 percent,16 and rural income only grew 4 percent in total between 2001 and 2007 (URT 2009). Trade as a share of GDP increased from 37 percent in 2002 to 51 16 Disaggregated data were not available using the extreme poverty line. 17 percent in 2005 (World Development Indicators), with export of exploited natural resources a driver. 2.33 The PRSCs supported implementation of the government‘s 2001 Agricultural Sector Development Strategy, and associated Agriculture Sector Development Program, which identified improving agricultural productivity and profitability as a key objective. Fluctuations in agricultural sector growth were more associated with changes in rainfall rather than changes in policy. The 5 percent growth rate was not sustained after the PRSCs closed, and growth was overall flat during the 2000-2010 decade, as shown in Figure C. Rather than serving as evidence of increasing productivity, much of the growth could be attributed to expanding the area of land cultivated rather than increases in land yields, especially given modest or nonexistent gains in use of improved seeds, fertilizer, and area developed for irrigation. Figure 2-3.Sector growth in Agriculture and Mining and Quarrying 18.0 16.0 Annual Growth (percent) 14.0 12.0 10.0 8.0 6.0 4.0 2.0 0.0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Agrigulture Mining and quarrying Source: Tanzania Poverty and Human Development Report 2011 2.34 The government‘s Agriculture Sector Development Program included a broad array of measures to address the identified constraints in the sector, including insecure land rights, lack of clarity on responsibility for regulatory decisions, limited access to agricultural credit, high levels of agricultural taxation, and limited access to markets due to poor transport infrastructure. The PRSCs specifically sought to support this program with an extremely ambitious agenda, including reforms of crop boards17 particularly with respect to input supply, product quality, and competitiveness; micro-finance legislation; the institutional arrangements governing rural roads; local government taxation; property rights; and overall coordination of public interventions in the agricultural sector. A separate investment project—the Participatory Agricultural Development and Empowerment Project— complemented this policy work by financing small agricultural development subprojects 17 There were eight crop boards covering tobacco, cotton, pyrethrum, sisal, coffee, cashew nut, tea, and sugar. 18 planned at the community level, but there was no broad agriculture sector project until the second series. 2.35 In practice, policy reforms supported by PRSCs 1-3 were structural in nature, and not likely to generate a rapid supply response. Regardless, implementation of those reforms was slow and weak, and the sector‘s temporarily improved growth performance owed little to the credits. The passing of the Land (Amendment) Act in 2004 provided for improved clarity of property rights and better securitization of land, but the PRSC-2 trigger to support its implementation was delayed one year, and its benefits did not have time to affect growth during the period. Slow consensus-building between the European Union and Ministry of Lands and Housing Services delayed the preparation of the land sector strategic plan, which was to receive financial support from EU‘s European Development Fund. As a consequence, the reform was redefined as a trigger for PRSC-3, but there was no evidence of immediate ―operationalization‖ as envisaged. Similarly, the approval of the strategy to reform two crop boards was only partially met and was repeated as a trigger for the follow-up PRSC series. The cabinet approved changes to the financing of crop boards, but did not endorse specific reforms of their role and functions, including regulation, governance structures, and accountability to stakeholders.18 2.36 In view of the delays in getting the Bank‘s Agricultural Sector Development Program (2006) approved internally and the initial slow pace of implementation, it appears that the range of issues undertaken by the Bank in this area overstretched its capacity, since this involved coordinating the activities of many sector experts. 2.37 The overall efficacy of PRSCs in achieving the first objective was modest—strong growth was not matched by a broadening of its impact, and there is little evidence that the improved growth performance can be attributed to measures supported by the credits. OBJECTIVE 2: SUPPORTING A RESULTS ORIENTATION OF PUBLIC SERVICE DELIVERY 2.38 PRSC support for this objective was particularly poorly defined. The dialogue was to focus on establishing a robust monitoring and evaluation (M&E) system to feed into assessment of the results of sectoral programs (without directly supporting those programs). The processes for preparing annual public expenditure reviews and medium term expenditure frameworks were to provide a mechanism for monitoring the use of resources through public expenditure management and revenue mobilization. Most of the Bank‘s support was focused on monitoring the use of resources rather than helping to establish the M&E system, where the intention appeared to be to remain engaged in the government-identified priority sectors with an eye to shifting the Bank‘s investment lending support in these sectors into future PRSCs. 18 The joint evaluation of General Budget Support (ODI and DAL 2005) in their assessment of the implementation of this policy noted that it was not so clear that there was a strong political thrust in favor of it from the government side. The report further concluded that agricultural policy could have been more coherent if the three ministries involved directly in the sector were subjected to the common discipline of the MTEF/Budget Guidelines/PER processes. 19 2.39 Monitoring and evaluation. On the M&E side, there were some tangible results. First, and most importantly, when the Bank and government agreed to bring the PRSC under the umbrella of general budget support in 2003, the Bank introduced two new outcome- oriented areas for monitoring under the joint performance assessment framework (PAF)— reduction of income poverty and environmental sustainability—and added detail to the policy matrix that ―brought a greater policy content and a higher level of rigour‖ (ODI and DAL 2005). Second, the government, Bank, and other development partners set up a system for monitoring progress in some sectors through sector working groups. Third, building on the 2002 Poverty and Human Development Report, two additional reports were produced in 2003 and 2005, and have been published every two years since. These reports present and analyze the available data on progress toward achieving poverty reduction strategy goals, organized around the objectives of the national development strategy. By 2005, Bank analytical work served as a major input to analysis. 2.40 Outcome data were widely available in the health and education sectors. While data collected in the course of administering programs—much of which was not considered to be of high quality—provided the backbone of information in these sectors, it was verified through regular Demographic and Health Surveys. The data indicated expanded access to education, reductions in infant mortality, and declining incidence of AIDS.19 However, the quality, frequency, and availability of data on outcomes of other government programs, such as water and electricity, were weak, relying heavily on data collected at the local level at the point of service delivery. And meaningful data on the incidence of income poverty were too infrequent, with household budget surveys conducted in 1991, 2000/01, and 2007. The government and development partners were surprised by the lackluster income poverty results of the 2007 survey, and were forced to rethink their development approach. 2.41 Another avenue to support the transformation of the public service to results- orientation was by introducing annual plans, performance budgets, and linking each to the results of new service delivery surveys. In all ministries, departments, and agencies (MDAs), the government approved strategic plans and action plans, as well as budgets that were linked to the medium term expenditure framework. Service delivery surveys, with beneficiaries as respondents, were completed for all MDAs, but the results were of uneven quality and because the surveys were designed and implemented by five separate firms, the aggregation of results produced by the President‘s Office-Public Service Management was of limited use (United Republic of Tanzania 2004a). More importantly, they were not captured in the strategic planning process, and so the critical link between policy and budgets on the one hand, and results on the other was missing. 2.42 Monitoring use of resources. The program was partially successful in helping to monitor the use of resources, with the Bank playing an important role. The PRSCs drew extensively on the annual Public Expenditure Reviews, and provided the main focal point for operational dialogue around public sector reform. The government adopted a 19 This should not be construed as an impact of the PRSC program, which only indirectly contributed to progress in the education and health sectors by helping to protect them from budget cutbacks through dialogue and monitoring the use of resources, and building the infrastructure for monitoring progress. 20 recommendation from the Bank-supported 2003 PER that requires each agency to explicitly link proposed budget allocations with PRSP objectives, which allows observers to understand resource allocation in terms of broad development goals. In addition, through public expenditure management, the Bank supported the development of three-year rolling budget guidelines, strengthening policy units in MDAs for priority sectors, improvement in expenditure tracking and reporting on budget execution, and upgrading of budget processes for local government authorities. 2.43 The share of the budget allocated to priority sectors was flat as a percentage of GDP during the period, although actual allocations rose with GDP in real terms. Priority sectors were protected even after HIPC support was irrevocably granted, but they did not grow relative to other sectors. It should be noted that this was during a time of relative plenty, as development partners were increasing their support to Tanzania, rather than an adverse fiscal environment where priority sectors may have been more susceptible to budget-cutting. 2.44 There was improvement in budget execution, as resources spent were more in line with the approved budget. There was a reduction in the total absolute deviation between approved budget allocations and actual expenditures at the level of Ministries, Departments, and Agencies, as shown in Figure D.20 The aggregate deviation indicator, while indicating important progress, hides a shortcoming for the quality of expenditures. According to the public expenditure review (PER) of 2004/05, which covered most of the period of the first series of PRSCs, ―over-spending‖ units leaned toward more of administrative expenditures (public administration, defense, foreign affairs), while ―under-spending‖ units constituted those with more expenditures directly going to growth and poverty reduction (health, education). 20 The PRSCs used a different indicator—the share of ―votes‖ (MDA budget allocations) with budget deviations greater than 5 percent. The government and development partners later abandoned this indicator as insufficiently comprehensive. 21 Figure 2-4. Total Absolute Deviation between Approved Budget Allocations and Actual Expenditures Declined Under the Operation 30% 25% 20% 15% 10% 5% 0% Source: PER 2007, PER 2011 2.45 The PER identified other shortcomings, including that (i) there had been no major shift in investments toward infrastructure, although it had been identified as the main constraint for growth; (ii) despite a strengthened formulation of medium term budget strategy, there was still a need to transform the budget guidelines into a comprehensive instrument for guiding public finances over the medium-term; and (iii) little progress had been made in compiling actual expenditure data for the local government authorities. The slow progress reflected weak capacity in local and central governments and the lack of a standard reporting format. 2.46 An important dimension of enhanced public service was the mobilization of resources for local authorities. PRSC-2 supported actions aimed at the rationalization of local government taxes and levies. In the name of supporting private sector development the government abolished ―nuisance taxes‖—small taxes levied by local government authorities on the marketing of agriculture—in the 2002/03 budget, and introduced compensatory grants to local government authorities to replace the lost revenue. However, the effectiveness of these grants was undermined by: (i) delays and unpredictability in the flow of funds from the Ministry of Finance; (ii) continued erosion of local governments‘ fiscal autonomy due to the inadequacy of other revenue sources; (iii) the inability of councils to transfer a share of the compensation grants to villages; and (iv) a weakened drive to strengthen good governance. 2.47 Did the results orientation of public service delivery improve with the better availability of data on results and resource allocation? Where data was available at regular intervals, as in health and education, government counterparts paid close attention and focused on achieving quantitative objectives. The Poverty and Human Development Reports, while far from perfect, were outcome-oriented served as important inputs to the dialogue on general budget support. This was an important step toward orienting the development 22 program to focus on results. Overall, there was important progress toward achieving Objective 2, and while the results-orientation agenda is incomplete, efficacy of Bank assistance is rated substantial. OBJECTIVE 3. ENHANCING PUBLIC SECTOR PERFORMANCE 2.48 The objective to enhance public sector performance, minimize leakage, and strengthen accountability was the centerpiece of the PRSC program. The PRSCs were considered successor operations to the PFM Reform Program and the Public Service Reform Program, which had previously been the primary focus of dialogue on these issues. It was expected that measures in this area—concentrated on public financial management, civil service pay, procurement, and decentralization—would enhance public sector capacity to implement poverty reduction programs in the priority sectors and generate additional funds for poverty reduction. The government, World Bank and other development partners believed that strengthening the civil service and public financial management (including procurement and decentralization) would benefit society by enabling better service delivery and would be indispensable for the success of the budget-support strategy for delivering aid. 2.49 The available data indicate that there were good outcomes on the public sector objectives, especially in policy areas for which the PRSC included prior actions to supplement the dialogue, as elaborated below. Much of the progress in implementing reforms was attributed by government officials and Bank staff to the open and intensive dialogue surrounding the government-led preparation of the annual public expenditure review, as well as the mutual trust that accompanied it. 2.50 Figure E summarizes trends in one measure of public sector performance, the Worldwide Governance Indicators for government effectiveness, voice and accountability, and control of corruption. These indicators are not ideal—ratings are based on perceptions and normalized such that the global average is constant at zero—but they do show trends in perceptions of governance relative to global changes.21 The indicators suggest an improvement in the control of corruption and in voice and accountability, but not in government effectiveness which was flat during the period. 21 The Worldwide Governance Indicators aggregate scores from a number of primary sources on governance—mostly surveys—and then compares each country‘s aggregate score for the year with that of over one hundred other countries in the world. The indicator shows how many standard deviations the country‘s score is above or below the mean for all countries. Tanzania, like most low- income countries, is below the mean, so a less negative score indicates improvement. While giving a good estimates and confidence intervals for a country‘s relative position in a given year, the WGI is less useful for comparisons over time, because the changes result not only from changes in a country‘s absolute performance, but also from changes in the primary indicators that are used in calculating the country‘s raw score and from changes in the collection of comparator countries. 23 Figure 2-5. Trends in Worldwide Governance Indicators for Aspects of Public Sector Performance 2002 2003 2004 2005 2006 2007 2008 2009 2010 0 -0.1 -0.2 -0.3 -0.4 -0.5 -0.6 -0.7 -0.8 -0.9 -1 Government effectiveness Accountability Control of corruption Source: Worldwide Governance Indicators 2011 2.51 Prior actions with respect to public service performance were well-linked to the desired outcomes, and implemented almost in their entirety. Public financial management (PFM) was seen as the critical entry point, increasing transparency and leverage of reformers in MoF to press for other reforms. Moreover, minimally acceptable PFM is considered a pre- condition for provision of general budget support. Progress was made on most of the components of the PFM Reform Program, particularly in areas supported by parallel investment projects (such as the integrated financial management system, external resource management, and procurement). The 2006 Public Expenditure and Financial Accountability Report (World Bank 2006b) observed that performance in Tanzania was stronger than the average for East African comparators for cash management and policy-based budgeting, with average performance in accounting and reporting, comprehensiveness and transparency, procurement, and budget credibility. Although one cannot say definitively whether the PRSCs led to the better performance or whether those issues were picked up in the PRSCs because performance was good anyway, the PRSC series concentrated its triggers and prior actions on these issues, and the dialogue and support can be plausibly linked to that progress.22 On the other hand, fiduciary concerns—external scrutiny and auditing, which were not in focus under the PRSC—got lower scores in the Public Expenditure and Financial Accountability (PEFA)23 ratings, while establishing the PFM Reform Program as a 22 As noted in the PPAR on PRSCs 4-8, the core budget areas were not singled out as targets in the second series, and the 2009 PEFA scores declined in all five. The revenue administration and fiduciary areas, on the other hand, had triggers in the second series, and their PEFA scores all improved from 2006 to 2009. Irish Aid and DFID (2011, p.36). 23 Public Expenditure and Financial Accountability assessments (PEFAs) are internationally standardized assessments. They started in the early 2000s. The grading from A to E is benchmarked 24 coordination mechanism took more time than expected. Table 2-2 summarizes relevant PEFA ratings covering these and other public sector performance issues. Table 2-2. Selected 2006 PEFA Ratings on Public Sector Performance Public Expenditure and Financial Accountability indicator Score Public financial management cash management: Aggregate expenditure A outturn compared to original approved budget Public financial management cash management: Composition of D expenditure outturn compared to original approved budget Public financial management cash management: Aggregate revenue A outturn compared to original approved budget Public financial management cash management: Stock and monitoring of A expenditure payment arrears Public financial management policy-based budgeting: Orderliness and B participation in the annual budget formulation process Public financial management policy-based budgeting: Multi-year B perspective in fiscal planning, expenditure policy and budgeting Anti-corruption: Ethics and anti-corruption measures are in place and are C effective Payroll: Effectiveness of payroll controls C+ Procurement: Competition, value for money and controls in procurement C+ Decentralization: Transparency of inter-governmental fiscal relations C Audit: Effectiveness of internal audit C Audit: Scope, nature and follow-up of external audit D+ Source: 2006 Tanzania Public Expenditure and Financial Accountability Review 2.52 Civil service pay reform was part of a longer agenda that had included an earlier phase of downsizing in the 1990s. That experience was deemed to have succeeded better than in most countries (PER 2002; IEG 2008) largely because it came at a time when privatization and private sector growth opened job opportunities in the private sector, while inflation reduced the real value of public sector salaries, especially at the upper levels, leading to loss of staff. In the 2000s, the remaining problems were the inadequate salaries at upper levels to compete with the private labor market, the use of allowances to compensate for this, and insufficient incentives for teachers and health specialists to work in under-served areas. While the pay raises during PRSCs 1-3 met the triggers, they were not well targeted and did not make progress on the difficult issues, which reflected a problem with program design in terms of targeting.24 Rather than enhancing pay for positions that were most difficult to fill, such as service-delivery personnel in remote areas, employees in Dar es Salaam and other cities were permitted to access travel allowances that exceeded the cost of travel. This was, in effect, a salary supplement. But it was concentrated among civil servants with strong connections who were not on the front lines, and gave them incentive to travel excessively away from their jobs (Twaweza, 2010). with A‘s for advanced economies like Denmark and Chile, so a B or C is a good mark for a low- income country. 24 The program document for PRSC-3 noted: ―At present the gains made in reducing the number of allowances at the beginning of the program seem to be eroding.‖ (World Bank, 2005, p. 37). 25 2.53 Procurement received substantial attention, with prior actions in PRSC-2 and 3. PRSC-1 included an annex with the recommendations of the 2003 Country Procurement Assessment Review, which were prepared ―in parallel‖ with the PRSC and found that ―procurement is a neglected function‖ by the government (Country Procurement Assessment Review, 2003, p. 121). Procurement reform was necessary for reducing corruption and generally for increasing the value for money in public investment, whose expansion the PRSCs intended to finance. The main measure was the transfer of procurement activity from a central tender board to ministries, agencies and local government, and the establishment of a Public Procurement Regulatory Authority (PPRA). The PPRA would enforce regulations and provide an independent check on the procuring entities, and reduce leakage of government resources by providing a more competitive process. By 2006, the PEFA review gave good marks for the improved set up of the procurement process, although there was not a track record yet of implementation (implementation proceeded rapidly during the second series, with the share of agencies complying with PPRA regulations reaching roughly 70 percent). The Auditor General‘s report on Central Government for 2006/07 also gave good marks for procurement, but noted that ―justification for use of less competitive methods is weak and is not yet routinely monitored,‖ an issue that still remains unresolved. 2.54 With regard to decentralization, PRSC-2 intended to support the strengthening of capacity of the central administration to monitor activity by local governments and extending decentralized procurement authority to local governments, in order to improve the transparency and efficiency of the public sector.25 The central ministry charged with oversight of local government authorities hired three staff and procured information technology equipment in the central office and some regional offices to enhance the monitoring function. The Public Procurement Regulatory Authority was intended to regulate the local government tender boards. That approach did not, however, address the largest areas of nominal sub-national spending—wages in education and health—as the local governments had little discretion in these areas. As the program document for PRSC-2 (World Bank, 2004a, p. 40) noted, ―the lack of a clear and unified position with respect to decentralization of human resources management raises concern.‖ This assessment concurs. In PRSC-3 there were no triggers or prior actions for decentralization, but rather descriptions of delays and vague plans to address the issue in the future. There were no specific indicators identified to measure whether or not the decentralization and oversight measures specifically improved public sector performance, but the very limited attention to the issue by the PRSC to this objective suggests a modest contribution to achieving the objective. 2.55 Overall in the most important public service areas with Bank support, there were performance improvements, and so efficacy is rated substantial. OBJECTIVE 4: STRENGTHENING ENVIRONMENTAL MANAGEMENT 2.56 Tanzania faced a number of important environmental challenges at the outset of the PRSC series. Chief among these were land degradation from human activity—including 25 See, for instance, United Republic of Tanzania 2004b for the earliest Local Government Fiscal Report in which the Prime Minister‘s Office-Regional and Local Governments monitors local government budgets. 26 deforestation, soil erosion, overgrazing, degradation of water resources, and loss of biodiversity, which were aggravated by poor agricultural practices—and pollution from improper treatment and disposal of solid and liquid wastes, especially in expanding urban areas. These challenges cut across sectors, including energy and mining, forestry, agriculture, and water and sanitation at a minimum, and the country lacked proper instruments to enforce existing legislation and policies of central and local authorities, especially with respect to resource management in agricultural and range land, forestry, wildlife, and minerals. 2.57 The Bank and government introduced a new objective under PRSC-2 to strengthen environmental management because, it was asserted, the poor were heavily dependent on the environment for both income generation and consumption.26 In addition, the cross-sectoral nature of the challenge required a cross-ministry response in order to be effective. Neither the Bank‘s PRSC, nor the government under the Performance Assessment Framework, identified indicators to measure the achievement of this objective. The late introduction of the topic and limited focus in the program suggest this was considered a lower-priority objective by the government. 2.58 There were no triggers in the series directly related to the objective; rather the PRSC used dialogue to support the government‘s efforts to address environmental issues identified in the National Environmental Action Plan (1994) and the National Environmental Policy (NEP) adopted in 1997. This was done through three approaches: (a) mainstreaming of environmental concerns into the dialogue on poverty reduction through the budget process and sectoral policy dialogue; (b) improving understanding of the poverty-environment linkages in Tanzania; and (c) strengthening institutional capacity to integrate environmental assessment procedures into sectoral strategies and policies. 2.59 In December 2002, the Cabinet approved an Institutional and Legal Framework for Environmental Management Program, which provided the basis for Environmental Impact Assessment, cross-sectoral coordination, and integration of environment management into local government planning processes. A number of key sector policies and strategies in forestry, agriculture, fisheries and mining that followed reflected an increasing focus on internalizing environmental sustainability objectives and empowering the poor for improved resource management. 2.60 During PRSC-1 and PRSC-2, the Bank undertook analytical work that identified entry points for mainstreaming environmental issues into other activities, including an environmental review of all taxes and levies associated with the agricultural sector (which served as the basis for policy reforms of the crop boards), and increased attention to removing land tenure uncertainties. During PRSC-3 the Bank supported an intensive consultation process leading to the passage of the Environmental Management Act (EMA), a new legal and institutional framework for environmental management. 26 The mechanism for this dependence was not discussed in the program document, but the heavy reliance of the poor on rain-fed agriculture and fishing for income and sustenance, and the economy‘s dependence on natural resources was likely the source of this assertion. 27 2.61 The Bank provided capacity support to the Vice President‘s Office and the National Environmental Management Council for the development and implementation of sector- specific guidelines for environmental assessment in roads and agriculture, and pursued procedures for strategic and sectoral environmental assessment. It also contributed to an initial needs assessment for implementation of the EMA as part of the first public expenditure review on environment. The production of a large number of key sector policies and strategies in forestry, agriculture, fisheries and mining that followed reflected an increasing focus on internalizing environmental management at the sectoral level as directed by the EMA. But implementation of strategies remains weak because of the undeveloped institutional structures and the lack of qualified personnel. In addition, Environmental Impact Assessment (EIA) training manuals and guidelines were prepared. Implementation of this aspect of the EMA seems to have been effective, as the number of projects undergoing EIAs that complied with the regulations increased from 8 in 2005/06 to 166 in 2010/11 (United Republic of Tanzania 2011b) and the number of districts allocating budget to EMA implementation rose from 0 in 2005 to 121 (out of 132) in 2009, according to the 2010 PAF. 2.62 The lack of pre-identified outcome indicators or reporting on environmental outcomes, combined with the focus of the PRSC series on legislative and process reforms through dialogue, makes assessment of development outcomes and attribution to Bank support difficult. PRSC-2 specified five output indicators—the share of sectors with environmental units, the share of districts that have officers with environmental functions, the share of district-level environmental officers that have been trained, the share of districts with committees of environment, and commencement of activities to implement 50 percent of obligations under international environmental conventions and agreements—to monitor, none of which was tracked, nor available in the field. Targets for three of these indicators were reduced in the program document for PRSC-3 without explanation. For instance, the target for the share of sectors with environmental units was reduced from 90 percent to 40 percent. 2.63 Because there are no measured outcomes from the support provided on environmental issues under the PRSC series (in the absence of clear results framework), but some evidence of increased attention to environmental management across local government authorities, efficacy is rated modest. Ratings OUTCOME 2.64 The objectives of the series of PRSCs were relevant to the conditions in Tanzania, particularly the primary focus on improving public sector performance, including issues of public financial management, given the weak environment for development in the country. The objectives were closely linked to both the Bank‘s and the government‘s development strategies, and the circumstances of the country at the time. Objectives of the PRSC series are substantially relevant. 2.65 The PRSCs for the most part identified important policy areas to achieve its objectives. But for several objectives, it was not obvious that these policy changes would 28 contribute much to the desired outcomes, with prior actions and proposed triggers often preliminary in nature and unlikely to deliver results during the implementation period. Moreover, the results framework was rudimentary, without a clear articulation of how progress in adopting the proposed policy measures would help to achieve the program‘s objectives and without consistent outcome indicators for each operation and for the overall series. Relevance of design is rated modest. 2.66 The PRSCs had a mixed result in terms of the achievement of the four key objectives of the program. On sustaining and accelerating economic growth and broadening its impact, efficacy was modest because poverty reduction was slow and the PRSCs did not directly contribute to robust growth. On supporting results orientation of public service delivery, efficacy was substantial because there was more focus on poverty reduction in the joint donor performance assessment framework, greater availability of data on health and education outcomes, and a more systematic sector-level planning process, but there was also great variability in the quality of data on other government services and attention remained focus on monitoring budget allocations to priority sectors rather than measuring outcomes. Efficacy for improving public service performance, the primary focus of the series, was substantial. Outcomes on public financial management were strong in almost every area, with the exception of procurement (which improved following reforms under the program), audit (not a focus of the program), and tax collection (also not a focus of the program), while ratings on transparency and anti-corruption improved and outcomes for managing decentralization services were weak. On strengthening environmental management, efficacy was modest—little could be attributed to Bank support as its role was primarily in bringing the issue into the dialogue with development partners and no data on outcomes were available. 2.67 How far would Tanzania have gone in achieving its objectives without the PRSCs? The intense public expenditure review dialogue and careful macroeconomic management, to which the PRSC was a direct contributor, contributed to the stabilization and acceleration of growth. Public sector performance improved as a result of improved budget execution and public financial management, although later improvements from procurement reforms may have been achieved on a slower track if addressed sector-by-sector. Tanzania‘s IDA envelope would have otherwise been committed through investment loans. While implementation of investment projects may have delivered important development outcomes at the sector level, they would be unlikely to achieve the public sector performance results of the PRSCs, most of which required national level dialogue with government officials holding a cross-sectoral perspective. Table 2-3. Summary of Outcome Ratings for PRSCs 1-3 Dimension of rating Rating Relevance of objectives Substantial Relevance of design Modest Achievement of objectives Sustaining and accelerating growth and broadening its impact. Modest Supporting results orientation of public service delivery. Substantial Improving public sector performance. Substantial Strengthening environmental management. Modest Overall outcome rating Moderately satisfactory 29 2.68 Overall, the outcome of PRSCs 1-3 is moderately satisfactory. Table 2-3 summarizes the subcomponents of the rating. In the face of mixed ratings on both the series‘ relevance and its achievement of objectives, the overall rating falls above-the-line at moderately satisfactory for two reasons. First, the program‘s design was modest in its relevance, but introduced without the benefit of the current Bank guidelines, which were put into effect in late 2004. Moreover, the relevance of design improved over time and maximized continuity with the previous two adjustment credits. Second, the third objective—improving public sector performance—is assigned the greatest weight, given its prominence in the program design and dialogue, and the Bank‘s role as the primary development partner focusing on these issues, while the fourth objective—strengthening environmental management—is assigned the least weight due to its limited emphasis in the PRSC design, dialogue with government, and the government‘s program. RISK TO DEVELOPMENT OUTCOME 2.69 At the time the PRSC series closed, the primary risks to development outcomes of PRSCs 1-3 were internal, and related to government ownership and a strong participatory framework. Many of the cross-cutting reforms were yet to have been fully implemented throughout government. If ownership were to fray or the broad-based participatory oversight (through technical committees and sector working groups) were to diminish in importance and critical actions to implement difficult policy reforms were to be abandoned, outcomes would be at serious risk. At the close of the series, the government‘s commitment appeared to be secure, although a change in Administration, and the movement of key reformers within government has coincided with a lesser focus on the reform agenda since that time. Development partner participation remained strong, and the broad-based participatory process in formulating the government program was likely to continue. 2.70 In the event, participatory oversight has continued apace but broad government ownership has since deteriorated, as the next phases of many reforms have been delayed and commitment to macroeconomic stability has frayed. Shortcomings of the first series on public financial management, such as inadequate reporting by local governments, was not taken up in the second series. An additional important risk was the limited capacity of government in policy and planning, which could undermine complex policy formulation and implementation as the reform process moves to a higher order. Assessment by the GBS partners (DAIMA and ODI, 2004) noted that some key senior policy staff lacked strong institutional and technical backing, and were swamped by administrative tasks. This risk is more likely to affect outcomes in the future, and is of particular concern because many of the achievements came short of implementation. 2.71 One external risk to development outcome was the government's continued heavy dependence on external resources to finance the reform effort. During 2003-05, while total government expenditures increased from 22 percent to 28.4 percent of GDP, the total revenues only increased from 12.7 percent to 14.2 percent, and much of the growth during the period could be attributed to government consumption fuelled by external support. Were the flow of concessional finance to slow, weak domestic revenue mobilization would likely slow down the reform effort, an issue that would become increasingly constraining if policy 30 performance were to falter. Since the series of operations closed, after rising to 15.6 percent of GDP in 2008/09, revenue fell again to 14.6 percent in 2009/10 (IMF 2011). At the same time, the scale of international budget support is tightening as a response to weakening commitment to reforms, but domestic revenue cannot yet fill this gap. 2.72 Overall, risk to development outcomes is significant. BANK PERFORMANCE 2.73 Quality at Entry was moderately satisfactory. The framework adopted under PRSCs 1-3 was tightly aligned with the government‘s development framework as outlined in the PRSP. The focus on macroeconomic and cross-cutting governance issues was strategically relevant. The existing dialogue surrounding the joint authorship of annual public expenditure reviews served as an important jumping-off point for the series. Developing a reform program based on a harmonized framework with government and other development partners also meant dealing with trade-offs with respect to what to include and what to exclude. On the whole, the Bank made strategically relevant choices of policies to support, focusing on issues that were the most important for income and non-income poverty reduction— agriculture and rural development, private sector development, public finance management and service delivery. The PRSCs were sometimes used opportunistically—elevating important reforms to the national dialogue if they stalled at the sector level. The weakest aspect of the program with respect to quality at entry was the results framework, which was poorly developed and had negligible value for monitoring progress of the program. 2.74 Program design incorporated lessons learned in the design and implementation of support under the SAC and PSAC. Three were relevant to the PRSC. First, it was suggested that forging a strong working relationship with the government and supporting capacity building could lead to successful implementation of an operation that is broad in scope and complexity (internal Bank evaluations). Second, promoting government ownership and using national systems where possible could facilitate implementation performance and efficiency (consensus under Paris Declaration). Third, monitoring and evaluation aided by quantitative performance indicators could help identify the impact of programs and promote accountability (internal Bank evaluations).27 In this respect, significant design characteristics of the PRSCs included: (i) the adoption of an annual single tranche, ex-post framework that assured the predictability of resource flow to government budget linked to prior actions;28 (ii) support for a coherent policy dialogue across sectors through the budget framework, thereby avoiding fragmentation of assistance to critical sectors; and (iii) efficiency in policy dialogue and performance assessment by aligning with other development partners in the use of a common performance assessment framework (PAF). This design does involve a potential trade-off between resource flow predictability and adjusting the credit commitment 27 See also, IEG‘s review of PRSCs (IEG, 2010). 28 The PRSC design (unlike that of the SAC) limited the number and scope of conditions (triggers and prior actions, thereby helping to ensure that the actions were manageable prior to approval by the Bank‘s Board. 31 in an event of failure to implement triggers for the subsequent operation, an issue that became problematic later in the second PRSC series. 2.75 The design of PRSCs 1-3 was consistent with the harmonized support mechanism of the poverty reduction budget support (PRBS), which initially had nine donor partners and later became the general budget support (GBS) group.29 GBS was guided by the government‘s PRSP, and aimed to provide a coherent policy dialogue at cross-cutting levels and in sector wide programs, efficiency in policy dialogue and performance assessment, and enhanced flexibility and predictability of resource flows (ODI and DAL 2005). Even so, the Bank was able to inject critical issues—such as environmental sustainability—into the dialogue that did not receive much attention in the PRSP. The development partners used a common performance assessment framework (PAF). The PRSC review process was thus conducted in the context of the annual review process of the PRBS, which involved a mid- term review around March-April and an annual review in October, and included inputs from various parts of government, civil society and the PRBS development partners. 2.76 The program was built on the analytic and participatory work that went into the development of the PRSP, which included the Bank‘s April 2001 Country Economic Memorandum, May 2001 Country Financial Accountability Assessment and April 2003 Country Procurement Assessment Review. There was also significant cooperation with other development partners as discussed in the first chapter, including through the shared use of a Performance Assessment Framework as well as co-financing from the German government through KfW. 2.77 Quality of supervision was moderately satisfactory. Formally, there were two supervision missions per year for each PRSC, but implementation support was intensive and ongoing, delivered primarily through regular consultations surrounding preparation of annual PERs. The Performance Assessment Framework served to coordinate general budget support across multiple donors. On public sector reform issues, supervision of the PRSCs was closely linked with supervision of related investment loans for public finance management reform, public service reform, and local government reform. Some delays occurred in the implementation of PRSC-3 (especially with respect to the reforms of the commodity boards), but they reflected an underestimation on the part of Bank staff of the strong interests opposed to reform and the time required to reach consensus. Although data on supervision are not readily available, Bank staff total time (preparation and supervision) devoted to each PRSC declined from 120 staff weeks for PRSC-1 to 104 staff weeks for PRSC-3. The Bank should have, however, adapted the programs to better monitor and evaluate progress over time: the results framework should have been overhauled to better link policy actions and outputs to intermediate and final outcomes. 2.78 Overall, the preparatory work for the PRSC, underpinned by analytical work by the Bank and supervision both by the substantial Bank team in the country office and staff in 29 The PRSC series was the first policy lending instrument by IDA for Tanzania that was fully integrated into the harmonized donor support mechanism (World Bank, 2007). The PAF was expanded considerably to accommodate the Bank‘s priorities under PRSC-1 (Overseas Development Institute and Daima Associates 2005). 32 Washington assured a rating of moderately satisfactory for Bank performance, despite a weak results framework. BORROWER PERFORMANCE 2.79 Government performance: The Ministry of Finance played a coordinating role and served, along with the Office of the Vice-President and the Prime Minister‘s Office as the key central agencies in driving the PRSC process in Tanzania. A small but strong network of highly-committed reformers in these departments provided leadership in both the formulation and the implementation of policy actions included in the PRSC through the intensely collaborative annual public expenditure reviews30. They guided the harmonization of support from development partners and facilitated the establishment of a common framework through the PAF. Government ownership of the program and its commitment to the joint PRBS process was also a key factor ensuring prompt implementation. When delays in implementation of policy actions occurred, as under PRSC-2, the government, under leadership of the Ministry of Finance, took swift action to bring back the program on track. The performance of the central agencies was satisfactory. 2.80 Implementing agencies: The Ministry of Finance also led implementation, but with support from several line ministries. These agencies worked in the context of sector working groups to hash out policy issues and to monitor implementation. Sector policies were decided upon at the technical committees, which were instrumental in generating indicators and assessing progress. But because monitoring was weak and there were delays in implementing reforms in the agriculture sector in part due to a lack of ownership, the performance of implementing agencies is rated moderately satisfactory. 2.81 Overall, the performance of both the central agencies, spearhead by the Ministry of Finance, and the implementing MDAs is rated moderately satisfactory. MONITORING AND EVALUATION 2.82 Design: There was no clear M&E framework at the outset of PRSC-1, nor were any key indicators identified. As part of the agreement under the multi-donor framework, the Bank waited on the framework being developed by the government and its development partners, with significant Bank input. The follow-up PRSC-2 and 3 in the series only included a rudimentary results framework, with virtually no consistent monitoring indicators for each operation and for the overall series, while the PAF was too complex to monitor effectively. 2.83 Implementation: The monitoring for the three PRSCs was conducted in the context of the yearly PER process and joint GBS Annual Review, and a poverty monitoring system that regularly reported on the implementation of the PRS through yearly Poverty and Human Development Reports. Many of the indicators specified in the PAF were collected, but the data often lacked relevance for the PRSC. No doubt implementation was hampered by the 30 The joint Bank and Ministry of Finance team working on the PERs met on a bi-weekly basis. 33 poor and changing M&E design, but nevertheless, there was not a clear set of data collected or presented in the program documents or Implementation Completion and Results report. 2.84 Utilization: Due to the infrequent availability of outcome data, the system was focused on inputs and outputs (meeting budget targets for social sectors, etc.). The program was not adjusted in response to the availability of information generated by M&E. 2.85 The PRSC results framework was rudimentary and although it evolved over the course of implementation and with increasing assistance and involvement of the development partners, it was still inadequate for tracking program implementation, especially with respect to evaluating development outcomes. The quality of the M&E system for the series was negligible. 3. Series 2: PRSC 4-8 Objectives, Design, and Relevance 3.1 Objectives. The program document of PRSC-4 (World Bank 2006) did not state the series‘ objectives explicitly, but it noted that the proposed operation was the first in a series of five annual operations to support the implementation of MKUKUTA, implicitly adopting its objectives. In this respect, the objectives of the series were identical to the pillars of the MKUKUTA:  accelerate growth and the reduce income poverty;  improve quality of life and social well being; and  improve governance and accountability. 3.2 Beginning with PRSC-6, the Bank program articulated the objectives of the PRSC series as ―(i) sustaining high and shared economic growth, and (ii) expanding the effective delivery of basic services through the government budget.‖ The first of these is roughly equivalent to the first pillar of MKUKUTA. The second of these is intermediate in nature, and intended to support the second and third pillars of MKUKUTA. Given this restatement serves as an output-oriented consolidation of the original more outcome-focused MKUKUTA objectives, this evaluation takes the earlier formulation as the program objectives for assessment. 3.3 The program did not distinguish between the objectives of individual operations and those of the full series. That said, the emphasis of the operations changed over time. There was an initial focus on core cross-cutting issues around budget execution and public sector reform, as well as agriculture and rural development. The program document for PRSC-6, after a change in team and country leadership, emphasized a shift to focusing on agriculture and rural development and private sector development, seeing these as the binding constraints to poverty reduction. But in practice, policy areas in infrastructure were introduced at this time, and agriculture and private sector development all but disappeared as part of the program. 34 3.4 Relevance of objectives. The second PRSC series (PRSCs 4-8 covering the period 2006-2010) was deeply grounded in Tanzania‘s National Strategy for Growth and Reduction of Poverty (NSGRP, 2005-2010, more commonly referred to by its Kiswahili acronym, MKUKUTA). It was equally linked to the Bank‘s country assistance strategy (CAS) (World Bank, 2007a) covering 2007-2011, which by that time, because of the evolving donor alignment, was an annex to the broader Joint Assistant Strategy for Tanzania. Like MKUKUTA, the CAS was focused on providing support for growth and reduction of income poverty as well as supporting improvement of quality of life, social well-being, and governance and accountability, and included four pillars: (i) private sector and infrastructure development, to support the government‘s objective of making the private sector the engine of growth; (ii) sustainable rural development, to improve the livelihood of most of the poor, who live in rural areas; (iii) improved social infrastructure, to facilitate improvement in social indicators through enhancement in the access to essential public services by the poor; and (iv) public service reform and institution building, to increase the effectiveness of public service delivery and improved governance. 3.5 In addition to consistency with MKUKUTA and the CAS, the objectives of the PRSC series corresponded with the multi-donor performance assessment framework (PAF)31 built around the MKUKUTA. The program‘s objectives were not articulated to specify its area of focus within the government program. This distinction would have facilitated greater focus on the Bank‘s contribution to such a broad program. Moreover, carefully defined objectives could have helped to guide the reform program as circumstances changed, a particularly important issue in Tanzania due to the unusually long duration of the series. Relevance of objectives is rated substantial. 3.6 Design. The series was initiated in the context of a global push, with the Bank‘s commitment at the highest level, for stronger aid coordination and country ownership of development programs. This agenda, which was articulated as the 2005 Paris Declaration, directly shaped the program, as was the case in many other PRSC programs.32 Indeed, the Bank and other partners articulated a desire to increase general budget support to gradually become the primary vehicle for resource transfer to Tanzania, maximizing country ownership over the use of resources. Moreover, development partners adopted a joint strategy partway through the series, with the government‘s development strategy as its core. 3.7 The PRSC design evolved during its implementation, as new policy areas were added over time. Initially, in service of the first objective, the program included macroeconomic stability and debt management, improving the business environment, and agriculture and rural development as policy areas. Transport and energy were added under PRSC-6, and 31 The revised framework of the PRSC and the PAF responded to the recommendations of the 2004 annual review of the PRBS/PRSC, which emphasized the application of the principles of the Paris Declaration. 32 IEG‘s evaluation of PRSC experience [IEG, 2010] finds evidence that overall the PRSC instrument served to strengthen country ownership, shifted the focus of operations towards public management reforms, eased conditionality, and provided more predictable funding. However, in several countries there were challenges to donor harmonization, and little evidence of success of pro-poor service delivery. 35 natural resources was added under PRSC-7. Under the second objective, the PRSCs included health, education, and water as policy areas from the outset. Under the third objective, public sector reform was the policy area, and it included issues of procurement, auditing, civil service pay, budgeting, and core public financial management. 3.8 Policy areas were derived from the dialogue emanating from the MKUKUTA and were primarily intended to catalyze policy reform in the specific sectors, particularly those that had stalled in the course of sector-specific dialogue. An exception was in health, education, and water. In these policy areas, the program did not promote particular policy changes, but rather provided support for an incipient process meant to build up and strengthen the policy dialogue in those sectors, and monitor results. It helped to introduce annual sector reviews in education and water, following the model in health that was already well-established. 3.9 The program was backed by Tanzania‘s comprehensive monitoring system, designed with input by both government and development partners, which covered outcome, output, process, and input indicators. The structure comprised three interrelated tiers: (a) The MKUKUTA monitoring system, which focused on the measurement of impact, outcomes, and proxy indicators; (b) the sector programs, such as those for education and health, which focused on the tracking of output indicators in the context of ongoing projects/programs under implementation; and (c) the PER/MTEF processes, which focused on inputs, with emphasis on budget allocation and budget execution. In order to strengthen domestic accountability and facilitate participation, the PRSC also supported regular reporting to the public in all areas under its scope. 3.10 The program made several different, and not always consistent, attempts to link program activities to final outcomes. Because the initial program documents did not sufficiently map out the chain leading to desired results, the program document for PRSC- 6—after two full years of implementation— introduced ―outcome areas‖ with associated indicators and targets that were directly mapped to the three MKUKUTA program objectives. This was too late in the process to drive the Bank‘s agenda. Program documents for PRSC-7 introduced another framework for this linkage, mapping seven ―operational policy areas‖ to the two restated objectives. The program document for PRSC-8 maintained this terminology, but re-stated and expanded upon the operational policy areas. Finally, the Implementation Completion and Results Report—the Bank‘s self-evaluation—reformulated the MKUKUTA objectives as program objectives (―PDO 1: Reduce income poverty . . . PDO 2: Improve key human development objectives . . . PDO3: Improve the capacity of the government to design and implement development policies in a transparent and efficient manner‖), and adopted the outcome areas identified beginning with PRSC-6 to help assess the program‘s achievements. 3.11 There were some clear shortcomings in the program design. First, the sectoral scope of the series was too wide-ranging and lacked clear prioritization. The triggers and prior actions associated with the PRSCs covered almost all the policy areas highlighted in the MKUKUTA. Rather than selected to jointly form a single coherent program focused on achieving realistic goals, many of these policy actions were selected to fill gaps when the reform dialogue at the sector level stalled, and were not always drawn from the joint 36 Performance Assessment Framework.33 Second, as with the first series of PRSCs, a workable results framework was only fully developed by the third operation of the series despite the specification of a variety of indicators previously. This results framework, though not repeated in the program documents for later operations in the series and so not entrenched in the ongoing monitoring of the program, was more manageable than in the first series, identifying intermediate indicators and attempting to link planned policy actions to expected outcomes. Third, the PRSCs—and the broader general budget support—failed to address directly the decentralization program, even though the program maintained service delivery objectives and local governments were expected to take on many aspects of that agenda. This agenda was largely pursued through a Bank-financed local government support project, without linkage to the PRSCs. Finally, the dropping and dilution of many triggers during the course of implementation points to a poor assessment of government commitment to implement the planned reforms at the preparation stage of each operation. 3.12 Relevance of design is rated modest. Implementation 3.13 The second PRSC series included 5 operations, associated with total financing equivalent of USD 1,025 million during 2006-11, including supplemental financing to PRSC-7 in response to the global financial crisis. Table 3-1 outlines the timing, approval and closing modalities associated with the five operations. The size of PRSC-5 and of PRSC-7 was reduced by USD10 million each due to the fact that critical triggers were not met. PRSC- 8 was far smaller than anticipated in the CAS and less than the other operations in response to cumulative delays in the reform progress. Overall, IDA commitments for PRSC-4-8 were lower than anticipated in the 2002 and 2007 CASs by approximately USD 315 million, even including the unanticipated Supplemental Credit, because of the shortfalls in policy implementation on the part of the government. The 2007 CAS envisaged an increase in PRSC commitment from about USD190 million in 2006/07 to as much as USD400 million in 2009/10, with the expectation that, as the government‘s policy dialogue with the Bank and other development partners matured, the PRSCs would eventually replace investment lending in several sectors, for example health, education, water, and agriculture. But this vision of the PRSC was moderated when GBS was no longer seen by the Bank and other development partners to be effective as the primary vehicle for resource transfer to the government. 33 The Bank team reports that some triggers were chosen outside the PAF because the formulation in the PAF was not suitable to be monitored as a trigger. 37 Table 3-1. Timing and Scale of PRSCs 4-8 Operation Proposed Actual Approval Effectiveness Closing Delays Implementing amount amount date date date agency Ministry of PRSC-4 $200m $200m 05/09/2006 07/18/2006 06/30/2007 None Finance Ministry of PRSC-5 $200m $190m 04/24/2007 08/27/2007 06/30/2008 None Finance Ministry of PRSC-6 $200m $160m 10/21/2008 11/17/2008 11/30/2009 Yes Finance Ministry of PRSC-7 $200m $190m 06/09/2009 08/20/2009 11/30/2010 None Finance PRSC-7 Ministry of NA $170m 12/22/2009 NA 11/30/2010 None supplement Finance Up to Ministry of PRSC-8 $115m 09/28/2010 11/19/2010 08/31/2012 None $400m Finance Source: 2002 and 2007 Country Assistance Strategies, World Bank Business Warehouse database 3.14 Triggers and prior actions constituted the main architecture for ensuring consistency between policy dialogue and implementation. There were numerous slippages in the government‘s compliance with the triggers of the PRSC 4-8 series and this progressively deteriorated over the course of the program (Table 3-2). The difficulties in meeting the triggers cut across the three objectives of the program. For example, 16 of the 22 triggers under the first objective of growth and the reduction of income poverty, 8 of the 11 triggers for the second objective to improve the quality of life and social well-being, and 13 of the 19 triggers for the third objective of governance and accountability, were fully completed. Implementation difficulties with triggers associated with agriculture and public financial management were particularly prominent, especially after PRSC-6. Table 3-2. Progress on the Triggers for PRSC Program 2 (in numbers) PRSC 7 PRSC 4 PRSC 5 PRSC 6 PRSC 7 PRSC 8 Program Suppleme Total (Cr4170) (Cr4294) (Cr4526) (Cr4635) (Cr4813) ntal Triggers 10 9 11 11 11 52 Completely Met 10 7 8 5 7 37 Partially Met 0 2 3 4 1 10 Not Met 0 0 0 2 3 5 Credit Amount (USD million) 200 190 160 190 170 115 1025 Source: Modified from program document for PRSC-8 (World Bank, 2010c), Table 8. 3.15 In addition to Bank financing, thirteen donors (African Development Bank, European Union, Canada, Japan, Netherlands, Switzerland, Germany, Ireland, Norway, Denmark, Finland, Sweden, and DFID) also provided budget support in the context of the GBS framework initiated at the beginning of the PRSC program, which included a Partnership Framework Memorandum, a Performance Assessment Framework (PAF), and an agreed Joint Review Process. Disbursement from the GBS increased from USD 393 million in 2005/06 to a high of USD 918 million in 2009/10, before dropping to USD 616 million in 2010/2011 (Development Partner Group 2011b). 38 3.16 Supervision and preparation of the subsequent PRSC credits and grants took place in close collaboration with other donors (GBS partners, the IMF and the African Development Bank) and was consistent with the MKUKUTA review mechanism. This involved the use of the PAF (or results matrix) defined by the monitoring and evaluation (M&E) framework of the MKUKUTA.34 Little progress was made in strengthening Tanzania‘s routine data system under the first PRSC series, which required sustained attention. The new Performance Assessment Framework defined for PRSC 4-8 differed significantly from the previous one.35 It covered six focus areas: the three pillars or clusters of the MKUKUTA, macroeconomic stability, resource allocation and budget consistency, and public financial management. The PAF was supportive of the outcome orientation of the MKUKUTA and of domestic processes by focusing primarily on defining a small number of indicators that allowed for an assessment of overall progress and the definition of local processes, which was central to the Joint Assistance Strategy for Tanzania between the GBS partners and the government. Overall, while the M&E system for the MKUKUTA was elaborate, it produced M&E outputs of uneven relevance. The system also suffered from the unavailability or low frequency and/or quality of some of the critical outcome indicators, such as poverty estimates, access to infrastructure, or learning achievements. 3.17 Sectoral and thematic reviews preceded and fed directly into the GBS Annual Review, which provided two inputs to the budget guidelines process. First, it provided feedback from GBS donors on strategic issues for consideration in the budget guidelines process. Second, it provided a firm commitment of the amount of GBS for the next fiscal year to allow government to prepare its budget within a relatively stable and firm external financing framework.36 3.18 Partway through the series, trust between the Bank and government counterparts began to erode, according to multiple interviews. This trust had been built prior to and during the first PRSC series on a technocratic level, through intensive collaboration surrounding preparation of the jointly-produced annual public expenditure reviews. But when the process for preparing PERs receded as the primary entry point for dialogue, replaced by the GBS Annual Review, this technocratic basis for collaboration was removed. As key reformers in government with whom the Bank had built trust moved on to new posts, there was no longer a solid foundation through which to build a common understanding on a technocratic basis going forward. The Bank reports that as part of a third PRSC series, the process for preparing PERs was revamped, and now is more demand driven with a group of champions for reform in the government. 34 For most of the program areas, supervision and dialogue were carried out on a continuous basis in the context of sectoral or thematic programs. This included the PER process, the PFMRP, the MKUKUTA, the Agriculture Sector Development Program, the Business Environment Strengthening in Tanzania program and sectoral programs. 35 In the 2004 annual review o f the PRBS/PRSC it was decided to start work on a new PAF and a redesign of the PRBS (DAIMA and ODI, 2004). 36 As noted above, the Bank disbursed USD 10 million less than its commitment for both PRSC-5 and PRSC-7. 39 Achievement of the Objectives 3.19 As first articulated in the program document for PRSC-6, there were six primary indicators to measure progress against the three program objectives, as summarized in Table 3.3. None of the targets for these six indicators was fully met, although two were nearly met. Table 3-3. Key Outcome Indicators for PRSC Development Objectives Objective Indicator Baseline Target Latest Result Growth and reduction Rural poverty headcount 38.6% 34% (2010) 37.6% (2007) Not met of income poverty a/ (roughly USD 0.32/day) (2001) Nearly Improvements in Under five mortality 112 (2004) 79 (2010) 81 (2009/10) met quality of life and social 70% well-being Literacy (age 15+) 69% (2002) 80% (2010) Not met (2009/10) -0.31 Nearly WGI on accountability 0 (2010) -0.09 (2010) (2005) met Governance and -0.73 -0.20 WGI on control of corruption -0.49 (2010) Not met accountability b/ (2005) (2010) WGI on government -0.35 0 (2010) -0.50 (2010) Not met effectiveness (2005) Source: Household Budget Surveys, Worldwide Governance Indicators, Implementation Completion and Results Report a/ The only key indicator identified for growth and reduction of income poverty was related to rural income poverty. In the event, economic growth was strong through the period, never dipping below 6 percent per annum. b/ Governance and accountability indicators are from the Worldwide Governance Indicators, which combine the views of enterprise, citizen, and expert survey respondents in industrial and developing economies to measure governance performance on a comparative basis, with an average worldwide score of zero and ratings ranging from -2.5 to +2.5, Standard errors for the 2010 measurements are 0.11, 0.16, and 0.17 respectively. 3.20 Analysis on the achievement of each objective follows, with greater detail provided in Annex D. OBJECTIVE 1: GROWTH AND REDUCTION OF INCOME POVERTY 3.21 The sole outcome indicator identified in project documents for this objective was to reduce the rural poverty headcount. Economic growth remained strong throughout the period, never dipping below 6 percent per annum even in the face of the global financial crisis. The latest reliable data available for income poverty are from 2007.37 They show little change in the poverty headcount and a very low elasticity of poverty to GDP growth. The components of the country‘s growth have not changed appreciably since 2007, and so a sharp reduction in income poverty since then is highly unlikely to have occurred. An alternative measure of poverty, the multi-dimensional poverty index constructed for the United Nations Development Programme‘s Human Development Report, showed gains in all three components of the index—health, education, and standard of living—and a 4.1% annual reduction in the poverty rate between 2008 and 2010 (Alkire and Roche 2013). 3.22 The program document for PRSC-6 also identified three outcome areas, each with associated indicators: 37 The government introduced a panel survey instrument, but its limited design makes it unable to accurately measure poverty trends. 40  Improved business environment by maintaining a conducive macro framework and reducing administrative barriers.  Increased access of private sector to key inputs, in particular credit and backbone services (electricity, roads, information and communication technologies).  Improved livelihoods in rural areas, where the majority of the poor work, through improved agricultural policies and expansion of rural infrastructure. PRSC-6 introduced these outcome areas to define the Bank‘s objective better. This section addresses these outcome areas in turn, considering the PRSC‘s contribution to progress made on the reform agenda intended to improve these outcomes 3.23 Progress in the three outcome areas were likely to contribute to growth and poverty reduction. A strong macroeconomic framework in the late 1990s and early 2000s provided the stability necessary for economic growth. Studies have concluded that growth would have been faster had not poor infrastructure—particularly an unreliable electricity supply and poor transport facilities—constituted a significant barrier (see for instance Millennium Challenge Corporation 2011 and Shkaratan 2010). And the vast majority of the poor live in rural areas, meaning that improving livelihoods there could directly reduce income poverty. 3.24 However, progress in each of these outcome areas was modest. The macroeconomic environment deteriorated and the business environment did not show signs of improvement. Private sector access to key inputs, including electricity and credit, did improve, but by less than had been anticipated. In rural areas, agricultural growth was weak and development of rural infrastructure was limited. 3.25 Improving the business environment by maintaining a conducive macroeconomic framework and reducing administrative barriers. The PRSC focus regarding macroeconomic stability was on strengthening public expenditure management. Broad macroeconomic indicators were chosen to measure progress—the fiscal deficit as a percentage of GDP and the rate of inflation. The fiscal deficit remained below 5 percent until 2008/09, when it rose above its target level due to poor revenue performance, a surge in discretionary spending, and the fiscal stimulus put in place to withstand the global crisis (as advised by the Bank and development partners). The balance dipped to 5 percent again in 2011/12 (IMF 2013). This counter-cyclical policy was appropriate. By 2011, Tanzania‘s debt to GDP ratio was close to 40 percent, up from the lowest level of around 25 percent on the onset of the program in 2005. Both revenue and expenditure performance led to higher than expected accumulation of debt. The tax revenue to GDP ratio was flat since 2007, and there was a rapid increase in expenditures unrelated to the fiscal stimulus following the crisis. Inflation accelerated, driven in part by global food and fuel price spikes, reaching into double digits by the end of the period, and persisting since. 3.26 Regarding the business environment, there were a number of outstanding issues from the first PRSC series. These included formulating a private sector development strategy, a second phase of labor legislation, and improving the investment climate. 3.27 The program document identified improvement in Tanzania‘s Doing Business ranking as the outcome indicator, with a MKUKUTA target value of 99 (and a PRSC target 41 of 130) compared to the baseline of 136. This indicator is inherently weak—rankings are relative to performance in other countries, data are revised frequently, conclusions are sometimes based on interviews with only a few informants, and many indicators measure the business environment de jure rather than de facto. The MKUKUTA target was not met— overall improvement in the Doing Business indicator under the second series of PRSCs was insignificant—although the much more modest PRSC objective was achieved. Tanzania‘s ranking remains poor even by regional standards, suggesting the PRSC target was much too modest. 3.28 There was progress on an absolute scale for some of the sub-indicators, notably for the indicator of the cost of starting a business (down from 187 percent of per capita GDP in 2005 to 37 percent in 2010) and trading across borders, both in terms of number of documents and number of days needed for transactions. For imports the number of documents was reduced from 16 in 2005 to 7 in 2010, and the number of days was reduced from 51 to 31. For exports the number of days was reduced from 30 to 24. There was, however, deterioration for other sub-indicators, including the number of days required for getting construction permits that increased from 276 to 303 days, and the cost of getting a construction permit that increased from around 100 percent of per capita GDP in 2005 to around 1,200 percent in 2010. 3.29 The Bank‘s program missed the opportunity to be shaped by the 2006 Investment Climate Assessment, whose findings were released only in 2009. The implementation of private sector development reforms under PRSC4-8 was slow and faded out after the relatively easy low-cost reforms—removal of local fees for businesses—were picked up through PRSC-5. The limited gains from these early reforms proved to be difficult to maintain because the losses to local jurisdiction due to elimination of fees and taxes were not properly compensated for, leading to a conflict between local authorities and the reform agenda. Indeed, government informants indicated these legal changes were to be reversed. The complexity arising from the heavy legacy of over-regulating and overtaxing regulatory institutions that were present in Tanzania for the past decades was not properly taken into account in the design. 3.30 The preparation of a private sector development strategy was never completed, and the second phase of labor legislation that was also expected to be part of the policy dialogue was delayed. The Business Activities Registration Act was adopted by Parliament, allowing two of the three triggers in this policy area to be converted to prior action. But the government could not reach political consensus regarding the direction of reforms in business regulation, perhaps due to the lack of a reform champion, and thereafter commitment to the program waned. This circumstance precluded the newly adopted law from becoming an effective legislative framework and translating the system into a one-stop-shop for business registration, the end-goal output for support in this area. As such, the final trigger was dropped, and PRSC-7 was reduced in size by USD 10 million. Despite the purported emphasis on this policy area in early program documents, no additional triggers were considered in private sector development. A parallel investment project—the USD95 million Private Sector Competitiveness Project launched in 2006—appears to have yielded limited results in part due to the slow progress of PRSC-supported reforms. 42 3.31 Increased access of private sector to key inputs, in particular credit and ‗backbone‘ services. The second outcome area for the objective of growth and reduction of income poverty covers the program‘s contributions to energy and transport sectors. 3.32 A reliable and affordable supply of electricity was considered essential by the MKUKUTA for economic growth and poverty reduction in Tanzania. The PRSC, as articulated in the program document for PRSC-5, sought to help the government improve access to electricity for the private sector, reduce vulnerability of the sector to climatic conditions, achieve greater installed capacity, and help Tanzania Electric Supply Company (TANESCO) to achieve full cost recovery. The PRSC included three relatively vague triggers in the energy sector, which were focused on achieving financial recovery for TANESCO. In parallel, three investment projects complemented the PRSC by financing the expansion of generation capacity and provide emergency power to avert prolonged periods of load-shedding. All three triggers and associated prior actions were considered completed on time. 3.33 There was a mixed record during the period. On the positive side, Tanzania was able to expand generating capacity and reduce reliance on hydro power. Capacity increased from 980 MW in 2007 to 1072 MW in 2010 (short of the MKUKUTA target of 1278 MW), and 1309 MW in 2012. The share of population with access to grid and non-grid sources of electricity increased from a range of 8.1 to 11.3 percent in 2005 to a range of 10.1 to 14.2 percent in 2010, depending the assumptions you make on the number of people per household (it should also be noted that it is assumed that all access points are functional once they are constructed). On the negative side, the tariffs that can be charged by TANESCO are far below costs for renting much of the recently added additional capacity, making full cost recovery impossible. With regard to availability of the power supply,38 TANESCO data indicate a modest improvement in the duration of outages, but this has not been accompanied by reductions in frequency. The burden of power outages on the economy is estimated to have been as high as 4 percent of GDP (World Bank 2010). 3.34 Efforts to improve the autonomy of TANESCO yielded limited results. Stakeholders perceive TANESCO‘s commercial focus to have improved substantially, but still far short of that to be expected from a private commercial entity. Collection improved and the new regulator approved tariff increases, bringing them more in line with costs. But transmission and distribution losses remained high, rising from 21.0 percent in 2007 to 23.3 percent in 2010, and tariffs have not risen fast enough for TANESCO to achieve operational cost recovery by 2010 as targeted by PRSC-8. Indeed, as the unit cost of energy has increased, so has TANESCO‘s operational losses since 2010 (IMF 2013). The Energy and Water Utility Regulatory Authority is seen by stakeholders to have relatively strong capacity to monitor the industry, and was named best regulator in 2010 in a study commissioned by the Africa Forum for Utility Regulators, but it was unable to fully enforce rules for competitive procurement and reported that TANESCO‘s cost of providing power was higher than it would have been without sole source selection. 38 The Bank considered capacity utilization, which improved from 62% to 85%, as the indicator for power availability but this was an inappropriate indicator. 43 3.35 In the transport sector, the focus of the PRSCs was on improving the condition of the road network, and later, port efficiency. Prior actions covered four areas: enactment of the Roads Act; development of a Transport Sector Investment Plan (TSIP) with adequate provision for maintenance; adoption of a framework for public-private-partnership (PPP); and actions to reduce port congestion. These were intended to strengthen the institutional and policy framework for the transport sector, prioritize investment and maintenance spending, and reduce transport costs for the private sector. 3.36 Improvement in the condition of national and regional roads was far short of the target through 2010, with the share of national and regional roads in good condition rising from 47 percent in 2005 to 53 percent in 2009/10 (against a target of 75 percent), and no change in those in fair condition at 35 percent (against a target of 20 percent).39 The share of passable rural roads increased from 50 percent to 59 percent, far short of the PRSC and MKUKUTA targets of 70 and 75 percent. The shortcomings were likely accounted for by the shortfall in maintenance financing. There was an impressive burst in resources made available by a doubling of the fuel levy in 2007/08 following significant pressure by development partners,40 but after that the need to fund maintenance continued to increase but the funds available covered less than two-thirds of routine maintenance and less than 60 percent of periodic maintenance.41 Moreover, roads contracts were signed by government agencies that amounted to twice the allocated funding in 2009/10, a problem that cut into available resources for other development projects and led to the suspension of the signing of any new contracts. A framework for public-private partnerships, which was expected to encourage the private sector to fill the financing gap of 60 percent in a Transport Sector Investment Plan, was delayed and ultimately lacked clarity on the division of government responsibility among the various implementing agencies. To date, private sector investment in infrastructure has been minimal, and no parallel Bank-financed projects have sought to jumpstart these types of investments. 3.37 By contrast, there was real improvement in the efficiency of container operations at the Dar es Salaam port, with container dwell time falling from its peak of 20.1 days in 2008 to 12.3 days for local containers and 16.3 days for transit containers even as traffic increased. The issue of port congestion was elevated to the PRSC when there was no consensus at the sector dialogue level. The PRSC contributed to the improvement in port efficiency by encouraging an increased tariff and alternative means for port storage, but longer term issues regarding port efficiency do not appear likely to be addressed given the lack of commitment to the agenda by key counterparts. The Bank and government have agreed to pursue a process by which the role of the Ports Authority evolves from operator to landlord, but the Ports Authority does not intend to carry this out, which was the crux of the difference of opinion at the sector dialogue level in the first place. Moreover, port clearance requires approval by many players in the government, many of whom have not decentralized 39 More recent data indicates some improvement in 2011, although still well below the target. 40 The levy is extremely high by regional standards, but widespread tax evasion erodes the revenue collected (Gwilliam and others 2011). 41 Severe weather and the apparent reclassification of district roads as national and regional roads also may have contributed to this result. 44 authority to the ―single window‖ at the port, while many procedures still need to be automated and infrastructure remains insufficient (Tanzania Revenue Authority 2009). 3.38 Improved livelihoods in rural areas, where the vast majority of the poor live, through improved agricultural policies and expansion of rural infrastructure. The third outcome area for the objective of growth and reduction of income poverty covers agriculture and rural development and natural resources sectors, as well as some aspects of the transport and energy sectors which were addressed in the previous section. 3.39 Strong growth in the agricultural sector was expected to make possible sustained economy-wide GDP growth of 6-8 percent, and help to reduce income poverty, but this did not come to pass. Agricultural growth averaged 4 percent per year during 2006-10, short of the PRSC program target of 6 percent and far short of the MKUKUTA target of 10 percent. Even this modest outcome is uncertain as doubt was expressed by stakeholders about the integrity of the data on agricultural growth. Taking the data as accurate, agricultural growth is still only marginally faster than that of population growth, and appears to be, as under PRSCs 1-3, derived mainly from continued expansion of cultivated areas rather than increased yields, unrelated to the policies pursued under the PRSCs. 3.40 There was a mixed record in the areas of support. On crop board reform, the government halted the price-setting function of crop boards, limiting their functions to regulatory activities, an important step. But the government did not consolidate these gains, as amendments to crop board legislation were not adopted and strategic plans for crop boards were not completed. On performance-based budgeting at the local level, all 133 Districts qualified for performance-based ―top-up grants‖ for agriculture-related public investments by the end of the period. 3.41 On land reform, there was some progress, albeit slow. Decentralized land registration was scaled up in both urban and rural areas. About 7,000 villages out of about 11,000 were surveyed and registered, although participatory village land-use plans were prepared in only 1,000 of them. But the issuance of certificates of customary rights of occupancy for individual customary landowners was not carried out. The development of efficient land registries has been lagging, in part because local government authorities lack the capacity to develop land registry systems that match their responsibilities under the law. This has led to many land disputes and lack of transparency in land management practices. There is no instrument to track women‘s ownership of land, although the country is working toward such data with the implementation of a new information management system with support from a parallel Bank investment project. These limitations reflect the inadequate implementation of the land laws, particularly titling and registration (supported under PRSC-8), which has been fragmented through a series of small and somewhat disjointed pilots. 3.42 Government commitment with respect to the policy dialogue in agriculture declined appreciably under PRSCs 4-8, and several of the triggers were dropped or significantly diluted. Even in those areas where some progress was made, such as land registration, and reorientation of the crop boards towards regulation and away from marketing, there are, as yet, no perceptible outcomes in terms of improved access to credit and reduction of 45 ―nuisance taxes‖—small taxes levied by local government authorities on the marketing of agriculture. 3.43 Natural resource issues were highlighted in PRSC-7 and PRSC-8, in response to the findings of the Bank‘s Country Economic Memorandum of 2007. Defined to include tourism, mining, forestry, and fisheries, the sector was considered to be a significant income supplement to many Tanzanians, although much of it operated in the informal economy. The PRSC sought to initiate improvements in transparency and accountability in resource management through reforms of the licensing and concession regimes. 3.44 Outcomes with regard to transparency and the regulatory environment for all four sub-sectors were mixed. In forestry and wildlife, the level of revenues received from concessions increased from TZS 23 billion in 2007/08 to TZS 33 billion in 2009/10 (IEG 2011)42 pointing to gains following improved transparency in allocating licenses and concessions. But the more basic objective of creating an institutional framework for long term sustainable management was not achieved. In tourism and coastal resources management, the CAS Completion Report—the Bank‘s self-evaluation of the country program—for the 2007-2011 CAS period concluded that Bank support was unlikely to achieve its development objectives of improving the regulatory framework. In mining, despite the issuance of a new Mining Act (United Republic of Tanzania, 2009), no significant outcome in terms of improved regulatory framework was achieved, although mining sector growth accelerated, with an average annual growth during 2000-09 estimated at 13.7 percent. Tanzania is seeking validation for the Extractive Industries Transparency Initiative, which would help generate a more transparent framework, a situation widely supported by operators in the sector. Overall, despite meeting the triggers of the PRSCs and support from several IDA technical assistance projects, including GEF resources, improvement in the regulatory environment was limited. 3.45 To summarize, the overarching objective of poverty reduction did not materialize, and sustained growth cannot be attributed to PRSC-supported reforms. Apart from some real successes (including improving port efficiency and reducing crop board control of price- setting) intermediate outcomes in each of the three outcome areas was weak, and not improved by the slow pace of reforms supported by the PRSCs. Efficacy on this objective is rated modest. OBJECTIVE 2: IMPROVEMENTS IN QUALITY OF LIFE AND SOCIAL WELL-BEING 3.46 This objective aimed to improve quality of life and social well-being through improvements in access to education, health, and potable water supply, as well as mitigating the effects of HIV/AIDS. The PRSC supported the objective through efforts to enhance the sector dialogue among development partners and improve the monitoring and evaluation of service delivery. There was no direct support to the underlying government program in the sectors, nor dialogue regarding specific policies. Rather, the PRSC sought to encourage 42 The proposed performance indicator—the level of revenues received from concession and license in forestry as percentage of their estimated values—was not measured, nor was it sufficient to assess achievement of objectives in the sector. 46 sector-based dialogue among a number of development partners, and strong monitoring of the program, conditioning Bank support on a ―satisfactory‖ annual sector review in health, education, and water. As such, there was a long chain between the Bank‘s PRSC support, and development outcomes in the country. 3.47 The program document for PRSC-6 identified three outcome areas for this Objective, each with associated indicators:  Improved health status of the population.  Expanded enrollment and enhanced quality in all levels of education.  Improved well-being through better access to clean water. 3.48 Improved health status of the population. During 2006-2010 (PRSCs 4-8), health outcomes continued to improve. In spite of the declining budget share, per-capita public expenditure increased (due to the substantial increase in overall public expenditure) and was anticipated to reach USD 14 per person by 2010/11. The 2010 Demographic and Health Survey documents continued marked declines in infant and under-five mortality, putting Tanzania on track to achieve its Millennium Development Goals in these areas (the trend for under-five mortality is shown in Figure 3.1). These improvements were likely the result of the government program—there was significant increase in the percent of children under two receiving three doses of vaccines (from 71 percent to 88 percent in 2010) and the percent of births attended by a skilled health worker (from 41 percent in 2006 to 51 percent in 2010 and 63 percent in 2012, but still short of the PRSC target of 65 percent). Targeted vaccination rates and AIDS prevalence among 15-24 year olds which were set for 2010 had already been achieved by 2008, and national HIV prevalence fell from 4 percent in 2003 to 2.4 percent in 2008. Child nutrition indicators also continued to improve and there was some evidence of a beginning decline in maternal mortality (although large margins of estimation error preclude a definitive trend).43 Significant disparities across districts remain and, as highlighted by a review of the General Auditor‘s Office, little progress has been made in matching workload to staff deployment. Little progress is evident on other issues raised by a 2007 joint evaluation of the health sector in Tanzania (COWI and others 2007), including, for example, poor hospital management, and a weak Public-Private Partnership framework. 43 The 2010 Demographic and Health survey reported a decrease in child stunting from 38 to 35 percent and a marginal decline in underweight from 22 to 21 percent. The incidence of moderate anemia declined from 43 to 28 percent. The DHS also reported that the estimated confidence interval upper limit for maternal mortality declined from 578 to 545 per 100,000 over the same period. 47 Figure 3-1. Trend in Under-five Mortality in Tanzania Source: Tanzania DHS Surveys, Tanzania National Bureau of Statistics & Macro International via de Savigny and others 2011. 3.49 During 2007-09 the Bank had intended to shift its financial support to the health sector from the health sector basket fund (a pooling arrangement supported by most health sector donors) to general budget support through the PRSCs relying on the budget to ensure adequate allocations to the sector. This shift was conditional on (i) increased budget allocation to the health sector commensurate to the IDA resources shifted to the PRSCs; (ii) progressive improvements in procurement systems; and (iii) satisfactory implementation reviews of a health care waste management plan. In the event, there was progress on the second and third conditions, but health budget allocations (in terms of share of total expenditure) remained relatively steady and perhaps increased slightly, but this was insufficient for the Ministry of Health continue its support for the shift to GBS, and led the Bank to drop its plan to use the PRSCs as its main instrument for financial support to the sector.44 Nevertheless the PRSC attention to the health sector policy dialogue through the annual reviews did underscore the Bank‘s concern with health sector performance and contribute modestly to health sector outcomes. 3.50 Expanded enrollment and enhanced quality in all levels of education. Bank support for education through the PRSCs was underpinned by the first ever annual sector review in the sector prepared by the government with inputs from technical working groups, including civil society and development partners, and modeled on the experience with health sector reviews. The annual triggers in the education sector were the completion of a satisfactory annual sector review dealing with sector performance issues. 3.51 Overall, there were significant achievements in expanding, and then maintaining, high levels of enrollment at the primary and secondary level, and mixed evidence on quality. Primary net enrolment ratios peaked at 97.2 percent in 2007/08 from a baseline of 73 percent, falling back to 94 percent in 2010/11, just short of the ambitious MKUKUTA target of 44 A development consistent with the fact that, in 2009, the Bank proceeded to prepare a second installment of Additional Financing for the ongoing Health Sector Development Program II/ 48 universal enrollment. Primary completion and the pass rate on the primary school leaving exam fluctuated during the period (the trends for all three are shown in Figure 3-2). In addition, the number of qualified primary school teachers increased and there was improvement in pupil/teacher and pupil/classroom ratios, each a proxy for the quality of education. Independent quality assessments were mixed: a large household-based survey by a regional NGO45 found competencies of primary school leavers below those in Uganda and Kenya, while a set of school-based surveys by a government-sponsored regional initiative46 concluded that Tanzanian students performed above regional averages in reading and arithmetic and significantly improved performance between 2003 and 2007. Analytic work undertaken for the 2010 PER found that education developments in Tanzania represented good value for money overall considering that Tanzania would achieve the Millennium Development Goal for education at relatively low cost; but it also found that quality issues constituted the greatest threat to maintaining good value for money in the sector and that there was scope for improvement by shifting resources to those under-served districts which had demonstrated higher than average efficiency. Figure 3-2. Indicators of Progress in Education. 100 13 year olds who enrolled at age 7, Percent of 7 year olds, percent of 95 90 percent of exam-takers 85 80 75 70 65 60 55 50 45 40 2003 2004 2005 2006 2007 2008 2009 2010 2011 Net enrollment rate Primary completion Pass exam Source: Poverty and Human Development Report 2011 3.52 Public satisfaction with the extent to which the government was addressing educational needs showed a declining trend, with 85 percent indicating fairly or very well in 2005, followed by 79 percent in 2008 and 56 percent in 2012 according to the Afro barometer. 3.53 The secondary education net enrollment ratio continued to rise but at a slower rate than earlier in the decade growing from 24.4 percent in 2007 to 34.5 percent in 2010. Even this slowing growth put great strains on the overall system, with the capitation grant of TSh 45 Uwezo. 46 Southern and Eastern Africa Consortium for Measuring Educational Quality. 49 25,000, intended to contribute to learning materials and other non-salary support to secondary schools, being paid out at a small fraction of the mandated amount due to the financial squeeze on the secondary sector from the drive to universal primary enrollment and the remarkable expansion of the tertiary sector. After an initial deterioration in the ratio of secondary school pupils to qualified teacher, the ratio improved slightly toward the end of the period. Aggregate resources for secondary education appear to have been squeezed between primary and higher education student loans, and spread across a much larger cohort of students. The consequence was seen in a plummeting Form IV pass rate down from 26.7 percent in 2007 to 11.4 percent in 2009. 3.54 The MKUKUTA target for higher education was exceeded as the gross enrollment ratio jumped from 0.27 percent in 2005/06 to 8 percent in 2010/11. However, this remarkable growth threatens the fiscal sustainability of the whole sector through the commensurate growth of a student loan scheme which lacks adequate repayment mechanisms and is inherently regressive. 3.55 The Bank‘s contribution to these outcomes through the PRSCs was indirect, through enhancing the quality of the sector dialogue. Bank staff presented a proposed set of criteria to define a satisfactory sector review as part of the PRSC-5 documentation,47 but although the sector review undertaken for PRSC-6 was more substantive than the previous year, it fell short of these criteria in a number of aspects, primarily related to an inadequate integration of progress on results with reporting on outputs and financial allocations. Under PRSC-7 and PRSC-8, the Bank made it progressively more explicit that the annual sector review should focus on performance and on the linkage between results and sector funding and policies. The quality of the sector reviews did improve and were deemed satisfactory by the development partners. However other developments made the determination of sector review quality more difficult. The government clarified that it would only respond to a binary determination – satisfactory or not satisfactory, and the development partner sector working group decided to give equal weights to outcomes, resources and dialogue in determining the overall rating for the annual performance reviews. In the view of Bank staff and some development partners, the combination of these factors reduced the leverage of the annual sector review on GBS and the PRSCs by narrowing the assessment range. 3.56 As with health, the Bank dropped its intention to shift its primary financial support for the education sector from sector operations to the PRSC because it determined that its leverage was needed to improve sub-sector allocations and focus attention on quality issues. Bank staff concluded that this could be best accomplished through sub-sector financial support and policy dialogue combined with general budget support through the PRSC with its emphasis on the broad-based annual sector review carried out in conjunction with all other 47 These included: (i) timely inputs to the budget process; (ii) involvement of a wide range of stakeholders; (iii) assessment of performance against agreed benchmarks and targets; (iv) use of reliable information stemming from sector MISs, Government of Tanzania statistics or MDA implemented projects; and (v) a review of both retrospective and prospective actions. The definition of ―reliable information‖ subsequently proved problematic as it could be interpreted to exclude third party monitoring and data collection. 50 development partners in the sector.48 Overall, there was significant progress in expanding enrollment and mixed outcomes with regard to quality, but the direct contribution of the PRSC to education sector outcomes was modest. 3.57 Improved well-being through better access to clean water. Survey data, less frequently available but more robust than administrative data, show an improvement in the share of households using protected water sources between 2007 and 2010, from 45 percent to 57.8 percent in rural areas and 73 percent to 86 percent in urban areas. 49 This fell slightly short of the targets envisaged in the PRSC. There was also some improvement by 2010 in the share of the rural population which could access clean and safe water within 30 minutes from 40 percent to 48 percent over the same period, but this constituted less than half the rural population. Similar analysis for the urban population was subject to data limitations, but there appears to be no improvement in the share of urban population with access of clean and safe water within 30 minutes.50 3.58 For sanitation, the use of pour flush toilets rose from 1 percent to 1.6 percent in rural areas and from 6 percent to 15.3 percent in urban areas during 2007 and 2010, but overall there was an increase in the share of the population with no access to a toilet (even a pit latrine), from 7.5 percent to 13.7 percent. This suggests reduced access to the most basic sanitation facilities, primarily in rural areas, even as improved sanitation facilities became more common. The 2011 Poverty and Human Development Report emphasizes that the challenge to provide hygienic and safe sanitation for all Tanzanians is large. 3.59 With the exception of PRSC-6 when the water sector review was considered fairly satisfactory as in health and education, the triggers and prior actions were all met. Initially, the focus on jump-starting the annual sector review process was useful, particularly when standards were introduced for what constitutes a ―satisfactory‖ sector review. But over time, annual review meetings became excessively focused on negotiating the rating of the review process (satisfactory or unsatisfactory) rather than a substantive discussion of results and the policy actions critical to improve development outcomes. Ultimately, all parties faced strong incentives to rate programs satisfactory, given their interest in maintaining dedicated funding to the sector. Even when considering outcomes, administrative data were particularly weak, and consistently overestimated access to water in rural areas when compared to surveys, for two reasons. First, access was poorly defined and was sensitive to the assumption made on how many people on average had access to an individual household connection or water outlet. Second, the data provided from government sources assumed that all connections or outlets were functional, given the absence of a mechanism to identify functionality, but 48 The 2010 Secondary Education Development Program II APL marked the shift in strategy away from the PRSCs. 49 See the Water Sector Status Report 2011, Household Budget Survey 2007, and the Demographic and Household Survey 2010. 50 Urban figures for access in 2007 do not include Dar es Salaam. Its inclusion would have raised the baseline for 2007and decreased (or even make negative) the percentage change for 2010. 51 informants suggested there are a significant number of non-functioning water points, particularly in rural areas.51 3.60 In summary, there were some remarkable gains across all three sectors, but particularly in health outcomes and access to education. The PRSCs succeeded in upgrading and providing standards for the dialogue among the government and various development partners in each sector. This was an important contribution to improving the process, but far from sufficient to generate the overly ambitious desired outcomes articulated in the program documents. Reforms supported by the PRSCs were limited to early stage changes to process, and required significant changes to policy for the objectives to be achieved. The large balance of this work was undertaken by the government through sector interventions supported by the Bank and other development partners. Overall, because of the tenuous link between Bank support through the PRSC and these outcomes, efficacy is rated modest. OBJECTIVE 3. IMPROVED STATE OF GOVERNANCE AND ACCOUNTABILITY 3.61 The PRSC‘s efforts to support reforms of public sector performance and improved service delivery focused on strengthening governance and accountability. The program documents identify three public-sector reform indicators from the Bank‘s World Governance Indicators (on accountability, control of corruption, and government effectiveness). The first two improved somewhat, but did not meet their target, and the third deteriorated. These are weak indicators of outcomes, as they are scored relative to other countries, and so do not pick up absolute improvement or deterioration.52 The evaluation thus looks more closely at other indicators. 3.62 The program document for PRSC-6 identified the following three outcome areas in governance and accountability:  enhance the accountability of the state;  improve the effectiveness of public administration through better incentives for public servants and decentralization; and  improve public financial management. 51 A one-time water point mapping exercise, currently underway, will survey the existence and functionality of rural water points. But given this will only be an end-of-period single observation, household survey data is the most useful source to measure progress. 52 The Worldwide Governance Indicators aggregate scores from a number of primary sources on governance—mostly surveys—and then compares each country‘s aggregate score for the year with that of over one hundred other countries in the world. The indicator shows how many standard deviations the country‘s score is above or below the mean for all countries. Tanzania, like most low- income countries, is below the mean, so a less negative score indicates improvement. While giving a good estimates and confidence intervals for a country‘s relative position in a given year, the WGI is less useful for comparisons over time, because the changes result not only from changes in a country‘s absolute performance, but also from changes in the primary indicators that are used in calculating the country‘s raw score and from changes in the collection of comparator countries. 52 3.63 In the area of enhancing accountability of the state, primarily through control of corruption, anti-corruption legislation was enacted, a greater share of investigated cases were prosecuted, and a smaller share of court cases remained outstanding for more than two years. But these figures are probably misleading because only about one-sixth of allegations over the 2006-2010 period were investigated by the Prevention and Control of Corruption Bureau. 3.64 Indirect evidence suggests corruption may have increased. During the period, there was a series of high-level corruption cases, accounting for large sums of misappropriated funds. Perception ratings also point to a deterioration: the International Country Risk Guide rating fell from 3.5 in 2007 to 2.5 in 2010 on a 1-4 scale, with 4 indicating the lowest perception of corruption and the Bank‘s Country Policy and Institutional Assessment rating fell from 3.5 in 2007 to 2.5 in 2010 on a 1-5 scale, with 5 indicating the lowest perception of corruption as shown in Table 3-4. Transparency International‘s Corruption Perception Index (2011) showed improvement in 2011 after a rapid fall from the peak in 2007. Tanzania‘s score on the Worldwide Governance Indicators (relative to performance in other countries) improved but fell short of the program‘s target. The Afro barometer found the share of people believing the government was fighting corruption well fell from 62 percent in 2005 to 56 percent in 2008 to 34 percent in 2012. Implementation of new anti-corruption legislation—intended to extend the focus of anti-corruption efforts beyond national ministries to local government, civil society, and the private sector—faced strong opposition. An independent evaluation of anti-corruption efforts in Tanzania found that GBS dialogue has had modest influence on combating corruption (Norad and others, 2011). Table 3-4. Tanzania-ICRG and CPIA Ratings 2005 2006 2007 2008 2009 2010 2011 ICRG (1-4 scale) Bureaucracy Quality 1 1 1 1 1 1 1 Corruption 2 2.5 3.5 3 3 2.5 2.5 CPIA (1-5 scale) Budgetary and financial management 4.5 4.5 4.0 3.5 3.5 3.5 Revenue mobilization Efficiency 4.0 3.0 4.0 4.0 4.0 4.0 Public administration 3.5 3.5 3.5 3.5 3.5 3.0 Public sector transparency, 3.5 3.5 3.5 3.0 3.0 2.5 accountability, and corruption Sources: ICRG website; Bank CPIA ratings. 3.65 The second outcome area covers the PRSC‘s contributions to improving the effectiveness of public administration through better incentives for public servants and decentralization. With respect to improving civil service effectiveness, the PRSC-supported reform objectives focused on remuneration with the intention of putting an effective public service framework in place to provide a foundation for service delivery and improvements in poverty. The objectives were not met. The ICRG rating for Bureaucratic Quality remained steady, and the CPIA rating for Quality of Public Administration declined (see Table 3.4). This is consistent with the Worldwide Governance Indicator on government effectiveness, which deteriorated from -0.35 (2005) to -0.45 (2008), in contrast to the program target of 0.0 for 2010, although again this indicator was not appropriate to measure trends over time. 53 3.66 There were two main reasons for this result. First, there were weaknesses in design, as reflected in the repetition of public service pay reform triggers, some of them imprecise. Second, the strategy remained to increase pay across-the-board, an approach which had not succeeded during the first PRSC series, and which empirical evidence elsewhere showed as unlikely to succeed (IEG 1999 and 2008; Evans 2008). Success in meeting the intermediate outcome target of increasing pay as a proportion of the government‘s pay target (from 86 percent to the PRSC target of 90 percent, but not the MKUKUTA target of 100 percent) should be interpreted with caution. An alternative approach of targeting pay increases to positions with the greatest need (for example, in remote areas) and in agencies critical for delivery of prioritized services – which experience and analysis elsewhere shows to have greater chances of success – was not adopted. The program document for PRSC-8 notes the persistence of problems related to the structure of public service pay, emphasizing the distorted incentives that arise from allowances and poor targeting. 3.67 Local government reform. With Bank support, the government sought to strengthen the link between budgets of local authorities and MKUKUTA, while advancing decentralization. This was a long-standing challenge given Tanzania‘s history of extreme centralization. The share of the budget transferred directly to local government authorities increased, following the de-concentration of the secondary education budget, and the share of local government expenditures calculated on a formula basis increased from 16 percent in 2006/07 to 25 percent in 2009/10 (GBS Annual Review 2010). A system of intergovernmental transfers, which provides performance-based discretionary grants (Council Development Grants and Capacity Building Grants) to local government authorities, was disbursing by 2010. There was, however, limited progress made in providing autonomy in human resources management at the local government level. Moreover, transparency and accountability of use of funds was undermined by multi-channeled funding at the local level, and there were continued large inequities in the allocation of funds among districts (World Bank 2010c). Although the PRSC series had the objective of improving the effectiveness of decentralization, none of the five PRSCs directly supported policy measures through triggers or prior actions to promote this, a serious design shortcoming. Moreover, decentralization efforts seem to have declined in priority as an activity area for public sector reform, according to interviews with country officials. Although more money in the aggregate is going to local governments, the targeting to rural areas and accountability to central authorities has weakened (there never were substantial mechanisms for accountability to local populations). 3.68 Reforms in the third outcome area, improving public financial management, focused on procurement, external and internal audit, and budget and core public financial management (PFM) issues. Budget management is deteriorating, while auditing and procurement are becoming more thorough. 3.69 For procurement reform, the strategy was to implement the new law passed during the first PRSC series—to stop using a central tender board, to allow ministries and local governments to manage their own procurement, and to create a regulatory body to set and enforce standards for efficient and non-corrupt procurement. The number of agencies participating in the Public Procurement Regulatory Authority evaluations grew from 20 in 2006/07 to 100 in 2007/08 and to 174 in 2010/11. The proportion of agencies found in 54 compliance with PPRA regulations rose from 39 percent in 2006/07 to 50 percent in 2007/08 and reached a plateau around 70 percent in 2009/10 and 2010/11, while the number of agencies included rose from 91 in 2009/10 to 174 in 2010/11 (PPRA Annual Performance Evaluation Report, 2011). The 2009 Public Expenditure and Financial Accountability review rated competition, value for money, and controls in procurement as a B in 2009 compared to D+ in 2005. But that report and the 2011 DfID Risk Assessment noted the continuing problem of low compliance with procurement laws and regulations, stating that ―anecdotal evidence and the national corruption survey still question the integrity of the actual [procurement] selection process at both central and local government level‖ (DfID 2011). 3.70 External Auditing was strengthened through the approval of the Public Audit Act in 2008, increasing the independence, powers and responsibilities of the Controller and Auditor General. The PEFA score for the scope, nature, and follow-up of external audits improved from D+ in 2005 to B in 2009, and the country is one condition from ascending to (African Organization of Supreme Audit Institutions) AFROSAI level 3 status, which only two other sub-Saharan African countries have achieved. To reform internal audit in 2008 the government moved the internal audit function from the office of the Accountant General to make it independent, with direct reporting to the Permanent Secretary of the Treasury. Auditing showed progress in terms of increased coverage capacity, independence of the National Audit Office, and meeting regional and international standards. The audit office is well respected in the region, as evidenced by Tanzania being elected chair of AFROSAI East for 2011-2014 and the audit office winning the Swedish National Audit Office 2010 award for the best performance audit report, and sitting as a member of the United Nations Audit Panel. Its annual reports were on time since 2005 and contained substantive analysis and frank critiques. Outcomes reported by the audits (a measure of the quality of public financial management) showed progress in increasing the proportion of ―unqualified opinion‖ until 2009, which reversed in FY 2010, but recovered for FY 2011. Some important ministries, like Finance and Education, were still not getting ―unqualified opinions‖. Also ―adverse opinions‖ in audits of local governments continued to be prevalent (TWAWEZA, 2010). These reflected PFM problems in the audited agencies, but also the ability of the auditors in identifying them. However, government responsiveness to National Audit Office recommendations has been lacking. The audit office report on FY2010/11 said the following: ―Most of the recommendations issued in the previous year‘s General Report of the Central Government were not responded to, which is an indicator of lack of seriousness on the part of the Government in implementing those recommendations.‖ (p. xxiv) 3.71 The government‘s umbrella PFM reform program began to slow down in 2006-07 as government commitment deteriorated, so that there was little further progress regarding budget and core public financial management issues. The focus was on ensuring that the approved budget was in line with policy objectives and actual expenditures were consistent with the approved budget. Outstanding issues from the first series of operations, such as inadequate budget reporting by the local governments, were not followed up in the second series. 3.72 Starting in July 2008, there were improvements in specific systems—cash management, payment systems, coverage of the Integrated Financial Management System, and budget classification. Despite this, the CPIA and Public Expenditure and Financial 55 Accountability ratings for PFM deteriorated over the period, and the Open Budget ratings did not make net improvement.53 Recurrent budget deviation at vote level improved, but remained above the target of 10 percent, as shown in Figure 3-3 (the same as Figure 2-4. in Chapter 2). The main recipients of reallocated funds were those Ministries with large staff complements—the Education, President‘s Office and Cabinet, Defense, and Police—because most of the reallocations were for salary arrears and adjustments. Figure 3-3. Total Absolute Deviation between Approved Budget Allocations and Actual Expenditures Declined Under the Operation 30% 25% 20% 15% 10% 5% 0% Source: Public Expenditure Reviews 2007 and 2011 3.73 The early years of the second series of PRSCs saw the adoption of budgets in line with MKUKUTA priorities, as well as satisfactory budget execution. The situation changed in later years. But commitment to budget execution and alignment of expenditures to MKUKUTA deteriorated and outcomes were worse than expected in PRSCs-7 and -8, although the situation seems to have improved in 2009/10 (Public Expenditure Review 2011). The allocations for capital spending remained low, about 18 percent of the total budget, and allocation for infrastructure maintenance declined, despite unreliability and low accessibility of electricity and rural roads. Spending on wages and salaries remains high at 32 percent of total; this is down from the peak of 34 percent in 2008/09 but still above the target and the pre-2008 levels. 3.74 The PRSC 4-8 series was unable to greatly influence policy performance to enhance governance and accountability in the public sector. Policy dialogue and assistance for reforms in civil service, local government, procurement, control of corruption, public finance management and economic stabilization resulted in little progress, as reflected in the key indicators. Support to external and internal audit yielded substantial improvements in the 53 The International Budget Partnership has published Open Budget Survey ratings that assess budget transparency biennially since 2006. 56 system, an important achievement, but government response to audit recommendations was minimal. Overall the efficacy of governance and accountability reforms in Tanzania was modest. Ratings OUTCOME 3.75 The relevance of the objectives of the PRSC 4-8 is assessed as substantial. Objectives were strongly aligned with, indeed identical to, the government program and consistent with the objective of the Bank‘s 2000 CAS, as well with subsequent CASs. In addition, the objectives of the PRSC series were drawn to correspond with the multi-donor performance assessment framework (PAF). But the program did not articulate its priorities within the government‘s program in order to guide the program as circumstances changed. 3.76 The relevance of design of the PRSC series was modest. The framework appropriately included a modest number of triggers and prior actions. These were derived from the policy dialogue emanating from the MKUKUTA. But the sectoral scope of the series was too wide-ranging and lacked clear prioritization—the program covered almost all the policy areas highlighted in the MKUKUTA. The results framework, much improved from the first series, was only fully developed by the third operation. Moreover, the operations should have more directly addressed the challenges of decentralization in delivering services, with better linkages to the parallel investment project. 3.77 The achievement of the objectives of PRSC 4-8 was mixed. With respect to growth and reducing income poverty, efficacy was modest. There are no up-to-date data to measure progress toward the primary objective—to reduce poverty in rural areas. It is very unlikely that the target was met because the drivers of economic growth have not changed much since 2007 when analysis showed a very low elasticity of poverty to GDP growth. Efficacy of improving quality of life and social well-being is also modest because of good outcomes in health and mixed outcomes in education and water and sanitation, and limited plausibility linking the PRSCs to these outcomes. Efficacy of improving governance and accountability was modest, with little progress in improving public financial management, improving the effectiveness of public administration, and enhancing the accountability of the state. 3.78 How far would Tanzania have gone in achieving its objectives without the PRSCs? Implementation of many of the specific policies promoted through dialogue under the program was slow—for instance in agriculture, transport, and private sector development. While there was some initial progress in governance reforms, there was substantial backsliding in the later years. Thus the program appears to have contributed little to national- level results. Without the PRSCs, Tanzania‘s IDA envelope would have been delivered either through additional sector investment loans, or a larger scale of the investment loans already under implementation. Given that one of the primary barriers to progress was related to government ownership—as agreed policy actions were dropped, diluted, and even likely to be reversed—it is unlikely that the specific reforms pursued under the PRSC series would have been better implemented through AAA or investment lending. Indeed, many of the sector-specific policy issues were ―elevated‖ to the PRSC precisely because they had not 57 been addressed at the sector level. If these sector level reforms had not been elevated, they would have faced the same political economy and ownership constraints as they did under the PRSC. Cross-cutting reforms were pursued through the PRSC dialogue specifically because they could not be addressed at the sector level. Aside from policy and institutional changes, allocation of Tanzania‘s IDA resources to investment lending may have yielded more tangible results at the sector level, for instance through direct investment in new infrastructure or the social sectors. 3.79 There were significant shortcomings in the program‘s achievement of objectives, particularly with regard to the first and third objectives. Despite strong growth, the likely poverty impact was minimal. None of the targets for improvement in governance were achieved. Given this modest efficacy, and some shortcomings in the design of the program, the outcome of the PRSC series is rated moderately unsatisfactory. Table 3-5 summarizes the subcomponents of the rating. Table 3-5. Summary of outcome ratings for PRSCs 4-8 Dimension of rating Rating Relevance of objectives Substantial Relevance of design Modest Achievement of objectives Growth and reducing income poverty. Modest Improve quality of life and social well-being. Modest Governance and accountability. Modest Overall outcome rating Moderately unsatisfactory RISK TO DEVELOPMENT OUTCOME 3.80 As a result of weakening government commitment to the established modalities for dialogue under the PRSC, modest ownership of specific policy actions (as triggers were increasingly dropped or diluted), and high aid dependence during implementation, the risk to development outcome is significant. 3.81 A fundamental internal risk that could affect the development outcome of PRSC 4-8 is waning government commitment with respect to the established modalities for dialogue on policy reforms, especially the analytic processes that underpinned the reform effort. Specifically, the PER—which provided a platform for the Bank, key government counterparts, as well as other development counterparts to engage in an intense dialogue on public finance management and service delivery issues—came to be perceived by government officials as externally owned. The PERs were still produced on an annual basis with officially joint authorship between the Bank and the government, but the primary vehicle for policy dialogue on core issues has been transferred to joint annual reviews of general budget support, a process that is seen by some senior government officials as insufficient.54 Under the follow-up PRSC series, the Bank reports that it has worked with the 54 This development could be traced to three factors: First, some development partners preferred a broader authorship and viewed the PER as a Bank product. Second, changes in the composition of 58 government to revamp the PER process, and that early signs suggest that the government has begun to expand its commitment to the dialogue. The proliferation of parallel strategy documents, such as the National Development Plan, manifests parallel movements for ownership.55 In sectors where specific policy actions were targeted, implementing agencies did not champion the reforms in question, so that triggers were met, if at all, in letter rather than in spirit, and anticipated outcomes were not realized. 3.82 The second internal risk pertains to implementation modalities at the sectoral levels. For example, risk to development outcome in education remains high. While the evidence is mixed, the record of achievement in primary schools is highly variable and household-based survey data suggest that performance levels may be lower than the school-based survey data gathered through a government-sponsored regional initiative indicate.56 Similarly, the declining share, in recent years, of public expenditures devoted to the health sector, even as population grows at a rapid pace, suggests that the development outcomes are at risk. 3.83 The key external risk to sustaining the high levels of service delivery supported by the PRSCs is that the country remains highly dependent on aid flows even to finance recurrent expenditures. Domestic resource mobilization is low, although increasing, and so the government has faced challenges as some donors have begun to reduce their general budget support. BANK PERFORMANCE 3.84 Quality at Entry: Of the five operations in the series, the Bank‘s Quality Assessment Group assessed only PRSC-5 for quality at entry, and gave it a highly satisfactory rating. The operation was commended for best practice in its design of the mechanisms for harmonizing aid. Indeed, development partners and the Bank were on the leading edge of implementing the Paris Declaration principles of donor harmonization, with a joint strategy and a complex system for dialogue. Preparatory work by Bank staff on all PRSCs was intensive. Fiduciary aspects were addressed directly—for example, under PRSC-6, the handling of potential fiduciary problems stemming from alleged corruption at the Bank of Tanzania was made a condition for negotiations. And the overall approach was strategically relevant. 3.85 The Bank carried substantial analytical and advisory activities (AAA). Areas covered included nutrition analyses, private sector development, trade, growth, poverty, agriculture, and public expenditures. While these reports were of high quality, the Bank in an internal review was concerned that they did not always reach the policy makers. AAA was less intensive during the latter part of the series (the last Country Economic Memorandum and staff in both the Bank and government sides have weakened the dynamics of the partnership. Third, the new decision-makers on the part of government viewed the process as too intensive, with excessive transaction costs. 55 In agriculture, the Agriculture Sector Development Program is being overshadowed with the recent ―Agriculture First‖ agenda. 56 The household-based survey was fielded by the regional NGO Uwezo. The school-based survey was fielded by the Southern and Eastern Africa Consortium for Measuring Educational Quality. 59 Poverty Assessment were begun in 2007 and completed in 2010). The annual PER was prepared jointly and in close consultation with key government officials at the start of the series. The technical quality of the PERs was high, and the government used them to guide its budget operations. However, the participation of the Ministry of Finance team in their preparation and the accompanying policy discussions declined significantly during the second series as the process came to be seen as too intensive, a change that was seen to diminish the quality of the dialogue.57 3.86 The chain linking the objectives of the program and expected activities to help achieve them was not adequately specified across sectors, and the monitoring and evaluation system was not designed to provide regular feedback on progress. This is a challenge in implementing PRSCs while subscribing to Paris Declaration principles—the Bank adopted and adjusted a country-owned and donor-harmonized results framework with significant shortcomings in the belief that it would lead to better results than a stronger framework with less ownership. This is consistent with a finding from the Bank‘s recent retrospective on development policy lending, which found weaker results frameworks in countries with joint budget support (World Bank 2012b). On balance, these were moderate shortcomings that require elevated attention. Quality at entry is rated moderately satisfactory. 3.87 Quality of Supervision: Supervision and preparation of the annual PRSCs took place in close collaboration with other donors and were conducted in accordance with the review mechanisms established under MKUKUTA, with the Bank playing a central role. For most of the program areas, supervision and dialogue were carried out on a continuous basis in the context of sectoral or thematic programs. This included the PER process, PFM Reform Program, PSRP/MKUKUTA, Agriculture Sector Development Program, the Business Environment Strengthening program and sectoral programs. A joint general budget support (GBS) review was conducted on an annual basis for an overall assessment of progress made in each of the program areas. Progress toward achieving the objectives of general budget support was centered on the performance assessment framework (PAF), for which the Bank was the leading technical adviser. The harmonized GBS system helped the government reduce transaction cost with development partners, and gave the Bank a mechanism to materially affect the thrust of the government‘s partner-financed program. At the same time, as has been documented in other countries (IEG 2010; World Bank 2012b), the coordination process among development partners was complex and costly in terms of Bank staff time and resources as the system evolved with increasingly comprehensive Performance Assessment and M&E frameworks, and required broad consultation for even minor adjustments. The government‘s reform program established quantitative and qualitative benchmarks for the implementation period of the MKUKUTA. Starting with the third operation, the PRSC program adopted its own set of targets that were linked but not identical to MKUKUTA targets (the M&E system is assessed below). 3.88 Both MKUKUTA and the joint donor strategy were based on the presumption that income poverty had been reduced during the years of rapid growth in the early 2000s. When, after a long period in which no new reliable data on poverty was available, it became clear 57 The Bank reports that it has since renewed the process for joint preparation of PERs, with early signs of success. 60 that this was not the case, the strategies themselves lost credibility, according to government and development partner informants, and development partners reformulated their programs. This abrupt change may have been a rush to judgment, as the various measures of the poverty record, based on different price indices, suggested different trends. Yet the Bank did not prioritize supporting the country to collect robust data more frequently. 3.89 The program document for PRSC-6 suggested a renewed focus on agriculture and rural development and private sector development in order to stimulate poverty-reducing growth, a welcome change. In practice, however, the only trigger for private sector development for the final three operations of the series was dropped, while the three triggers for agriculture were diluted, dropped, and partially completed. 3.90 The Bank was confronted with a deterioration of government commitment to the reform effort, and lacked the leverage to reverse it. There was a tension between achieving the implementation of important policy reform objectives on the one hand, and the stated commitment—consistent with the Paris Declaration—of external partners, including the Bank, to provide regular, predictable financing on the other. In the event, the Bank reduced the size of two operations by USD10 million late in the budget cycle to signal to the government that there were consequences to missing triggers,58 and responded to the government‘s protest to this unpredictability by drafting a memo that explained the Bank would adjust indicative amounts based on in-year program performance. The adjustments appear to have had an unintended effect of eroding the trust of government counterparts without materially affecting the program. Government officials considered the reductions as an unwarranted nuisance, but did not view them as a major impediment to financing or implementing the program. Given this mixed performance, including moderate shortcomings, the quality of supervision was moderately satisfactory. 3.91 Overall, Bank performance is rated moderately satisfactory. BORROWER PERFORMANCE 3.92 The Ministry of Finance successfully managed the process of conducting the annual MKUKUTA national policy dialogue, ensuring that the process was participatory for NGOs, academics, and other development partners. It also successfully managed the deliberations of the complex general budget support annual review process, starting from the sector review, including interactions with the GBS donor group to agree on assessment outcomes of current performance assessment frameworks and jointly formulate contents of future PAFs. 3.93 But the commitment in central ministries to pursue important cross-cutting reforms, particularly in the governance and accountability objective, deteriorated as a new government administration and project counterparts took leadership during the initial year of the series, and key reformers associated with the first PRSC series process moved into different roles. 58 At USD115 million, the size of PRSC-8 was much lower than the USD400 million initially anticipated in the CAS. This reduction, however, was primarily the result of general disenchantment with GBS and the PRSC as the primary instrument for policy dialogue rather than a shortfall on specific policy actions. 61 Moreover, the willingness or ability of central ministries to influence line ministries in the implementation of PRSC-related reforms in other areas was less than anticipated, and so the program was beset by delays and modest backsliding. 3.94 This preferred modality by the government for policy dialogue—implementation and monitoring of the MKUKUTA and the GBS—created potential conflict of interest in sector reporting, undermining the quality of the dialogue. Government representatives in the annual sector reviews of specified policy areas were primarily from line ministries, while their counterparts from the development partner agencies were primarily sector specialists. Development partners reported that this was conducive to a lack of candor in performance rating, since acknowledgment of poor results would be expected to lead to reduced financial support for the sector concerned. Further, some policy actions agreed to by the central ministries were not then pursued with vigor by line ministries. Finally, over time government officials became accustomed to and increasingly expected annual general budget support disbursements from development partners without due regard to the performance of the program. The expectation that it was the responsibility of development partners, under their Paris Declaration commitments, to provide the predictable flow of resources in this fashion generated significant tension in the relationship with the Bank, and contributed to the erosion of trust. 3.95 During the PRSC period non-income indicators of poverty, such as access to education and health services, improved. Public spending on education and health also increased slightly as a share of GDP. While borrower performance in supporting these basic services is remarkable, this was not a significant focus of the PRSC policy dialogue and prior actions. 3.96 Borrower performance thus had significant shortcomings, and is assessed as moderately unsatisfactory. MONITORING AND EVALUATION 3.97 Design: The design of the M&E for the PRSC was weak at the start of the series because it needed to await the development of the M&E for the government‘s MKUKUTA to which it is aligned. Over time an elaborate and comprehensive M&E structure for MKUKUTA was jointly developed by the government and development partners. But counterparts indicated it was largely donor-oriented, producing M&E outputs intended to satisfy donors rather than influence policies. Because it was targeted as a support for the GBS system of aid coordination, it was integrated with the budget process, notably through (i) implementation of a new national statistical master plan under the auspices of the National Bureau of Statistics; (ii) implementation of an annual Living Standards Measurement Study and a panel survey instrument, started in 2009 that has yet to provide meaningful data; and (iii) supporting the strengthening of the internal reporting requirements of the government, in particular, the output and performance reports associated with the PAF and the now well- established biannual Poverty and Human Development Reports. The PAF is a matrix of commitments consisting of three categories monitored by the Cluster Working Groups throughout the year: (i) Underlying Process; (ii) Temporary Process Actions; and (iii) Outcome Indicators. The M&E system was over-designed, and not practically linked to the 62 capacity and timing constraints for implementation and use., though complex, was well thought through and had the potential for generating useful data 3.98 M&E Implementation: Gathering data through the government M&E system and the associated performance assessment framework for monitoring progress on MKUKUTA posed substantial challenges. The low frequency of the surveys was a major limitation in the generation of accurate data on a timely basis. Indeed, strategic priorities of both the government and development partners—relying on poverty simulations—were distorted due to infrequent feedback on progress on poverty reduction. Furthermore, the vague definition of certain concepts made interpretation of data difficult. For example, progress related to access to clean water or electricity depended heavily on assumptions regarding how many people can access a given water or electricity hook-up, and further presumed that all existing connections were functional. The M&E system associated with PRSCs 4-8, therefore, had significant weaknesses with regard to data generation, quality, and use. 3.99 Utilization: The data derived from the government M&E system served as the basis for annually reviewing progress of the MKUKUTA. However, much of the effort in preparing and carrying out the review focused on inputs and processes rather than outputs and results (World Bank 2009a). Little of the data generated, therefore, provide useful feedback on the effectiveness of the program. On the other hand, the Poverty and Human Development Reports provided a useful synthesis of the available data on the country‘s key development priorities, and its analysis was independent and candid. 3.100 The quality of monitoring and evaluation is rated modest. 4. Lessons 4.1 The implementation of PRSCs 1-8 spanned nearly a decade during a period of sustained macroeconomic stability and fast economic growth, offering lessons both for Tanzania and other countries with similar development challenges. 4.2 Lesson 1: Direct and sustained focus and clarity of vision over the entire course of implementation of PRSC programs that cover many years is required to facilitate lasting reforms particularly with regard to public sector governance. There is a significant risk of backsliding when attention is shifted to other priorities. Yet joint budget support is a messy attempt to reach a common understanding, always subject to change, with a wide range of domestic stakeholders and international development partners. In Tanzania, intense dialogue and very high government commitment led to early successes in improving public financial management and budget execution in the first PRSC series. But some achievements stagnated or were reversed in the second series, when attention was diluted to include a broad range of sectoral reforms and other aspects of governance. 4.3 Lesson 2: PRSCs can make little progress without a foundation of trust between the Bank and the government, and among development partners at the institutional and individual levels. Without sufficient trust, negotiated and agreed-upon triggers can quickly be transformed into more heavy handed conditions, whereby the level of future support is 63 determined by strict adherence to pre-identified policy changes. In these circumstances, a scaled-back operation, more tightly focused around areas of agreement might gain more traction than one used as an enforcement mechanism for previously agree reforms that no longer have political backing. 4.4 Lesson 3: PRSC programs risk becoming irrelevant if the design, scale, and scope are not adjusted as conditions change. The advance expectation that such changes will occur may smooth the transition should events warrant a scaling back of the program. Circumstances changed substantially during the second PRSC series in Tanzania: reform champions in government and among development partners moved on; the presumption that sustained growth of 6-8 percent would lead to significant poverty reduction proved false; a new President was elected; and the country was affected by higher food and fuel prices and by the global financial crisis. The mechanisms for dialogue also need to evolve so that the program can continue to support and influence policy decisions at the highest strategic level. To avoid misunderstanding, the Bank and government could plan in advance to recalibrate the program at a set point during implementation. 4.5 Lesson 4: The lack of well-defined and realistic objectives and a fully developed results framework makes adjustment to changing circumstances difficult. A results framework that clearly explains how policy actions are expected to lead to desired outcomes, in light of the support provided through investment lending and by other development partners, can provide a results-focused discipline to the program, improving selectivity in the process. 4.6 Lesson 5: Regular reporting on key outcome indicators is essential so that data can drive adjustments to the PRSC program. Adequate investment in statistical capacity and the creation of a robust monitoring and evaluation system that addresses each step of the results chain to identify the key drivers of success or failure, while remaining focused on a few key outcome indicators, can help guide long PRSC series as circumstances evolve. Policy makers need to be aware of trends in outcomes in order to adjust the program accordingly. In Tanzania, reliable data on, for example, income poverty were available only once every seven years or so. Interim poverty simulations suggested significant poverty reduction as a result of sustained growth, but these projections turned out to be inaccurate. Household surveys every 3-4 years could provide much more timely and meaningful data to policymakers to allow for evidence-based adjustments during implementation. 4.7 Lesson 6: Transitioning to the use of PRSCs as the primary instrument for resource transfer is only likely to be successful in rare circumstances, when strong central ministries can fully pursue a sector-inclusive dialogue representing the perspectives of key players throughout government. PRSCs themselves have very limited leverage in sector ministries, and the number and specificity of sector-specific triggers are necessarily curtailed when many sectors are covered. 4.8 Lesson 7: Elevating to the national dialogue sector-specific structural reforms that stalled at the sector level does not guarantee their implementation. Even as the depth of sector exposure and experience permitted the sector content to be brought opportunistically into the PRSC, there is no evidence that this accelerated sector reforms or 64 achieved greater cross-cutting momentum. Indeed, many of the same obstacles can prevent these reforms from taking hold, as was the case in energy, transport, agriculture, and water. Moreover, policy reforms committed to by central ministries are likely to be seen by implementing agencies, less vested in the PRSC program, as externally-imposed conditions, and subverted rather than adopted. 65 References Agrawal, Nisha, Zafar Ahmed, Michael Mered, and Roger Nord. 1993. ‖Structural Adjustment, Economic Performance, and Aid Dependence in Tanzania.‖ World Bank Policy Research Working Paper 1204, Washington DC. Alkire, Sabina and Jose Manuel Roche. 2013. ―How Multidimensional Poverty Went Down: Dynamics and Comparisons.‖ Oxford Poverty and Human Development Initiative, Oxford. Baum, Warren C. and Stokes M. Tolbert. 1985. Investing in Development – Lessons of World Bank Experience. Oxford: Oxford University Press Bourguignon, Francois. 2003. ―The Growth Elasticity of Poverty Reduction: Explaining Heterogeneity Across Countries and Time Periods‖ in Inequality and Growth: Theory and Policy Implications, ed. Eicher, Theo S. and Stephen J. Turnovsky. Boston: Massachusetts Institute of Technology Collier, Paul. 1991. ―Aid and Economic Performance in Tanzania‖, in Transition in Development: The Role of Aid and Commercial Flows, ed. Uma Lele and Ijaz Nabi, San Francisco: ICS Press. Confederation of Tanzania Industries. 2011. ―Challenges of Unreliable Electricity Supply to Manufacturers in Tanzania,‖ A policy research paper submitted to the Energy Sector Stakeholders in Advocacy for Ensured Reliable Electricity Supply to Tanzania Manufacturers. Dar es Salaam. COWI, Goss Gilroy Inc., and EPOS Health Consultants. 2007. ―Joint External Evaluation of the Health Sector in Tanzania, 1999-2006,‖ Denmark Ministry of Foreign Affairs. Copenhagen. de Savigny, D., Burford, M., Smith, T., Masanja, H., & Wak, G. 2010. INDEPTH 2010 AGM Millennium Development Goals Keynote Speech. Accra, Ghana. Development Consultants and Policy Analysts, DAIMA and Overseas Development Institute, ODI. 2004. ―Joint evaluation of General Budget Support – Tanzania 1995-2004, Final Report.‖ Dar es Salaam. Development Partners Group and United Republic of Tanzania. 2011a. ―Joint Government and Development Partners Roadmap to Improve Development Cooperation in Tanzania.‖ Dar es Salaam: Tanzania Ministry of Finance. Development Partners Group, DPG and United Republic of Tanzania. 2011b. ―Tanzania – Implementing the Unfinished Aid Effectiveness Agenda for Greater Development Effectiveness.‖ Dar es Salaam: Tanzania Ministry of Finance. Economic and Social Research Foundation. 2005a. ―Assessment of the Relevancy and Adequacy of Existing Policies on Promoting Agricultural Development.‖ Dar-es-Salaam: Tanzania Ministry of Agriculture and Food Security. Economic and Social Research Foundation. 2005b. ―Enhancing Aid Relationship in Tanzania: IMG Report 2005.‖ Dar es Salaam: Independent Monitoring Group. Foster, V, and J. Steinbuks. 2009. ―Paying the Price for Unreliable Power Supplies: In‐House Generation of Electricity by Firms in Africa,‖ World Bank Policy Research Working Paper Series 4913, World Bank, Washington, DC. Gwilliam, Kenneeth, Heinrich Bofinger, Richard Bullock, Robin Carruthers, Ajay Kumar, Mike Mundy, Alberto Nogales, and Kavitha Sethi. Africa’s Transport Infrastructure. Washington DC: World Bank. Helleiner, Gerald. 1999. ―Tanzania Assistance Strategy: Critical Issues,‖ Dar es Salaam: Tanzania Consultative Group. Helleiner, Gerald, Tony Killick, Ibrahim N. Lipumba, Benno Ndulu, and Knud Erik Svendsen. 1995. ―Report of the Group of Independent Advisors on Development Cooperation between Tanzania and its Aid Donors.‖ Copenhagen: Danish Ministry of Foreign Affairs IEG. 2005. 2004 Annual Review of Development Effectiveness. Washington, DC: World Bank. 67 ——. 2010. Poverty Reduction Support Credits: An Evaluation of World Bank Support . Washington, DC: World Bank. ——. 2011. Project Performance Assessment Report on Forest Conservation and Management Project in Tanzania. Washington, DC: World Bank. International Development Association and International Monetary Fund. 2000. ―Tanzania: Decision Point Document under the Enhanced Heavily I ndebted Poor Country Initiative.‖ Washington, DC: World Bank. Independent Evaluation Office (IMF) and Operations Evaluation Department (World Bank). 2004. ―The United Republic of Tanzania: Evaluation of the Poverty Reduction Strategy Paper (PRSP) Process and Arrangements under the Poverty Reduction and Growth Facility (PRGF).‖ Washington, DC: World Bank. (available at http://www.worldbank.org/oed/prsp or www.imf.org/ieo.) International Monetary Fund. 1996. ―United Republic of Tanzania: Staff Report for the 1996 Article IV Consultation.‖ Washington, DC: IMF. ——. 2006a. ―United Republic of Tanzania: Fifth Review Under the Three -year Arrangement Under the Poverty Reduction and Growth Facility.‖ Washington, DC: IMF. ——. 2006b. ―United Republic of Tanzania: Ex Post Assessment of Longer -Term Program Engagement.‖ Country Report No. 06/198, Washington, DC: IMF. ——. 2010. ―Tanzania: First Review under the Policy Support Instrument.‖ IMF Country Report No. 10/351. Washington, DC: IMF. ——. 2011. ―United Republic of Tanzania: 2011 Article IV Consultation and Second Review under the Policy Support Instrument.‖ IMF Country Report No. 11/105. Washington, DC: IMF. ——. 2013. ―United Republic of Tanzania: Fifth Review Under the Policy Support Instrument.‖ IMF Country Report No. 13/12. Washington, DC: IMF. Irish Aid & Department for International Development. 2011. ―Country Program Assessment 2004 -09.‖ Dar es Salaam: Iris Aid. Kitabu, Gerald. 2009. ―Factors hindering Tanzania‘s democratic process.‖ The Guardian of Tanzania, June 18. Marrissey, Oliver and Vincent Leyaro. 2007. ―Distortions to Agriculture Incentives in Tanzania.‖ Agricultural Distortions Working Paper #52. Nottingham: University of Nottingham. Ministry of Infrastructure Development. 2008. ―Ten Year Transport Sector Investment Programme.‖ Dar es Salaam: Tanzania Ministry of Infrastructure and Development. ——. 2009. ―Short Transport Sector Investment Plan (TSIP 3 -Year Rolling Plan 2009/10 – 2011/12).‖ Dar es Salaam: Tanzania Ministry of Infrastructure and Development. Ministry of Agriculture, Food Security and Cooperatives. 2011. ―Evaluation of the Performance and Achievements of the Agricultural Sector Development Programme.‖ Dar es Salaam: Tanzania Ministry of Agriculture, Food Security and Cooperatives. Millennium Challenge Corporation. 2011. ―Tanzania Growth Diagnostic: Partnership for Growth - A joint analysis for the Governments of the United Republic of Tanzania and the United States of America,‖ Washington DC: Millennium Challenge Corporation. Norwegian Agency for Development Cooperation, Asian Development Bank, Danish International Development Assistance, Swedish Agency for Development Evaluation, Swedish International Development Co-operation Agency, and U.K. Department for International Development. 2011. ―Joint Evaluation of Support to Anti-Corruption Efforts: Tanzania Country Report.‖ Oslo: Norwegian Agency for Development Cooperation. Nyangoro, Julius. 2006. ―The 2005 General Election in Tanzani a: Implications for Peace and Security in Southern Africa.‖ Institute for Security Studies Paper 122. Dar es Salaam. 68 Operations Evaluation Department. 1998. Tanzania, agriculture, and the World Bank: An OED Review . Washington, DC: World Bank. Overseas Development Institute and Daima Associates Limited. 2005. ―Joint Evaluation of General Budget Support – Tanzania 1995-2004, Final Report: Report to the Government of Tanzania and to the Poverty Reduction Budget Support Development Partners.‖ London and Dar es Salaam. Organisation for Economic Co-operation and Development, Development Assistance Committee. 2004. ―Evaluation Framework for General Budget Support: Framework for Country -level Case studies.‖ Report to the OECD-DAC Evaluation Network. Paris. Tanzania Electricity Supply Corporation, TANESCO. 2010. Annual Reports 2005-2010, Dar es Salaam. PovcalNet: the on-line tool for poverty measurement developed by the Development Research Group of the World Bank. (http://iresearch.worldbank.org/PovcalNet/index.htm?0) Shkaratan 2010. ―Africa Infrastructure Country Diagnostic Country Report. Tanzania‘s Infrastructure: A Continental Perspective.‖ Washington, DC: World Bank. Tanzania Ports Authority. 2009. ―Tanzania Ports Master Plan 2009.‖ Dar es Salaam: Tanzania Ports Authority. Tanzania Revenue Authority. 2009. ―Tanzania Time Release Study, 2009.‖ Dar es Salaam: Tanzania Revenue Authority. Transparency International 2011. Corruption Perceptions Index, 2007-2011. Berlin. United Nations Development Program. 2010. Human Development Report 2010. New York: United Nations. United Republic of Tanzania. 1998. ―The National Poverty Eradication Strategy.‖ Dar es Salaam: Tanzania Vice-President‘s Office. ——. 1999. Poverty and Welfare Monitoring Indicators, Dar es Salaam: Tanzania Vice-President‘s Office. ——. 2000. ―Poverty Reduction Strategy Paper.‖ Dar es Salaam. ——. 2001a. ―Poverty Monitoring Master Plan‖ Dar es Salaam. ——. 2001b. ―Poverty Reduction Strategy Paper – Progress Report 2000/01.‖ Dar es Salaam. ——. 2002. ―Agriculture Sector Development Programme Framework and Process Document.‖ Dar es Salaam. ——. 2003. ―Poverty and Human Development Report 2003.‖ Dar es Salaam: Tanzania Office of the Vice President. ——. 2004a. ―Summary Results from Service Delivery Surveys 2003/04.‖ Dar es Salaam: Tanzania President‘s Office-Public Service Management. ——. 2004b. ―Local Government Fiscal Review 2004.‖ Dar es Salaam: Tanzania Ministry of Finance. ——. 2005a. ―National Strategy for Growth and Reduction of Poverty.‖ (and MKUZA for Zanzibar). Dar es Salaam: Tanzania Vice President‘s Office. [http://www.povertymonitoring.go.tz/Mkukuta/MKUKUTA_MAIN_ENGLISH.pdf] ——. 2005b. ―Poverty and Human Development Report 2005.‖ Dar es Salaam: Tanzania Office of the Vice President. ——. 2006. ―Agriculture Sector Development Programme Document.‖ Dar es Salaam. ——. 2007, 2008, 2009, 2010, 2011. ―MKUKUTA Annual Implementation Report.‖ Dar es S alaam: Tanzania Ministry of Finance and Ministry of Planning, Economy and Empowerment. ——. 2008. ―The Roads Act, 2007.‖ Bill number 1 -2007-3. Dar es Salaam: Tanzania Parliament. ——. 2009. ―Household Budget Survey 2007 – Mainland Tanzania.‖ Dar es Salaam: Tanzania National Bureau of Statistics. 69 ——. 2010. ―National Strategy for Growth and Poverty Reduction II.‖ Dar es Salaam: Tanzania Ministry of Finance. ——. 2011a. ―Tanzania in Figures 2010.‖ Dar es Salaam: Tanzania National Bureau of Statistics. ——. 2011b. ―Poverty and Human Development Report 2011.‖ Dar es Salaam: Tanzania Ministry of Finance and Economic Affairs. Wobst, Peter. 2001. ―Structural Adjustment and Inter -sectoral Shifts in Tanzania: A Comparable General Equilibrium Analysis.‖ Washington, DC: IFPRI Research Report #117. World Bank. 2000. ―Memorandum of the President of International Development Association and International Finance Corporation to the Executive Directors on a Country Assistance Strategy for the United Republic of Tanzania.‖ Report No. 20426-TA. Washington, DC. ——. 2003. ―Program Document for a Proposed Credit in the Amount of US$32 Million to the United Republic of Tanzania for a Poverty Reduction Support Credit and Grant.‖ Report No. 25807 -TZ. Washington, DC. ——. 2004a. ―Program Document for a Proposed Credit in the Amount of US$60 Million to the United Republic of Tanzania for a Second Poverty Reduction Support Credit and Grant.‖ Report No.29616-TZ, Washington, DC. ——. 2004b. ―Investment Climate Assessment: Improving Enterprise Performance and Growth in Tanzania.‖ Washington, DC. ——. 2005. ―Program Document for a Proposed Credit in the Amount of US$150 Million to the United Republic of Tanzania for a Third Poverty Reduction Support Credit.‖ Report No.33166 -TZ. Washington, DC. ——. 2006a. ―Program Document for a Proposed Credit in the Amount of US$200 Million to the United Republic of Tanzania for a Fourth Poverty Reduction Support Credit (PRSC-4).‖ Report No. 35800-TZ. Washington, DC. ——. 2006b. ―United Republic of Tanzania Public Expenditure and Financial Accountability Review— FY05.‖ Report No. 36642-TZ. Washington, DC. ——. 2007a. ―Country Assistance Strategy for the United Republic of Tanzania: FY2007 -FY2010.‖ Report No. 38625-TZ. Washington, DC. ——. 2007b. ―Program Document for a Proposed Credit in the Amount of US$190 Million to the United Republic of Tanzania for a Fifth Poverty Reduction Support Credit (PRSC-5).‖ Report No. 38820- TZ. Washington, DC. ——. 2007c. ―Implementation Completion and Results Report on a Credit in the Amount of US$272 Million to the United Republic of Tanzania for a First Series of Poverty Reduction Support Credits (PRSCs 1-3).‖ Report No. ICR0000501. Washington, DC. ——. 2007d. ―Project Appraisal Document for a Proposed Credit in the Amount of US$105 to the United Republic of Tanzania for an Energy Development and Access Expansion Project.‖ Report No. 40771-TZ. Washington, DC. ——. 2007e. ―Tanzania – Sustaining and Sharing Economic Growth: Country Economic Memorandum and Poverty Assessment.‖ Report No. 39021-TZ. Washington, DC. ——. 2008a. ―Quality at Entry in FY06-06 (QEA8), A QAG Assessment.‖ Washington DC ——. 2008b. ―Program Document for a Proposed Credit in the Amount of US$160 Million to the United Republic of Tanzania for a Sixth Poverty Reduction Support Credit.‖ Report No. 45348 -TZ. Washington, DC. ——. 2009a. ―Program Document for a Proposed Credit in the Amount of US$190 Million to the United Republic of Tanzania for a Seventh Poverty Reduction Support Credit.‖ Report No. 48467 -TZ. Washington, DC. 70 ——. 2009b. ―Supplemental Financing Document for a Proposed Supplemental Financing in the Amount of US$170 Million to the United Republic of Tanzania for a Tanzania Seventh Poverty Reduction Support Credit.‖ Report No.51590-TZ. Washington, DC. ——. 2010a. ―International Development Association and the International Finance Corporation – Country Assistance Strategy Progress Report for the United Republic of Tanzania.‖ Report No. 52922 -TZ. Washington, DC. ——. 2010c. ―Program Document for a Proposed Credit in the Amount of US$115 Million Including US$15 Million in Pilot Crisis Response Window Resources to the United Republic of Tanzania for a Eighth Poverty Reduction Support Credit.‖ Report No. 56347 -TZ. Washington, DC. ——. 2011a. ―Implementation Completion and Results Report on a Series of Credits Totaling in the Amount of US$1,025 Million to the Untied Republic of Tanzania for a Second Series of Poverty Reduction Support Credits (PRSCs 4 through 8). Report No. ICR00001717. Washington, DC. ——. 2011b. ―International Development Association, International Finance Corporation, and Multilateral Investment Guarantee Agency Country Assistance Strategy for the United Republic of Tanzania for the Period FY 2012-2015.‖ Report No. 60269-TZ. Washington, DC. ——. 2012a. ―Tanzania Economic Update: Stairways to Heaven.‖ Report No. 73339. Washington, DC. ——. 2012b. ―2012 Development Policy Lending Retrospective: Results, Risks, and Reforms.‖ Washington, DC. World Bank, International Finance Corporation, and U.K. Department for International Development. 2010. Doing Business in the East African Community. Washington, DC: World Bank. World Development Indicators. 2012. Washington, DC: World Bank. Worldwide Governance Indicators. 2012. Washington, DC: World Bank. 71 ANNEX A Annex A. Basic Data Sheet TANZANIA 1ST POVERTY REDUCTION SUPPORT CREDIT (CR. 37721, GR.H0400) Key Project Data Appraisal Actual or Actual as % of estimate current estimate appraisal estimate Total project costs 132.00 132.00 100 Loan amount 132.00 132.00 100 Cofinancing -- -- -- Cancellation Cumulative Estimated and Actual Disbursements FY04 Appraisal estimate (US$M) 132.00 Actual (US$M) 132.00 Actual as % of appraisal 100 Date of final disbursement: 06/30/2004 Project Dates Original Actual Concept Review 05/08/2002 05/08/2002 Appraisal 02/26/2003 02/13/2003 Board approval 05/29/2003 05/29/2003 Signing 06/23/2003 06/23/2003 Effectiveness 07/18/2003 07/18/2003 Closing date 06/30/2004 03/31/2006 Staff Time and Cost Staff Time and Cost (Bank Budget Only) Stage of Project Cycle No. of staff weeks US$ Thousands (including travel and consultant costs) Lending FY02 29 145.57 FY03 85 345.41 FY04 2 1.55 Total 116 492.53 Supervision/ICR FY02 FY03 FY04 4 17.63 Total 4 17.63 73 ANNEX A Other Project Data Borrower/Executing Agency: Follow-on Operations Operation Amount Credit no. Board date (US$ million) 2nd Poverty Reduction Support Credit 3965-TA 150.00 07/29/2004 TANZANIA 2ND POVERTY REDUCTION SUPPORT CREDIT (LOAN 3965-TA & GRANT H118- TA) Key Project Data Appraisal Actual or Actual as % of estimate current estimate appraisal estimate Total project costs 150.00 150.54 100.36 Loan amount 150.00 150.54 100.36 Cofinancing Cancellation Cumulative Estimated and Actual Disbursements FY05 Appraisal estimate (US$M) 150.00 Actual (US$M) 150.54 Actual as % of appraisal 100.36 Date of final disbursement: 09/15/2004 Project Dates Original Actual Concept Review 11/18/2003 11/18/2003 Appraisal 03/29/2004 06/28/2004 Board approval 07/29/2004 07/29/2004 Signing 08/05/2004 08/05/2004 Effectiveness 09/15/2004 09/15/2004 Closing date 06/30/2005 06/30/2005 Task Team members Names Responsibility/Specialty Lending Supervision Denis Maro Biseko Public Sector Specialist AFTPR Serigne Omar Fye Sr Environmental Specialist. AFTS1 Indumathie V. Hewawasam Sr Environmental Specialist. AFTS2 Manush A. Hristov Counsel LEGAF Rogati Anael Kayani Consultant SDNCA Marius Koen Sr Financial Management Specialist OPCFM Allister J. Moon Lead Economist AFTP2 74 ANNEX A Denyse E. Morin Sr Public Sector Specialist. AFTPR Philip Isdor N. Mpango Senior Economist AFTP2 Emmanuel A. Mungunasi Research Analyst AFTP2 Mary-Anne D. Mwakangale Program Assistant AFCE1 Vedasto Rwechungura Program Officer AFTPS Mercy Mataro Sabai Sr Financial Management Specialist AFTFM Arlette Sourou Program Assistant AFTP2 Pascal Tegwa Sr Procurement Specialist. AFTPC Robert Townsend Senior Economist SASAR Michael D. Wong Sr Private Sector Development AFTPS Staff Time and Cost Staff Time and Cost (Bank Budget Only) Stage of Project Cycle No. of staff weeks US$ Thousands (including travel and consultant costs) Lending& Supervision FY03 2 11.13 FY04 75 377.96 FY05 9 43.71 Total 86 432.80 *Supervision was carried out concurrently with preparation of next operation in sequence Other Project Data Borrower/Executing Agency: Follow-on Operations Amount Operation Credit no. Board date (US$ million) 3rd Poverty Reduction Support Credit Loan 4110-TA 150.00 09/08/2005 TANZANIA 3RD POVERTY REDUCTION SUPPORT CREDIT (LOAN 4110-TA) Key Project Data Appraisal Actual or Actual as % of estimate current estimate appraisal estimate Total project costs 150.00 150.00 100 Loan amount 150.00 150.00 100 Cofinancing Cancellation Cumulative Estimated and Actual Disbursements FY06 Appraisal estimate (US$M) 150.00 Actual (US$M) 150.00 Actual as % of appraisal 100 Date of final disbursement: 06/30/2006 75 ANNEX A Project Dates Original Actual Concept Review 02/01/2005 02/01/2005 Appraisal 03/31/2005 03/31/2005 Board approval 09/08/2005 09/08/2005 Signing 09/22/2005 09/22/2005 Effectiveness 10/20/2005 10/20/2005 Closing date 06/30/2006 06/30/2006 Task Team members Names Responsibility/Specialty Lending Serigne Omar Fye Sr Environmental Special. AFTS1 Indumathie V. Hewawasam Sr Environmental Special. AFTS2 Manush A. Hristov Counsel LEGAF Rogati Anael Kayani Consultant SDNCA Marius Koen Sr Financial Management Special OPCFM Allister J. Moon Lead Economist AFTP2 Philip Isdor N. Mpango Senior Economist AFTP2 Emmanuel A. Mungunasi Research Analyst AFTP2 Mary-Anne D. Mwakangale Program Assistant AFCE1 Vedasto Rwechungura Program Officer AFTPS Alema E. Siddiky Consultant AFCF2 Arlette Sourou Program Assistant AFTP2 Robert Townsend Senior Economist SASAR Michael D. Wong Sr Private Sector Development AFTPS Supervision Denis Maro Biseko Public Sector Specialist AFTPR Serigne Omar Fye Sr Environmental Special. AFTS1 Indumathie V. Hewawasam Sr Environmental Special. AFTS2 Johannes G. Hoogeveen Senior Economist AFTP2 Manush A. Hristov Counsel LEGAF Allister J. Moon Lead Economist AFTP2 Denyse E. Morin Sr Public Sector Special. AFTPR Philip Isdor N. Mpango Senior Economist AFTP2 Emmanuel A. Mungunasi Research Analyst AFTP2 Mary-Anne D. Mwakangale Program Assistant AFCE1 Vedasto Rwechungura Program Officer AFTPS Dieter E. Schelling Lead Transport Specialist AFTTR Arlette Sourou Program Assistant AFTP2 Pascal Tegwa Sr Procurement Special. AFTPC Robert Townsend Senior Economist SASAR 76 ANNEX A Michael D. Wong Sr Private Sector Development AFTPS Staff Time and Cost Staff Time and Cost (Bank Budget Only) Stage of Project Cycle US$ Thousands (including No. of staff weeks travel and consultant costs) Lending & Supervision FY04 0.80 FY05 90 440.26 FY06 14 48.83 Total 104 489.89 Other Project Data Borrower/Executing Agency: Follow-on Operations Operation Amount Credit no. Board date (US$ million) 4th Poverty Reduction Support Credit 4170-TA & TF-90200 200.00 05/09/2006 TANZANIA 4TH POVERTY REDUCTION SUPPORT CREDIT (LOAN4170-TA) Key Project Data Appraisal Actual or Actual as % of estimate current estimate appraisal estimate Total project costs 200.00 200.00 100 Loan amount 200.00 200.00 100 Cofinancing Cancellation Cumulative Estimated and Actual Disbursements FY07 Appraisal estimate (US$M) 200.00 Actual (US$M) 200.00 Actual as % of appraisal 100 Date of final disbursement: 06/30/2007 Project Dates Original Actual Concept Review 12/12/2005 12/12/2005 Appraisal 03/14/2006 03/14/2006 Board approval 05/09/2006 05/09/2006 Signing 06/05/2006 06/05/2006 Effectiveness 09/03/2006 07/18/2006 Closing date 06/30/2007 06/30/2007 77 ANNEX A Task Team members Names Responsibility/Specialty Lending Denis Maro Biseko Sr. Public Sector Spec. AFTPR Parminder P. S. Brar Lead Financial Management Spec AFTFM Francis Ato Brown Sector Manager, Water MNSSD Rogati Anael Kayani Consultant AFTPC Rest Barnabas Lasway Sr. Education Spec. PA9SS Julie McLaughlin Sector Manager, Health, Nutrition SASHN Allister J. Moon Lead Economist WBIGV Denyse E. Morin Senior Operations Officer OPCIL Philip Isdor N. Mpango Senior Economist AFTP2 Emmanuel A. Mungunasi Economist AFTP2 Robert Townsend Senior Economist ARD Michael D. Wong Sr. Private Sector Development SASFP Robert Utz Senior Economist AFTP2 Supervision Modupe A. Adebowale Consultant AFTFM Parminder P. S. Brar Lead Financial Management Spec AFTFM Francis Ato Brown Sector Manager, Water MNSSD Serigne Omar Fye Consultant AFTSP Henry F. Gordon Sr. Sector Economist AFTAR Rogati Anael Kayani Consultant AFTPC Rest Barnabas Lasway Sr. Education Spec. PA9SS Baruany Elijah A. T. Luhanga Consultant AFTUW Julie McLaughlin Sector Manager, Health, Nutrition SASHN Allister J. Moon Lead Economist WBIGV Philip Isdor N. Mpango Senior Economist AFTP2 Emmanuel A. Mungunasi Economist AFTP2 Mary-Anne D. Mwakangale Program Assistant AFCE1 Edith Ruguru Mwenda Sr. Counsel LEGAF Vedasto Rwechungura Consultant AFTDP Dieter E. Schelling Lead Transport Specialist AFTTR Arlette Sourou Program Assistant AFTP2 Robert Townsend Senior Economist ARD Michael D. Wong Sr .Private Sector Development SASFP Robert Utz Senior Economist AFTP2 78 ANNEX A Staff Time and Cost Staff Time and Cost (Bank Budget Only) Stage of Project Cycle US$ Thousands (including No. of staff weeks travel and consultant costs) Lending and supervision FY06 33.5 149.58 FY07 0.2 0.94 33.7 150.52 Total Other Project Data Borrower/Executing Agency: Follow-on Operations Operation Credit no. Amount Board date (US$ million) 5th Poverty Reduction Support Credit 4294-TA 190.00 04/24/2007 TANZANIA 5TH POVERTY REDUCTION SUPPORT CREDIT (LOAN4294-TA) Key Project Data Appraisal Actual or Actual as % of estimate current estimate appraisal estimate Total project costs 190.00 190.00 100 Loan amount 190.00 190.00 100 Cofinancing Cancellation Cumulative Estimated and Actual Disbursements FY08 Appraisal estimate (US$M) 190.00 Actual (US$M) 190.00 Actual as % of appraisal 100 Date of final disbursement: 06/30/2008 Project Dates Original Actual Concept Review 10/12/2006 10/12/2006 Appraisal 01/10/ 2007 02/06/2007 Board approval 04/24/2007 04/24/2007 Signing 06/12/2007 06/12/2007 Effectiveness 08/27/2007 08/27/2007 Closing date 06/30/2008 06/30/2008 79 ANNEX A Task Team members Names Responsibility/Specialty Lending Denis Maro Biseko Sr Public Sector Spec. AFTPR Parminder P. S. Brar Lead Financial Management Spec AFTFM Francis Ato Brown Sector Manager, Water MNSSD Henry F. Gordon Sr Sector Economist AFTAR Johannes G. Hoogeveen Senior Economist AFTP2 Ralph Ake Karhammar Sr Energy Spec. AFTEG Rest Barnabas Lasway Sr Education Spec. PA9SS Baruany Elijah A. T. Luhanga Consultant AFTUW Julie McLaughlin Sector Manager, Health, Nutrit SASHN Allister J. Moon Lead Economist WBIGV Denyse E. Morin Senior Operations Officer OPCIL Philip Isdor N. Mpango Senior Economist AFTP2 Emmanuel A. Mungunasi Economist AFTP2 Christian Albert Peter Sr. Natural Resources Mgmt. Spec. AFTEN Vedasto Rwechungura Consultant AFTDP Mercy Mataro Sabai Sr Financial Management Spec. AFTFM Dieter E. Schelling Lead Transport Specialist AFTTR Pascal Tegwa Senior Procurement Specialist AFTPC Solomon Muhuthu Waithaka Sr Highway Engineer AFTTR Michael D. Wong Sr Private Sector Development SASFP Robert Utz Senior Economist AFTP2 Supervision Modupe A. Adebowale Consultant AFTFM Gregor Binkert Lead Specialist AFTP1 Denis Maro Biseko Sr Public Sector Spec. AFTPR Parminder P. S. Brar Lead Financial Management Spec AFTFM Francis Ato Brown Sector Manager, Water MNSSD Matthew D. Glasser Lead Urban Specialist SASDU Henry F. Gordon Sr Sector Economist AFTAR Pankaj Gupta Lead Financial Specialist AFTEG Johannes G. Hoogeveen Senior Economist AFTP2 Ralph Ake Karhammar Sr Energy Spec. AFTEG Rest Barnabas Lasway Sr Education Spec. PA9SS Baruany Elijah A. T. Luhanga Consultant AFTUW Julie McLaughlin Sector Manager, Health, Nutrit SASHN Donald Paul Mneney Senior Procurement Specialist AFTPC Allister J. Moon Lead Economist WBIGV Denyse E. Morin Senior Operations Officer OPCIL 80 ANNEX A Philip Isdor N. Mpango Senior Economist AFTP2 Emmanuel A. Mungunasi Economist AFTP2 Mary-Anne D. Mwakangale Program Assistant AFCE1 Edith Ruguru Mwenda Sr Counsel LEGAF Christian Albert Peter Sr Natural Resources Mgmt. Spe AFTEN Ravi Ruparel Sr Financial Sector Spec. AFTFE Vedasto Rwechungura Consultant AFTDP Mercy Mataro Sabai Sr Financial Management Spec. AFTFM Dieter E. Schelling Lead Transport Specialist AFTTR Arlette Sourou Program Assistant AFTP2 Pascal Tegwa Senior Procurement Specialist AFTPC Robert Johann Utz Senior Economist AFTP3 Linda Van Gelder Sector Manager EASPR Michael D. Wong Sr. Private Sector Development SASFP Staff Time and Cost Staff Time and Cost (Bank Budget Only) Stage of Project Cycle US$ Thousands (including No. of staff weeks travel and consultant costs) Lending and supervision FY07 69 282.53 FY08 4 17.97 Total 73 300.50 Other Project Data Borrower/Executing Agency: Follow-on Operations Operation Credit no. Amount Board date (US$ million) 6th Poverty Reduction Support Credit 4526-TA 160.00 10/21/2008 TANZANIA 6TH POVERTY REDUCTION SUPPORT CREDIT (LOAN4526-TA) Key Project Data Appraisal Actual or Actual as % of estimate current estimate appraisal estimate Total project costs 160.00 160.00 100 Loan amount 160.00 160.00 100 Cofinancing Cancellation 81 ANNEX A Cumulative Estimated and Actual Disbursements FY09 Appraisal estimate (US$M) 160.00 Actual (US$M) 160.00 Actual as % of appraisal 100 Date of final disbursement: 11/30/ 2009 *The data were taken from Client connection. Project Dates Original Actual Concept Review 10/25/2007 10/25/2007 Appraisal 03/25/2008 03/25/2008 Board approval 10/21/2008 10/21/2008 Signing 11/04/2008 11/04/2008 Effectiveness 11/17/2008 11/17/2008 Closing date 11/30/2009 11/30/2009 Task Team members Names Responsibility/Specialty Lending and Supervision Denis Maro Biseko Sr Public Sector Spec. AFTPR Francis Ato Brown Sector Manager, Water MNSSD Matthew D. Glasser Lead Urban Specialist SASDU Henry F. Gordon Sr Sector Economist AFTAR Johannes G. Hoogeveen Senior Economist AFTP2 Ralph Ake Karhammar Sr Energy Spec. AFTEG Josaphat Paul Kweka Senior Economist AFTP2 Rest Barnabas Lasway Sr Education Spec. PA9SS Baruany Elijah A. T. Luhanga Consultant AFTUW Julie McLaughlin Sector Manager, Health, Nutrit SASHN Denyse E. Morin Senior Operations Officer OPCIL Jacques Morisset Lead Economist AFTP1 Emmanuel A. Mungunasi Economist AFTP2 Mary-Anne D. Mwakangale Program Assistant AFCE1 Edith Ruguru Mwenda Sr Counsel LEGAF Christian Albert Peter Sr Natural Resources Mgmt. Spe AFTEN Ravi Ruparel Sr Financial Sector Spec. AFTFE Mercy Mataro Sabai Sr Financial Management Spec. AFTFM Dieter E. Schelling Lead Transport Specialist AFTTR Luis M. Schwarz Senior Finance Officer CTRFC Vera Songwe Adviser MDO Arlette Sourou Program Assistant AFTP2 82 ANNEX A Javier Suarez Cordero Senior Economist AFTFE Michael D. Wong Sr Private Sector Development SASFP Paolo B. Zacchia Lead Economist AFTP2 Staff Time and Cost Staff Time and Cost (Bank Budget Only) Stage of Project Cycle US$ Thousands (including No. of staff weeks travel and consultant costs) Lending and supervision FY08 51 193.21 FY09 14 72.54 Total 65 265.75 Other Project Data Borrower/Executing Agency: Follow-on Operations Operation Credit no. Amount Board date (US$ million) 7th Poverty Reduction Support Credit 4635-TA 190.00 06/09/2009 TANZANIA 7TH POVERTY REDUCTION SUPPORT CREDIT (LOAN4635-TA) Key Project Data Appraisal Actual or Actual as % of estimate current estimate appraisal estimate Total project costs 190.00 190.00 100 Loan amount 190.00 190.00 100 Cofinancing Cancellation Cumulative Estimated and Actual Disbursements FY10 Appraisal estimate (US$M) 190.00 Actual (US$M) 190.00 Actual as % of appraisal 100 Date of final disbursement: 11/30/2010 Project Dates Original Actual Concept Review 02/03/2009 02/03/2009 Appraisal 04/20/2009 04/20/2009 Board approval 06/09/2009 06/09/2009 Signing 01/29/2010 01/29/2010 Effectiveness 08/20/2009 08/20/2009 Closing date 11/30/2010 11/30/2010 83 ANNEX A Task Team members Names Responsibility/Specialty Lending and Supervision Sherri Ellen Archondo Senior Operations Officer AFTFE Denis Maro Biseko Sr Public Sector Spec. AFTPR Parminder P. S. Brar Lead Financial Management Spec AFTFM Francis Ato Brown Sector Manager, Water MNSSD Matthew D. Glasser Lead Urban Specialist SASDU Pankaj Gupta Lead Financial Specialist AFTEG Dominic S. Haazen Lead Health Policy Specialist AFTHE Johannes G. Hoogeveen Senior Economist AFTP2 Arun R. Joshi Sr Education Spec. AFTED Ralph Ake Karhammar Sr Energy Spec. AFTEG Rosemary Mukami Kariuki Sector Leader AFTUW Jane A. N. Kibbassa Sr Environmental Spec. AFTEN Josaphat Paul Kweka Senior Economist AFTP2 Rest Barnabas Lasway Sr Education Spec. PA9SS Baruany Elijah A. T. Luhanga Consultant AFTUW Ida Manjolo WBIVP Donald Paul Mneney Senior Procurement Specialist AFTPC Denyse E. Morin Senior Operations Officer OPCIL Emmanuel A. Mungunasi Economist AFTP2 Mary-Anne D. Mwakangale Program Assistant AFCE1 Edith Ruguru Mwenda Sr Counsel LEGAF Adam Nelsson Country Officer AFCE1 Christian Albert Peter Sr Natural Resources Mgmt. Spec. AFTEN Ravi Ruparel Sr Financial Sector Spec. AFTFE Mercy Mataro Sabai Sr Financial Management Spec. AFTFM Dieter E. Schelling Lead Transport Specialist AFTTR Robert Schlotterer Financial Analyst AFTEG Arlette Sourou Program Assistant AFTP2 Javier Suarez Cordero Senior Economist AFTFE Satoru Ueda Lead Water Resources Spec. AFTWR Yutaka Yoshino Economist AFTP2 Paolo B. Zacchia Lead Economist AFTP2 84 ANNEX A Staff Time and Cost Staff Time and Cost (Bank Budget Only) Stage of Project Cycle US$ Thousands (including No. of staff weeks travel and consultant costs) Lending FY09 39.72 202,024.76 Supervision (including PRSC 7 Supplemental financing) FY09 0.00 31,770.80 FY10 9.52 72,544.15 Total 49.24 306,339.71 Other Project Data Borrower/Executing Agency: Follow-on Operations Operation Credit no. Amount Board date (US$ million) 7th Poverty Reduction Support Credit- Supplemental 4635-1-TA 170.00 12/22/2009 TANZANIA 7TH POVERTY REDUCTION SUPPORT CREDIT SUPPLEMENTAL FINANCING (LOAN4635-1-TA) Key Project Data Appraisal Actual or Actual as % of estimate current estimate appraisal estimate Total project costs 170.00 170.00 100 Loan amount 170.00 170.00 100 Cofinancing Cancellation Cumulative Estimated and Actual Disbursements FY10 Appraisal estimate (US$M) 170.00 Actual (US$M) 170.00 Actual as % of appraisal 100 Date of final disbursement:11/30/2010 Project Dates Original Actual Concept Review 08/04/2009 08/04/2009 Appraisal 09/28/2009 09/28/2009 Board approval 12/22/2009 12/22/2009 Signing 01/29/2010 01/29/2010 Effectiveness 02/23/2010 02/23/2010 Closing date 11/30/2010 11/30/2010 85 ANNEX A Task Team members Names Responsibility/Specialty Lending and Supervision Javier Suarez Cordero Senior Economist AFTP2 Emmanuel A. Mungunasi Economist AFTP2 Jos Verbeek Lead Economist AFTP2 Sergiy Zorya Senior Economist AFTAR Yutaka Yoshino Economist AFTP2 Staff Time and Cost Staff Time and Cost (Bank Budget Only) Stage of Project Cycle US$ Thousands (including No. of staff weeks travel and consultant costs) Lending FY10 22.1 94,551.56 Total FY 09-10 22.1 94,551.56 Other Project Data Borrower/Executing Agency: Follow-on Operations Operation Credit no. Amount Board date (US$ million) 8th Poverty Reduction Support Credit 4813-TZ 115.00 03/09/2010 TANZANIA 8TH POVERTY REDUCTION SUPPORT CREDIT (LOAN 4813-TZ) Key Project Data Appraisal Actual or Actual as % of estimate current estimate appraisal estimate Total project costs 115.00 115.00 100 Loan amount 115.00 115.00 100 Cofinancing Cancellation Cumulative Estimated and Actual Disbursements FY11 Appraisal estimate (US$M) 115.00 Actual (US$M) 115.00 Actual as % of appraisal 100 Date of final disbursement: 08/31/2012 86 ANNEX A Project Dates Original Actual Concept Review 10/20/2009 10/20/2009 Appraisal 03/09/2010 03/09/2010 Board approval 09/28/2010 09/28/2010 Signing 10/08/2010 10/08/2010 Effectiveness 11/19/2010 11/19/2010 Closing date 08/31/2012 08/31/2012 Task Team members Names Responsibility/Specialty Lending Sherri Ellen Archondo Senior Operations Officer AFTFE Denis Maro Biseko Sr Public Sector Spec. AFTPR Madhur Gautam Lead Economist SASDA Dominic S. Haazen Lead Health Policy Spec. AFTHE Arun R. Joshi Sr Education Spec. AFTED Regina Astrid Martinez Fernandez E T Consultant AFTFE Yonas Eliesikia Mchomvu Transport Specialist AFTTR Emmanuel A. Mungunasi Economist AFTP2 Mary-Anne D. Mwakangale Program Assistant AFCE1 Michel Noel Lead Financial Specialist AFTFE Christian Albert Peter Sr Natural Resources Mgmt. Spec. AFTEN Dieter E. Schelling Lead Transport Specialist AFTTR Arlette Sourou Program Assistant AFTP2 Yutaka Yoshino Economist AFTP2 Paolo B. Zacchia Lead Economist AFTP2 Sergiy Zorya Economist AFTAR Supervision Yutaka Yoshino Economist AFTP2 Paolo B. Zacchia Lead Economist AFTP2 Yue Lie Young Professional AFTP2 Peter Miovic Consultant AFTP2 87 ANNEX A Staff Time and Cost Staff Time and Cost (Bank Budget Only) Stage of Project Cycle US$ Thousands (including No. of staff weeks travel and consultant costs) Lending FY10 32.38 222,330.90 FY11 9.55 65,460.77 Supervision including ICR preparation FY11 17.03 98,691.53 Total FY 10-11 58.96 386,483.20 Other Project Data Borrower/Executing Agency: Follow-on Operations Operation Credit no. Amount Board date (US$ million) 9th Poverty Reduction Support Credit 5072-TA 100.00 03/15/2012 88 ANNEX B Annex B. Complementarity of PRSCs with Other Bank Support 1. The PRSCs were implemented in parallel with a number of analytical products and investment projects financed by the Bank that were directly related to the PRSC objectives. This annex identifies those relevant activities, and assesses the extent to which the PRSCs and other Bank support were complementary. 2. Public Sector Reform. Analytical work was central to the Bank‘s public sector reform dialogue in Tanzania. During the first PRSC series and prior, annual Public Expenditure Reviews were produced through a highly participatory process, with meetings between the Bank and government counterparts almost every two weeks. The major objectives was to be sure that actual expenditures matched closely with what was in the approved budget and to have expenditures focused on poverty reduction, outcomes that were central to the early PRSCs as well. The PRSCs drew heavily on the reports as well as other analytical work during the first series, and provided the main focal point for operational dialogue around public sector reform issues. In this regard, the PRSCs and analytical work were highly complementary. 3. There were six relevant operations aimed at improving the performance of the public sector (highlighted in Table B.1). Table B.1. Bank Investment Lending for Public Sector Reform. Fiscal Project Name Objective Commitment Year 1999 Public Service Improve accountability, transparency, and resource $41.2 million Reform Program management for service delivery. 2005 Tax Administration Increase tax revenues without increasing tax rates. $12 million Reform 2005 Local Government Strengthen fiscal decentralization, improve $52 million Support Project accountability in the use of government resources, improve management of intergovernmental transfer systems, increase access to infrastructure and services in unplanned areas of Dar es Salaam, and improve revenue performance. 2006 Accountability, Improve access to judicial and legal services and in $40 million Transparency, and the accountable and transparent use of public Integrity Project financial resources. 2008 Performance, Results Enhance capacity, performance and accountability $40 million and Accountability of Ministries, Departments, and Agencies in the use Project of public resources and service delivery to levels consistent with timely and effective implementation of MKUKUTA programs. 4. The PRSCs may be considered as successor operations to the Public Financial Management Reform Program and Public Service Reform Program, which had previously been the primary focus of dialogue on these issues. Subsequent investment lending, topic- specific or agency-specific, was generally more important for the associated government agencies that the broad PRSC agenda. They tended to fill particular niches in the overall 89 ANNEX B program (on tax reform, local government, and judicial access, for example). But they were not central to the PSRCs, nor were government counterparts always invited in participate in the PRSC process. 5. Agriculture and rural development. Two pieces of analytical work—the 2000 "Agriculture in Tanzania since1986: Follower or Leader of Growth?‖ produced jointly with the International Food Policy Research Institute and a Bank-led 2005 Diagnostic Trade Integration Study—covered some of the same issues that later appeared in the PRSCs. These emphasized Tanzania‘s comparative advantage in maize, paddy and the traditional export crops, the potential of livestock production, the promise of agricultural research and extension, the need to reduce transport costs and rationalize and reduce agricultural export taxes, and the importance of crop board reform. While the there are some evident linkages between this analytical work and the focus of the PRSCs, issues of crop board reform and land were long-standing policy issues in the Tanzania agricultural sector with antecedents well beyond the time period of the PRSCs. 6. There were three agriculture lending operations approved during implementation of the PRSCs, summarized in Table B.2. Table B.2. Bank Investment Lending for Agriculture and Rural Development. Fiscal Project Name Objective Commitment Year 2003 Participatory Raise the production of food, incomes, and assets in $57 million Agricultural about 840 villages through the implementation of Development and small agricultural development subprojects planned Empowerment and managed by community members. Project 2006 Agricultural Sector Enable farmers to have better access to agricultural $90 million Development knowledge, technologies, marketing systems and Project infrastructure as a means to higher productivity, profitability, and farm incomes; and promote agricultural private investment based on an improved regulatory and policy environment. 2009 Accelerated Food Contribute to higher food production and $160 million Security Project productivity in targeted areas by improving farmers‘ access to critical agricultural inputs. 7. The lending operations supported by the Bank during the PRSC period appear to reflect contemporaneous priorities and approaches to agriculture in Tanzania. While successful implementation of the policy actions supported by the PRSCs would conceivably have improved the overall policy environment for agriculture and therefore enhance the likelihood that the three projects would be successful, the complementarity is quite broad and unspecific. The PRSC and investment lending in agriculture appear to have proceeded along two relatively independent tracks. 8. Private Sector Development. The private sector development agenda for the first series of PRSCs was shaped by sectoral analytical work, including the 2003 Investment Climate Assessment. But the linkage weakened in the second series. The findings of the 2006 90 ANNEX B Investment Climate Assessment were released only in 2009, a missed opportunity in terms of influencing the policy dialogue. 9. The PRSC focused on creating the general framework for a growing private sector, including the basic legislation and the national strategies that would facilitate effective interventions at operational level, including sector lending. At face value this took place, and the private sector development reforms under the PRSC program were followed by an investment loan—the USD95 million Private Sector Competitiveness project launched in 2006. While the project complemented the PRSCs at the conceptual level—supporting Tanzania‘s business environment strengthening program, building up private sector capacity to respond to regional and international opportunities, and improving access to financial services—implementation shortfalls in both resulted in missed opportunities. 10. Relatively poor outcomes in the competitiveness project reflected shortcomings in the legislation approved under the PRSC (such as the failure to change other acts to make the new legislation operational, and a delay in the adoption of subsidiary acts) as well as lack of political commitment and consensus about private sector development reforms. In addition, slow progress under the investment lending project reduced the space for potential intervention at policy level under PRSCs. 11. Health, Education, and Water. The second PRSC series was intended to provide additional leverage to the health, education, and water sectors dialogue among the Government, the Bank, and other development partners without adding additional layers of triggers, indicators, or review. The PRSC did not suggest any additional policy actions, but rather focused ―strictly on the monitoring and evaluation of effective service delivery‖ and addressing cross-cutting issues. PRSC prior actions were simply a rating of ―satisfactory‖ for annual reviews in each sector, defined as a mix of process, measurement, and outcome targets. 12. This approach reflected an assumption that development partners would adopt a sharp division between general budget support financing (including the PRSCs) and sector program support and policy dialogue. Sector financing would be absorbed would eventually be absorbed by general budget support, while policy support and dialogue would continue among sector working groups. In a sense this reflected the deepest potential complementarity between the PRSCs and sector investment and policy reform. Yet in practice, the ideal fell short as the coordinating role of the Ministry of Finance waned, and the line ministries in each of the sectors feared a shortfall in resources should their external financing mechanism be shifted to general budget support. As the ministries refused to recognize any link between general budget support and sector engagement with development partners, the Bank and other partners retained their support of sector arrangements. The Bank approved broad sector financing in all three sectors: additional financing totaling USD100 million for its Health Sector Development Program in 2008 and 2010; a USD150 million Secondary Education Development Program in 2010; and a USD200 million Water Sector Support Project in 2007. 13. Energy. The key issues in energy during the period were (i) frequent electricity shortages due to limited generation capacity and over-reliance on hydro-power; (ii) very low 91 ANNEX B accessibility to electricity; (iii) poor quality of electricity supply; and (iv) unsustainable financial position and poor management of the government-owned utility, TANESCO. 14. Three investment projects in the sector sought to address, to varying degrees, the first three issues, but the challenges persist. The Songo Songo Gas Development and Power Generation project, approved in 2002 and closed in 2011, used private sector management and expertise to develop and operate a gas infrastructure system and a commercial market for gas. After a severe drought in 2003, the Emergency Power Supply project was a short-term intervention to help avert a prolonged period of severe load-shedding by acquiring emergency electric energy from a private independent power producer and purchasing fuel for TANESCO. The Tanzania Energy Development and Access Expansion project, approved in FY2008 and still active, financed investments in TANESCO‘s transmission and distribution networks and supported rural off-grid alternatives. 15. The PRSC‘s focus was squarely on improving the management and financial position of TANESCO, while encouraging an environment for private investment in the sector. This was seen as the major obstacle to each of the other key issues—generating capacity, access expansion, and quality of supply—and was not addressed through any of the other projects. Moreover, the IMF considered the large transfers from the general budget to TANESCO as a major obstacle to fiscal sustainability in the country, and so cost-reflective tariffs and financial recovery for the utility could contribute to a stable macroeconomic environment in addition to private sector development. 16. In this sense, the design of the PRSC was a strong complement to the Bank‘s other investments and policy goals. But in practice, the PRSC had trouble overcoming the political economy barriers to improving TANESCO‘s financial position dramatically. 17. Transport. The key issues in transport during the period were the low quality of the trunk and regional road network, overcrowding of the Dar es Salaam port, weak functioning of the railways, and limited access of rural areas to markets due to a lack of all-weather rural roads. It was thought that in addition to public investments, private sector resources could be leveraged to address these shortfalls in the transportation sector. 18. The PRSC focused on policy and institutional improvements, including on roads, ports, and public-private partnerships (PPPs). One important trigger was the completion of a Transport Sector Investment Program with adequate allowance for road maintenance. The Investment Program was completed and addressed maintenance to an extent, although it turned out that there was a one-year surge in maintenance funding that receded rapidly thereafter. The government anticipated a number of sectors would benefit from private sector investment to help fill gaps in infrastructure development and maintenance, including in ports, roads, and energy. But the country lacked a national policy on PPPs, and dialogue or investment in a single sector was insufficient to remedy the situation. The PRSC helped to push through an imperfect national PPP policy, a necessary (but insufficient) step to invite this type of investment. 19. The Bank had three open investment projects during implementation of the two PRSC series, as summarized in Table B.3, focused on financing specific transport investments. The 92 ANNEX B Central Transport Corridor Project and its successor financed upgrading strategic road links in the central corridor, a bus rapid transit system in Dar es Salaam, and repair of a runway at the airport on Zanzibar. The Transport Sector Support Project and its additional financing financed a three-year slice of the Investment Program that the PRSC helped to jumpstart and improve. This included including rehabilitation of paved trunk roads, improvement of regional airports, improvement of road safety, promotion of public-private partnerships, emergency road and bridge repair, improvement of the Zanzibar airport, and rehabilitation of the jetty at Songo Songo island. But no project addressed the shortcomings of the railways or has yet picked up on the national PPP policy. Table B.3. Bank Investment Lending for Transport Projects FY Project name Objectives Commitment Upgrade strategic road links, enhance road Central Transport Corridor 2004 management capacity and improve operations of $122 million Project Tanzanian Railways. Support Tanzania‘s economic growth by providing Second Central Transport 2008 enhanced transport facilities that are reliable and cost $190 million Corridor Project effective. Improve the condition of the national paved road Transport Sector Support 2010 network, lower transport cost on selected roads, and $270 million Project (TSSP) expand the capacity of selected regional airports. To improve the condition of the national paved road network, lower transport cost on selected roads and to 2011 TSSP Additional Financing $59 million Songo Songo Island, and expand the capacity of selected airports. 93 ANNEX C Annex C. PRSCs 1-3: Detailed Assessment of Progress by Objective. Objective 1: Sustaining and accelerating economic growth and broadening its impact. Table C.1. Indicators of achievement on intermediate outcomes on shared growth objective Indicator Baseline Target Final Macroeconomic Debt service to exports ratio 5.3% (2002) NA 4.3% (2005) stability Improved business Growth rate of industry sector 7.8% 8-10% 10.3% environment Growth rate of agriculture 4.6% 5% 5% Raised income in sector rural areas Rural poverty headcount 38.6% (2001) 34% (2010) 37.4% (2007) (roughly USD 0.32/day) 1. The Bank recognized that achieving the overall growth and poverty reduction objectives of the PRSCs required the maintenance of macroeconomic stability. The IMF took the lead on these issues under its Poverty Reduction Grant Facility (PRGF), with PRGF reviews serving as an input into the Bank‘s assessment of the macroeconomic situation, an implicit condition for the program. The Bank focused on ensuring sustainability of Tanzania‘s external debt following the substantial debt cancelation under HIPC. This was intended to maintain the accelerated economic growth that was seen to be in part a result of improved fiscal performance in the late 1990s. The program thus encouraged prudent management and contracting of external debt by emphasizing the following actions:  adoption of a national debt strategy (PRSC-1);  amendments to the Loans, Guarantees, and Grants Act Number 30 of 1974 to ensure a prudential debt contracting and management system for government and public independent institutions (PRSC-1);  amendments of Treasury Registrar‘s Ordinance (1959) and other laws that relate to the procurement of debt (PRSC-2);  a comprehensive review of all laws that relate to the procurement of debt to ensure consistency with the provisions of, and elimination of loopholes concerning the enforcement of the amended Loans, Guarantees and Grants Act Number 30 of 1974 (PRSC-2);  preparation of an annual borrowing and repayment plan (concessional and non- concessional loans and attention to limits), presentation of this to parliament as part of the annual budget (PRSC-2);  the development of a complete inventory of- and a verification and audit of- the contingent liability component of domestic debt (PRSC-2); 94 ANNEX C  the establishment of a National Debt Management Committee which meets quarterly to advise the Minister o f Finance on all proposals for new credit and to monitor the implementation of the National Debt Strategy (PRSC-3). 2. Table C.2 outlines the status of the triggers and prior actions used as catalyst in effecting policy actions for macroeconomic stability and debt management under the PRSC series. The policy actions were to be assessed by using three key indicators: the net present value of public external debt to exports; debt service (domestic and foreign) (ratio between actual and budgeted); and gross borrowing (domestic and foreign) (ratio between actual and budgeted). There was substantial improvement in macroeconomic management during 2003- 05. Reforms in debt management were successful, in part because the PRSCs overlapped with the final stages of HIPC initiative, as well as with the Multilateral Debt Relief Initiative process which created strong incentives for Tanzania to reform its debt management system and introduce institutions for prudent macro-policies. All envisaged debt targets were met, and the country had significant scope to expand its fiscal space at the end of the period. With debt cancellation under HIPC and the Multilateral Debt Relief Initiative, the debt to GDP declined to 60 percent in 2005, while debt service declined from the peak of almost 40 percent to exports to about 5 percent. The net present value of debt-to-export ratio was estimated to be 64 percent by 2006. These figures were considered well below debt sustainability thresholds (World Bank, 2007e, p14). Table C.2. Triggers and Prior Actions under Macroeconomic Stability and Debt Management Trigger Prior Action Status PRSC-1 Presented the amended Loans, Completed Guarantees and Grants Act Number 30 of 1974 for parliamentary approval to ensure a prudential debt contracting and management system for government and independent public institutions PRSC-2 Progress in the implementation of the Prepared annual borrowing and Completed recommendations of the national debt repayment plan (both concessional strategy according to a work plan and non-concessional loans), inclusive (including activities, timeframe, and of borrowing limits, and presented it responsibilities) to be approved by to Parliament as part of the annual government. budget Source: Program Documents for PRSC 1-2 (World Bank, 2003, 2004a). 3. Private sector development was perceived as essential for growth and poverty reduction. The PRSC built on the tangible results under PSAC, which included the advancement of the privatization agenda, with government disengagement from commercial activities almost completed and significant progress toward greater private sector involvement in the provision of utilities; progress in strengthening the business environment and reform of land legislation; and the establishment of an improved regulatory environment. The PRSCs focused on the three primary constraints, focused on the lack of a private sector development strategy and small and medium enterprise policy, rigid labor laws, and a need to improve the business environment. 95 ANNEX C 4. The lack of a comprehensive strategic framework to help with coordination and targeting of the various initiatives in support of private sector development was seen as major problem, as earlier efforts had been piecemeal. Under PRSC-1, the Ministry of Industry and Trade prepared a small and medium enterprise policy and launched an associated public awareness campaign. The government also initiated the preparation of an integrated private sector development strategy, which was approved under PRSC-2. 5. The lack of a single comprehensive labor law was a major problem as the labor legislation was encompassed in six different, outdated pieces of legislation. This fragmentation impeded market-oriented deployment of labor. The government was expected to prepare a new labor policy and approve it under PRSC-2, while in the context of PRSC-3 a new Labor Act was expected to be prepared and presented to parliament for approval. 6. There was a significant unfinished agenda in improving investment climate and business environment. In July 2002, the government approved the Business Environment Strengthening for Tanzania program for carrying forward the reform agenda to improve the business environment, partly supported by the PSAC-1. The PRSC focused on the implementation plan for the business environment program that targeted eight priority areas of action: (i) harmonization of local government taxation to remove excessive tax burden on private enterprise; (ii) resolution of outstanding issues emanating from the Land Act; (iii) review and amendment of labor laws constraining business operations; (iv) streamlining of work permit procedures; (v) review and amendment of licensing legislation to reduce the cost of business establishment and continuation; (vi) review and revision of export-import procedures to reduce time costs and corruption-related costs; (vii) development of competition policy, incorporating advice given by the Facility for Investment Climate Advisory Services; and (viii) design and implementation of a program for enhancing access to commercial courts by small and medium enterprises. 7. The industry growth rate for the 2003-05 period was strong, at 10 percent per annum, meeting the targeted range of 8 to 10 percent. This rapid growth was probably due to external factors—such as a supply-side response in the mining sector to high commodity prices— rather than government reforms supported by the Bank‘s program. Moreover, the impact of industrial growth on poverty is limited because the labor intensity of mining is low, and the manufacturing sector remains small and concentrated in Dar es Salaam. 8. By the available indicators, the business environment did not improve: the results from the two investment climate assessments roughly corresponding to the beginning and end of the period (2003, 2006) suggest no major change.59 Implementation of private sector development reforms under PRSCs 1-3 remained in its early stages, as shown in Table C.3. Implementation of the small and medium enterprise strategy was much slower than expected due to insufficient funding from the budget. The private sector strategy that was expected to be completed under PRSC-3 was delayed because the government wanted to first complete the Diagnostic Trade Integration Study, which was finalized in June 2005. 59 Most Doing Business indicators at disaggregated level are missing for 2004-05. 96 ANNEX C 9. Phase-1 labor legislation covering employment relations, collective labor relations, dispute resolution, and labor market institutions was submitted to the Parliament in November 2003, and approved in April 2004. A task force assigned to do the preparatory works for phase-2 legislation (covering occupational safety, health workers compensation, and employment and skill development) was expected to be finalized under PRSC-3 (2005), but was postponed for 2006 ( and was not followed up by the second PRSC series). 10. Following approval of the business environment program under PRSC-2 the efforts shifted focus to replacing the business licensing regime by a registration system. In particular, measures to rationalize business licensing announced in the budget for 2004/05 and subsequently implemented included the abolition of business license fees for health facilities operated by religious organizations and small businesses, as well as the replacement of annual business license fees by a one-off fee payable at the time of business registration. A draft bill on Business Activities Registration that outlined the direction of the business environment reforms was submitted to Parliament in July 2005. Table C.3. Triggers and prior actions for private sector development Trigger Prior Action Status PRSC-1 Approval of an implementation plan for the Business Completed Environment Strengthening in Tanzania program by the integrated framework steering committee. PRSC-2 Phase I labor legislation (Employment Relations, Completed Collective Labor Relations, Dispute Resolutions, and Labor Market Institutions) presented to parliament. Reviewed the business licensing system after Completed consultation with stakeholders, prepared a position paper on business licensing reforms, and submitted to Parliament amendments to the Business Licensing Act, 1972, introducing reforms of the business licensing system. PRSC-3 New business licensing New business licensing framework under Completed framework under implementation in a phased strategy. implementation in a phased strategy. Source: Program document for PRSC 1-3 (World Bank, 2003, 2004a, 2005). 11. Agriculture and rural development were considered critical in broadening the impact of growth to generate poverty reduction given the preponderance of poverty in rural areas. Implementation of the government‘s agricultural sector development strategy and, within it, the Agriculture Sector Development Program, both issued in 2001, focused on resolving the constraints facing agriculture through a transformation from subsistence to commercial agriculture, thereby help combat both rural and urban poverty. The underlying themes of the strategy were to create a favorable environment for commercial agriculture, to delineate public and private roles in the sector, and to decentralize service delivery responsibilities to local governments. Insecure land rights affecting the use of land as collateral for credit; limited access to agricultural credit in general, high levels of agricultural taxation, as well as limited access to markets as a result of poor transport infrastructure were identified as key constraints to agricultural sector growth. 97 ANNEX C 12. The government‘s Development Program included a broad array of measures to address the identified constraints in the sector. From there, the PRSCs specifically sought to support this program by focusing on reforms of crop boards particularly with respect to input supply, product quality, and competitiveness; micro-finance legislation; the institutional arrangements governing rural roads; local government taxation; property rights; and overall coordination of public interventions in the agricultural sector. The program document for PRSC-1 expected the proposed reforms to contribute to agricultural sector growth on a sustained basis by 5 percent annually. 13. The envisaged outcomes for the sector, derived from the program were: an increase in real agricultural GDP from 3.6 percent in 2000/01 to an average of 5 percent by 2010, resulting in the 2015 Millennium Development Goals targets of: (i) increased share of income or consumption accruing to the poorest 20 percent, (ii) reduced share of population living on less than USD1 per day, and (iii) increased ratio of trade to GDP. Growth performance in agriculture was achieved; despite the impact of the drought on agriculture in 2003, agriculture grew by about 4 percent and rose further to 6 percent in 2004 and 5 percent in 2005. There were no data to assess the poverty impact of growth, but a poverty assessment in 2008, suggests that the poverty reducing impact of agricultural growth could be marginal. Table C.4. Triggers and Prior Actions for Agriculture Trigger Prior Action Status/Comment PRSC-1 Finalized the ASDP framework and Completed process document and adopted it for implementation. PRSC-2 Prepared and issued a prioritized strategic Trigger not met; plan and a corresponding program to redefined as strengthen the administrative capacity for trigger for PRSC- implementation o f the Land Act and 3 Village Land Act. PRSC-3 Finalized and obtained government Finalized and obtained government Completed, but no approval of the strategic plan for approval of the strategic plan for evidence of operationalization of the Land and operationalization of the Land and operationalization. Village Land Acts. Village Land Acts. Building on the results of the Crop Building on the results of the Crop Partially Boards review, government approval of a Boards review, government completed strategy to reform two crop boards approval of a strategy to reform two consistent with ASDS. crop boards consistent with ASDS. Sources: Program documents for PRSCs 1-3 (World Bank, 2003, 2004a, 2005). Note: ASDP = Agriculture Sector Development Program; ASDS = Agriculture Sector Development Strategy 14. The implementation of policy reforms in the agriculture sector supported by the PRSCs was slow and weak as shown in Table C.4. The passing of the Land (Amendments) Act in 2004 provided for improved clarity of property rights and could help facilitate securitization of land. On the other hand, the trigger for PRSC-2 to support its implementation was only met under PRSC-3. The preparation of the land sector strategic plan was initially expected to start in October 2003 and be completed by March 2004, with financial support from EU‘s European Development Fund. The start of this activity was delayed as agreement on the work plan, scope of work, and budget for the preparation of the 98 ANNEX C strategic plan between the Ministry of Lands and Housing Services and EU took longer than originally envisaged. As a consequence, the reform was redefined as a trigger for PRSC-3. The strategic plan was discussed during a stakeholders meeting in March 2005. Comments from stakeholders were incorporated into the draft and the strategic plan was submitted to the Ministry, but no evidence of immediate ―operationalization‖ was subsequently reported. Similarly, the approval of the strategy to reform two crop boards was partially met and was converted to a trigger for the follow-up PRSC series. Cabinet approved reforms to the financing of crop boards, but did not endorsed specific reforms of role and function of crop boards, including their regulatory functions, governance structures, and accountability to stakeholders.60 Objective 2: Supporting results orientation of public service delivery. 15. The program documents suggested few particular indicators related to the achievement of this objective. A summary of these indicators, with one added, is shown in Table C.5. Table C.5. Indicators of achievement on results orientation objective. Indicator 2001/02 2002/03 2003/04 2004/05 Budget Share of priority sectors in total 46% 46% 45% focus on spending priority Number of HIPC expenditure 8 9 11 sectors tracking benchmarks met Budget Share of votes with budget 35% 20% execution deviations greater than 5 percent Monitoring Updated annual plans and SDSs of uneven quality and performance budgets for all carried out, but not fully evaluation MDAs taking into account captured in strategic results of service delivery planning process surveys (SDSs) M&E performance monitoring Serious lack of data on the system functioning in all MDAs outputs and outcomes of indicating increased outputs government services, with from public service exceptions in health and education; Frequency of Household Budget Latest: Latest: Latest: Latest: 2000 Surveys for poverty monitoring 2000 2000 2000 Source: Program documents for PRSCs 2-3 (World Bank, 2004a, 2005) Italicized indicators were not identified in program documents but collected for this evaluation. 16. Budget formulation and execution was a possible contributor to this objective, given the poverty reduction strategy‘s emphasis on increasing budgetary funding to the priority sectors, which included education, health, agriculture, roads, water, the judiciary, and 60 The joint evaluation of General Budget Support (ODI and DAL 2005) in their assessment of the implementation of this policy noted that it was not so clear that there was a strong political thrust in favor of it from the government side. The report further concluded that agricultural policy could have been more coherent if the three ministries involved directly in the sector were subjected to the common discipline of the MTEF/Budget Guidelines/PER processes. 99 ANNEX C HIV/AIDS, which were seen as having the greatest development impact. The PRSC series continued the reforms in public expenditure management through budget formulation initiated under PSAC-1 toward strategic allocation of public spending to meet PRSP targets, as indicated by target spending shares for social sectors. As noted in the table, the share of the budget allocated to priority sectors was flat during the period, with the actual allocations rising in real terms. In addition, the government adopted a recommendation from the Bank- supported 2003 PER that requires each agency to explicitly link proposed budget allocations with MKUKUTA objectives. 17. Key achievements under PSAC in public expenditure management included introducing cash budgeting in 1996/97 and adopting a participatory PER/MTEF approach to monitor, inform, and improve the quality of government expenditures, enhancing funding for key activities in the priority sectors, and ensuring a greater proportion of donor support was transferred though the budget. The PRSC process identified a number of residual areas on which to focus:  approved budget (delineating priority sector budget codes in the context of PRSC-2 and PRSC-3) in line with PRSP objectives and continued protection of priority sector allocations from cuts during budget execution;  three-year, rolling budget guidelines, consistent with PRSP priorities (and resource needs identified during the PRSP-sectoral review process), issued to Ministries, Departments and Agencies (MDAs) and Local Government Authorities (LGAs);  strengthening of policy units in all key or priority sectors of MDAs;  use of the IMF‘s Government Finance Statistics classification for preparation of LGA budgets, as well as for the FY2005 development budget, and preparation of 2004 LGA budgets using the MTEF format;  consistent reporting on budget execution starting in FY2003 as per identified expenditure budget codes for priority sectors and items;  expenditure tracking work undertaken and recommendations to strengthen the reporting system implemented;  provision of user-friendly information to the public on budget performance, including LGAs‘ budget and accounts information; and  quarterly budget execution reports for LGAs submitted to the MoF (via Prime Minister‘s Office–Regional Administration and Local Government). 18. The PRSC used public expenditure management prior actions that were consistent with the PRSP/MKUKUTA objectives as well as expenditure outturns with projected budget allocations and reporting of expenditures by budgetary codes. The government completed all prior actions related to public expenditure management reforms under the PRSCs (Table C.6), but the program was only partly successful in terms of outcomes. According to the PER of 2004/05, which covered the period of the first series of PRSCs, there were a number of areas in which progress had been less than expected. In particular, in terms of consistency of budget with actual expenditures, the PER found that the aggregate deviation indicator based on MDA‘s (spending units) might have hidden the fact that ―over-spending‖ units leaned toward more of administrative expenditures (public administration, defense, foreign affairs), while under-spending units constituted those with more expenditures directly going to 100 ANNEX C growth and poverty reduction (health, education). This pattern raises concern about whether the monitoring indicator on the share of votes with deviations greater than 5 percent was the adequate one to use for assessment. Table C.6. Triggers and Prior Actions for Public Service Delivery Trigger Prior Action Status PRSC-2 Approved budget 2003/4 in line with PRS Completed objectives, delineating budget codes for priority sectors and items Budget execution for 2002/03 and 2003/04 Completed (FQ1 and FQ2) in line with the approved budget and with PRS priorities, consistently reported as per identified expenditure budget codes for priority sectors, and in 2003/04 also by identified codes for priority items Local government taxes and levies rationalized. PRSC-3 Approved budget 2004/05 in line with Approved budget 2004105 in line with PRS Completed PRS objectives, delineating budget objectives, delineating budget codes for codes for priority sectors and items. priority sectors and items. Budget execution for 2003/04, and Budget execution for 2003/04, and 2004/05 Completed 2004/05 (FQ 1 and FQ 2), in line with (FQ 1 and FQ 2), in line with approved approved budget and with PRS budget and with PRS priorities, consistently priorities, consistently reported as per reported as per identified expenditure budget identified expenditure budget codes codes for priority sectors and items. for priority sectors and items. Source: Program documents for PRSC 1-3 (World Bank, 2003, 2004a, 2005). Note: All triggers and prior actions in this area were focused on budget preparation and execution, and were not directly linked to results orientation of public services. 19. The PER also concluded that public investment projects remained not quite in line with the growth focus of NSGRP. There had been no major shift in investments toward infrastructure despite the fact that infrastructure had been identified as the main constraint for growth. In addition, it noted that in spite of impressive efforts made in enhancing the link between policy and resource allocation by strengthening the formulation of medium term budget strategy, there was still the need to develop strategic nature of the Budget Guidelines so that they become a comprehensive instrument for guiding public finances over the medium-term. The PER also found that in spite of the recommendation made in the previous PER external evaluation, there had been little progress in compiling actual expenditure data for the local government authorities. The main reasons for slow of progress seemed to be weak capacity in local government authorities and the Prime Minister‘s Office-Regional Authorities and Local Government and the lack of a standard reporting format, which were areas of emphasis of the PRSCs. 20. With respect to resource mobilization issues, there was only one prior action targeted at the rationalization of local government taxies and levies in PRSC-2. While this policy action was completed, it introduced unintended consequences. The government abolished nuisance taxes in the 2002/03 budget, and introduced compensatory grants to local government authorities to replace the lost revenue. But these grants were characterized by: (i) delays and unpredictable flow of the compensatory funds from the Ministry of Finance; (ii) 101 ANNEX C inadequacy of remaining sources of revenue for local government authorities and a further erosion of their fiscal autonomy; (iii) the inability of councils to transfer a share of the compensation grants to villages; and (iv) a weakened drive to strengthen good governance. Some local government authorities invented mechanisms to cope with the abolition of various levies, including by collecting more from city service levies and property taxes, selling rights to erect bill boards, and dropping some projects. Objective 3: Enhancing public service performance. 21. The relevant triggers and prior actions for this objective included core public financial management (PFM) policies (implementing the recommendations of the Country Financial and Accountability Assessment and linking them with the PFM Reform Program working group and the technical assistance project) and pay enhancement for the Civil Service (linking with the Public Service Reform working group and technical assistance project), as shown in Table C.7. Prior actions for procurement reform—which could improve efficiency and help minimize leakage from the budget—and implementation of the recommendations of the Country Procurement Assessment Review were included in PRSC-2 and 3. PRSC-2 also supported capacity building in the president‘s office for monitoring local government finances, the expansion of which was being assisted by the 2004 Bank‘s Local Government Reform operation, part of which served as budget support for local governments. Table C.7. Triggers and Prior Actions under Enhanced Public Sector Performance Trigger Prior Action Status/Com ment PRSC-1 Approval by the Government of Tanzania of the revised PFM Completed Reform Program, which includes the key agreed recommendations of the Country Financial and Accountability assessment. Implemented pay enhancement for civil servants in line with the Completed approved budget for 2002/03. PRSC-2 Progress in the implementation Joint PFM Reform Program Steering Committee reviewed (i) Partially of the revised PFM Reform the establishment of a management structure and (ii) the Completed Program, including development detailed annual work plan and budget for the first phase of of a monitoring system and implementation of the PFM Reform Program. regular reports on progress. Pay enhancement in line with the approved budget for FY04. Completed Prepared a draft bill amending the Public Procurement Act of Completed 2001, reflecting Assessment recommendations. Local Government Authority Tender Boards constituted and established under the new Regulations. PRSC-2 Progress in strengthening and Progress in strengthening and sustaining capacity of PMO- Completed sustaining capacity of the RALG for collecting, collating, and analyzing administrative coordination and monitoring unit data according to an updated action plan to be approved by in PMO-RALG [Prime government. Minister‘s Office-Regional and Local Government] for collecting, collating, and analyzing administrative data according to an updated action plan to be approved by government. 102 ANNEX C PRSC-3 Pay enhancement consistent Pay enhancement implemented in the 04/05 budget execution. Completed with the approved budget for Pay enhanced by between 9.09 percent and 15 percent in line 2004/05, and the overall thrust with the medium term pay policy and the wage bill ceiling for of the pay reform strategy. FY 04/05. Progress in the implementation (i) MTEF - consistent budget coding system made good Completed of the PFM Reform Program. Progress. PF objectives are compatible with the MTEF objectives as well as consistent coding of PFM Reform Program project funded, basket funded, and government funded activities within the government development budget for financial year 2005/2006. (ii) Monitoring indicators for each component at output level: (achieved). For each component, core activities and monitoring indicators have been prepared in draft. The government establishes the The Procurement Bill was approved on November 12, 2004 and Completed Regulatory Authority for assented by the President in February 2005. The new procurement and decentralizes Procurement Regulations were gazetted. The Central Tender procurement to Procurement Board ceased to exist and in its place was established a Public Authorities. Procurement Regulatory Authority; and all MDAs shall carry out procurement functions within their respective approved budgets. Source: Program document for PRSC 1-3 (World Bank, 2003, 2004a, 2005). 22. Prior actions with respect to public service performance were implemented almost in their entirety. With respect to public financial management, progress was made on most of the components of the PFM Reform Program, particularly in areas supported by parallel projects (such as, further development of the integrated financial management system, external resource management and procurement). In the 2006 Public Expenditure and Financial Accountability report, stronger than average performance was observed in cash management and policy-based budgeting, with average performance in accounting and reporting, comprehensiveness and transparency, procurement, and budget credibility. Although PRSCs 1-3 concentrated its targets and prior actions on these issues, one cannot say whether the PRSCs led to the better performance in those issues, or whether those issues were picked up in the PRSCs because performance was good anyway.61 On the other hand, fiduciary concerns—external scrutiny, and auditing, which were not in focus under the PRSC—got lower Public Expenditure and Financial Accountability scores, while establishing the PFM Reform Program as a coordination mechanism took more time than expected. 23. Civil service pay reform was part of a longer agenda that had included an earlier phase of downsizing in the 1990s, which had succeeded better than in most countries (PER 2002; IEG 2008), largely because it came at a time when privatization and private sector growth opened job opportunities in the private sector, while inflation decimated the real value of public sector salaries, especially at the upper levels leading to loss of staff. In the 2000s, the remaining problems were the inadequate salaries at upper levels to compete with the private labor market, the offsetting arrangements to have allowances make up for low regular salaries, and the insufficient incentives for teachers and health specialists to work in 61 As noted in the PPAR on PRSCs 4-8, the core budget areas were not singled out as targets in the second series, and the 2009 PEFA scores declined in all five. The revenue administration and fiduciary areas, on the other hand, had triggers in the second series, and their PEFA scores all improved from 2006 to 2009. Irish Aid and DFID (2011, p.36). 103 ANNEX C under-served areas. While the pay raises during PRSCs 1-3 met the triggers, they were not well targeted and did not make progress on the difficult issues, which reflected a problem with program design in terms of targeting.62 Rather than enhancing pay for positions that were most difficult to fill, such as service-delivery personnel in remote areas, employees were permitted to access travel allowances that exceeded the cost of travel. This was, in effect, a salary supplement. But it was concentrated among civil servants with strong connections who were not on the front lines, and gave them incentive to travel excessively away from their jobs (Twaweza, 2010). 24. Procurement received substantial attention, with prior actions in PRSC-2 and 3. PRSC-1 included an annex with the recommendations of the 2003 Country Procurement Assessment Review, which were prepared ―in parallel‖ with the PRSC and found that ―procurement is a neglected function‖ (Country Procurement Assessment Review, 2003, p. 121). The centerpiece of procurement reform, which was important for improving the effectiveness of public investment and for reducing corruption, was the transfer of procurement activity from the central tender board to ministries, agencies and local government, and the establishment of a Public Procurement Regulatory Authority (PPRA). In accord with recommended best practice, the PPRA would enforce regulations and provide a somewhat independent check on the procuring entities, ensuring a more competitive process and reducing leakage of government resources. By 2006, the procurement assessment for the PEFA review was giving good marks for the improved set up of the procurement process, although there was not a track record yet of implementation. The Auditor General‘s report on Central Government for 2006/07 also gave good marks for procurement, but noted that ―justification for use of less competitive methods is weak and is not yet routinely monitored.‖ 25. Although there were no triggers in PRSCs 1-3 for explicit anti-corruption measures, the concern with corruption and possible remedies was mentioned over 40 times in the program document for PRSC-1. It was also a policy trigger for the high-case scenario in the 2001-03 CAS, which required the ―dissemination of the national and sector specific anti- corruption plans; implement sector specific action plans against corruption in the seven areas surveyed‖ (World Bank 2003). In general, the improved financial management, procurement and transparency during the implementation of the PRSC series seemed to have had beneficial anti-corruption effects, in line with the findings of the PSRC evaluation (IEG, 2010). Although subject to technical limitations, Tanzania‘s International Country Risk Guide (ICRG) rating on corruption improved from 2 (on a 1-6 scale) to 2.6 in 2006 and 3.5 in 2007 (before declining again), and the Bank‘s Country Policy and Institutional Assessment (CPIA) rating for transparency and anti-corruption was up to 3.5 in 2005-07, after which it also declined. 63 62 The program document for PRSC-3 noted: ―At present the gains made in reducing the number of allowances at the beginning of the program seem to be eroding.‖ (World Bank, 2005, p. 37). 63 The CPIA is based on Bank economists‘ and sector specialists‘ rating of 20 items in four categories. The rating scale is 1-6, with 6 representing the highest achievement. 104 ANNEX C 26. The Bank‘s support for decentralization in Tanzania (referred to as decentralization by devolution) came mainly through the LGRP credit, which was partly for local capacity building and partly for formula-based grants to support local government investment, which was where local government had the most control. To complement this, PRSC-2 supported the strengthening the capacity of the central administration to monitor activity by local governments (hiring 3 staff and for getting information technology equipment in the central office and some regional offices) and extending decentralized procurement authority to local governments.64 The Public Procurement Regulatory Authority was intended to regulate the local government tender boards. The approach did not address the largest areas of nominal sub-national spending—in education and health—as the local governments had little discretion in these areas. As the program document for PRSC-2 (World Bank, 2004a, p. 40) noted, ―the lack of a clear and unified position with respect to decentralization of human resources management raises concern and it is hoped that the study will provide a clear direction for the way forward.‖ This assessment concurs. In PRSC-3 there were no triggers or prior actions for decentralization, but rather reports of delays and plans to make more progress in the future. Objective 4: Strengthening environmental management. 27. The PRSC introduced a new objective under PRSC-2 to strengthen environmental management because, it was argued, the poor were heavily dependent on the environment for both income generation and consumption. There were no triggers in the series designed to achieve the objective; rather the PRSC used dialogue to support the government‘s efforts to address environmental issues identified in National Environmental Action Plan (1994) and the National Environmental Policy (NEP) adopted in 1997. This was done through three approaches: (a) mainstreaming of environmental concerns into the dialogue on poverty reduction through the budget process and sectoral policy dialogue; (b) improving understanding of the poverty-environment linkages in Tanzania; and (c) strengthening institutional capacity to integrate environmental assessment (EA) procedures into sectoral strategies and Policies. 28. In December 2002, the Cabinet approved an Institutional and Legal Framework for Environmental Management Program, which provides the basis for Environmental Impact Assessment, cross-sectoral coordination, and integration of environment management into local government planning processes. A number o f key sector policies and strategies in forestry, agriculture, fisheries and mining that followed reflected an increasing focus on internalizing environmental sustainability objectives and empowering the poor for improved resource management. 29. The Bank conducted under PRSC 1 and 2 various pieces of economic and sector work focused on identifying the linkages between key macro and structural reforms policies and the environment. This analysis identified measures at both national and local government levels to reduce poverty and enhance environmental sustainability of policy reforms. The 64 See, for instance, United Republic of Tanzania 2004b for the earliest Local Government Fiscal Report in which the Prime Minister‘s Office-Regional and Local Governments monitors local government budgets. 105 ANNEX C analysis highlighted two findings linking the environment and poverty: (a) it is important to ensure that poverty reduction programs do not damage the resource base and the environment of which poor depend for their livelihood; and (b) improving environmental conditions could help reduce poverty and promote sustainable pro-poor growth. It is not clear what the nature of the follow-up to these findings was. 30. Through support for an intensive consultation process, PRSC-3 assisted in the passage of the Environmental Management Act (EMA) by Parliament. An initial needs assessment for implementation of the EMA was conducted as part of the first PER on environment. A draft Environmental Impact Assessment training manual was prepared, as well as a Strategic Environmental Assessment for the PRSC. The ICR for PRSC 1-3 noted that there were ongoing discussions regarding the potential for initiating a Sector Wide Approach for environmental management to support the implementing the National Environmental Management Act, but this did not materialize. 31. PRSC-2 specified five indicators to monitor, none of which were tracked, nor were available in the field. Targets for three of these indicators were reduced dramatically in the program document for PRSC-3. For instance, the target for the share of sectors with environmental units was reduced from 90 percent to 40 percent. 106 ANNEX D Annex D. PRSCs 4-8: Detailed Assessment of Progress by Objective Objective 1: Growth and reducing income poverty. 1. The first outcome area covers the program‘s contributions to macroeconomic policy and improving the business environment. Indicators for this outcome are summarized in Table D.1—two targets were not met, and the modest PRSC target for the third was met while the more ambitious MKUKUTA target was not. Table D.1. Indicators for progress in improving the business environment and macroeconomic policy. Indicator Baseline Target Latest Latest Source Result (ICR) (IEG) Fiscal deficit (after grants) 6.5% Ministry Not met 6% 5% (from PSI) 6.4% as percent of GDP (2010/11) of Finance 4% Not met Inflation rate (average, 7% Ministry 5.1% (MKUKUTA) 10.6% CY) (2010/11) of Finance < 5% (PRSC) 99 Doing PRSC target Ranking on ―Doing 128 136 (MKUKUTA) Business met Business Indicators‖ 130 (PRSC) 2011 2. The Bank did not devote as much attention in this series of PRSCs to policies to support macroeconomic stability as in the first series, with the IMF continuing in its lead role through the Policy Support Instrument (PSI). The PRSC input was limited to strengthening of budget and public expenditure management, without specific triggers related to debt management. This PRSC series focused on two broad policy actions: (a) approved budget in line with policy objectives and (b) expenditure outturn consistent with approved budget as shown in Table D.2. These results were under the ―Resource Allocation and Budget Consistency‖ section of the result framework. Residual items from the first series of operations, such as inadequate budget reporting by the local governments, was not specifically followed up in the second series of PRSCs. Similarly, the PER reforms were categorized under the objective of ―improved state of governance and accountability.‖ The aggregate deviation of expenditures for recurrent budget was the only monitoring indicator applied here and differed from the indicator used in the first series of PRSCs, which was the percentage of agencies with deviation above 5 percent. 3. Fiscal sustainability and debt management were monitored in the context of discussion on fiscal deficit (with a target of below 5 percent of GDP) and CPI inflation (annual inflation below 4 percent). The fiscal deficit rose above its target level during 2008- 10 due to the poor revenue performance since 2008, a surge in discretionary spending, and the fiscal stimulus put in place to withstand the global crisis (as advised by development partners). By 2011, Tanzania‘s debt to GDP ratio was close to 40 percent, up from the lowest level of around 25 percent on the onset of the program in 2005, but below the MKUKUTA target of 50 percent. Both revenue and expenditure factors led to higher than expected contracting of debt. The tax revenue to GDP ratio did not improve since 2007; and there was 107 ANNEX D a rapid increase in expenditures (not related to the fiscal stimulus following the crisis). The PRSCs paid little attention to macroeconomic stability in the series, and so it cannot be plausibly argued that it contributed directly to these outcomes. Table D.2. Triggers and Prior Actions for Macroeconomic Stability Trigger Prior Action Status/Comment PRSC-4 Approved budget 2005/06 in line with Approved budget FY06 in line with Completed MKUKUTA (PRS) implementation, MKUKUTA implementation, delineating budget codes for budget delineating budget codes for budget activities related to MKUKUTA goals activities related to MKUKUTA and strategies. goals and strategies. Budget execution for 2004/05, and Expenditure outturn for FY05 Completed 2005/06 (FQ 1 and FQ 2), in line with consistent with approved budget. approved budget and with MKUKUTA implementation, consistently reported as per identified budget codes for budget activities related to MKUKUTA goals and strategies. PRSC-5 Approved budget for FY07 in line with Approved budget for FY07 in line Completed policy objectives (MKUKUTA}. with policy objectives (MKUKUTA). Expenditure outturn for FY06 consistent Expenditure outturn for FY06 Completed with approved budget. consistent with approved budget. PRSC-6 Approved budget for FY08 in line with Approved budget for FY08 in line Completed policy objectives (MKUKUTA, sector with policy objectives policies). (MKUKUTA, sector policies). Expenditure outturn for FY07 consistent Expenditure outturn for FY07 with approved budget. consistent with approved budget. PRSC-7 Approved budget broadly in line with Approved budget broadly in line Completed policy objectives (MKUKUTA, sector with policy objectives policies). (MKUKUTA, sector policies). Expenditure outturn consistent with Dropped Dropped approved budget and recurrent budget deviation reduced. PRSC-8 Expenditure outturn for FY08/09 Expenditure outturn for FY08/09 Completed consistent with approved budget and consistent with approved budget and recurrent budget deviation reduced. recurrent budget deviation reduced. Source: Program document 4-8 (World Bank, 2006, 2007b, 2008b, 2009a, and 2010c). 4. Improving the business environment was central to the effort of the MKUKUTA to promote shared growth. PRSC-4 highlighted residual actions from the first PRSC series that needed attention. These included supporting a private sector development strategy, a second phase of labor legislation, and improving the investment climate. Triggers and prior actions (Table D.3) were identified for reforms to improve the business environment, while the progress in the private sector development strategy and the second phase of labor legislation was expected to be achieved through policy dialogue. 5. The expected outcome was an improvement in Tanzania‘s Doing Business ranking, with a MKUKUTA target value of 99 (and a PRSC target of 130) compared to the baseline of 136. The MKUKUTA target was not met—overall improvement in the Doing Business indicator under the second series of PRSCs was insignificant—although the much more 108 ANNEX D modest PRSC objective was achieved. There was progress on an absolute scale for some of the sub-indicators, notably for the indicator of the cost of starting a business (down from 187 percent of per capita GDP in 2005 to 37 percent in 2010 ) and trading across borders, both in terms of number of documents and number of days needed for transactions. For imports the number of documents was reduced from 16 in 2005 to 7 in 2010, and the number of days was reduced from 51 to 31. For exports the number of days was reduced from 30 to 24. There was, however, deterioration for other sub-indicators, including the number of days required for getting construction permits that increased from 276 to 303 days, and the cost of getting a construction permit that increased from around 100 percent of per capita GDP in 2005 to around 1,200 percent of per capita GDP in 2010. Tanzania‘s ranking remains poor even by regional standards.65 Table D.3. Triggers and Prior Actions Improving Business Environment Trigger Prior Action Status/Comment PRSC-4 Draft bill on Business Activities Draft bill on Business Activities Completed Registration submitted to Parliament. Registration submitted to Parliament. PRSC-5 Progress in the reform of the Business Progress in the reform of the Completed Activities Registration and the Business Activities Registration Business Regulatory Licensing and the Business Regulatory Regime. Licensing Regime including reflecting private sector views in the redrafting of the business activities registration bill. PRSC-7 Business Activities Registration Act Dropped subsidiary regulations consistent with one-stop-shop good practice approved. Source: Program document 4-7(World Bank, 2006, 2007b, 2008b, 2009a). 6. The implementation of private sector development reforms under PRSC4-8 was slow and faded out after PRSC-5. Initial delays in preparation of the private sector development strategy were prompted by the need to complete a Bank-initiated Diagnostic Trade Integration Study, although completion of the trade study did not lead to follow-up activities on the private sector development strategy. Similarly, the second phase of labor legislation that was also expected to be part of the policy dialogue was delayed. On the other hand, two of the three triggers related to the legislation for business environment reforms were carried out and converted into prior actions (Table D.3). The legislation was submitted to the Parliament in July 2005 and was approved in January 2007. The expectation that the government would follow up on the adoption of legislation by introducing governmental decrees for establishing one-stop-shop business registration system did not materialize. The associated trigger was dropped and subsequently led to the USD 10 million reductions in the size of PRSC-7. There was minimal plausibility linking policy actions carried out under the PRSCs to the very modest improvements in the business environment as measured by the Doing Business Indicators. 65 Indicators that are based on country rankings are not ideal—a country may make significant progress, but slower than other countries, and this would not be reflected in the indicator. 109 ANNEX D 7. The second outcome area for the objective of growth and reduction of income poverty—increased access of private sector to key inputs—covers the program‘s contributions to energy and transport sectors, which were expected to provide backbone services. Indicators for this intermediate outcome are summarized in Table D.4. The country met PRSC targets for two of the five indicators, and nearly met two others. Table D.4. Indicators of progress on increased access of private sector to key inputs Indicator Baseline Target Latest Latest (IEG) Source Result (ICR) Credit to private sector 8.9% 13% 17% Met (% of GDP) Time taken for a Transit: 16.3; Ports container from off- 20.1 12 days 12.5 days Local: 12.3 Authority Nearly met loading until clearing (2010) port (import data) 1278 Installed MW power 1072 (2010) 889 (MKUKUTA)/ 1077 TANESCO Nearly met capacity 1309 (2012) 1091 (PRSC) Reduction in the duration, Availability of power 62% 71% 85% but not TANESCO Ambiguous supply a/ frequency of outages Share of national and (Good) 53% Good 47% Good 75% Good 56% Roads regional roads in good (Fair) 35% Not met Fair 35% Fair 20% Fair 33% Fund and fair condition (2009/10) Budget coverage of 116.9 64.2 Roads routine road Deterioration (2007/08) (2010/11) Fund maintenance need (%) Budget coverage of 54.0 57.9 Roads Modest periodic road (2007/08) (2010/11) Fund improvement maintenance need (%) Note: Italicized indictors were not included in the results framework. These are added here to provide context. a/ The Implementation Completion and Results Report defines availability of power supply to be electricity capacity utilization. 8. A reliable and affordable supply of electricity was considered essential by the MKUKUTA for economic growth and poverty reduction in Tanzania. The PRSC, as articulated in the program document for PRSC-5, sought to help the government improve access to electricity for the private sector, reduce vulnerability of the sector to climatic conditions, achieve greater installed capacity, and help Tanzania Electric Supply Company (TANESCO) to achieve full cost recovery. These reforms were expected to (i) address energy generation shortages by expanding capacity and reducing reliance on hydro power; (ii) improve TANESCO‘s autonomy and commercial focus to avoid repetition of recent energy crises; (iii) improve TANESCO‘s cash flow to reduce fiscal drain on federal budget by increasing tariffs to meet long run marginal cost, securing a commercial loan, and optimizing the mix between hydro and thermal plants; and (iv) improve the monitoring of service delivery and financial performance of TANESCO by the Electricity and Water Utilities Regulatory Authority, ensuring competitive procurement (World Bank 2007b). In this respect, the PRSC included three relatively vague triggers in the energy sector, which 110 ANNEX D were focused on achieving financial recovery for TANESCO (Table D.5). All three triggers and associated prior actions were considered completed. Table D.5. Triggers and Prior Actions on Energy Issues Trigger Prior Action Status PRSC-6 Progress in the implementation of the Progress in the implementation of the Completed TANESCO financial recovery plan, TANESCO financial recovery plan, notably in increasing revenues, and notably in increasing revenues, and strengthening governance of the sector strengthening governance of the sector by improved regulation by improved regulation PRSC-7 Further progress in implementation of Further progress in implementation of Completed the TANESCO financial recovery plan the TANESCO financial recovery plan by the government, taking all to strengthen TANESCO‘s revenue necessary actions regarding any base additional measures that may be required to strengthen TANESCO‘s revenue base, as appropriate by October 2008. PRSC-8 Strengthening TANESCO‘s revenue An energy tariff was successfully Completed collection to meet full operational cost submitted by TANESCO to the recovery in 2010 regulator to allow for revenue collection to meet full operational cost recovery in 2011, and full cost recovery by 2013. Source: Program document 6-8 (World Bank, 2008b, 2009a, and 2010c). 9. One of the two targets for electricity was largely achieved, but the other indicator was inappropriate and evidence suggests the goal was not achieved. During the period, Tanzania was able to expand generating capacity and reduce reliance on hydro power. Capacity increased from 980 MW in 2007 to 1072 MW in 2010 (Figure 13). The Bank team reported that installed capacity reached 1,309 MW by March 2012, the difference a result of new (and costly) emergency power supply. The growth through 2010 fell just short of the PRSC target of 1091 MW in 2010 and significantly short of the MKUKUTA target of 1278 MW for 2010, although production levels increased. The increase in capacity through 2010 was largely due to a Bank-financed project that introduced natural gas power plants operated by independent power producers, which helped to reduce TANESCO‘s reliance on hydropower from 57 percent to 52 percent, as shown in Figure 4. But private investment fell far below expectations. Since 2010, TANESCO has been renting high-cost emergency plants, with capacity of an additional 262 MW in 2012. Tariffs that can be charged by TANESCO are far below costs for renting this additional capacity. The government has a plan for further expansion, but development partners offered serious doubts that these investments would be financed. Further, the increases in capacity and production notwithstanding, the quality and reliability of electricity supply remained poor.66 66 A 2010 World Bank report noted that at about 22 MW per million people, installed capacity per capita in Tanzania is roughly on par with other low income countries whose average is roughly 24 MW per million (Africa Infrastructure Country Diagnostic 2010). But per capita electricity 111 ANNEX D 10. The Implementation Completion and Results Report indicated that electricity capacity utilization improved from 62 percent to 85 percent, but this is not an appropriate measure of the availability of the power supply. Power outages remain a fact of life in Tanzania. While there may have been modest improvement in recent years in the duration of outages according to TANESCO data, their frequency has changed little. A 2010 World Bank report estimated the burden of power outages on the economy as high as 4 percent of GDP, noting that ―Excessive levels of outages result in elevated demand for emergency power and a high level of own generation by firms. The latter pushes the costs of production up; complicates survival of businesses, especially small ones; and makes the economy less competitive. Emergency power supply, another major problem for development, poses a serious fiscal burden that amounts to 0.95 percent of GDP‖ (Shkaratan 2010: 8). A World Bank study on self-generation in Sub-Saharan Africa reported that, according to World Bank Enterprise Surveys, there are roughly 63 days with power outages per year in Tanzania, more than all but three other countries in the region, and as a result 59 percent of firms own their own generator (Foster and Steinbuks 2009). And a recent survey by the Confederation of Tanzania Industries found that on average over 2008-2010 ―24 working-hours of labour (all employees) and all other respective employed factors of production are lost each month owing to electricity supply interruptions‖ (CTI 2011: 18). That report also cites concerns among manufacturers about significant fluctuations in the voltage level. Figure D.1. Non-emergency Generating Capacity- 2007 and 2010. Source: TANESCO Annual Reports 11. Efforts to improve the autonomy of TANESCO yielded limited results. Stakeholders perceive TANESCO‘s commercial focus to have improved substantially, but not yet close to a private commercial entity. Investment in pre-payment systems improved collection but transmission and distribution losses remained high, rising during the period from 21.0 percent in 2007 to 23.3 percent in 2010. As such, operational cost recovery was not achieved consumption in Tanzania falls far below comparator countries, suggesting leakage and unreliability of power plants. 112 ANNEX D by 2010 as targeted by PRSC-8, but claims to the national budget were reduced considerably (TANESCO, Annual Reports 2005-2010).67 The establishment of the Energy and Water Utility Regulatory Authority has led to successive increases in electricity tariffs, bringing them closer in line with costs. The Authority is seen by stakeholders to have relatively strong capacity to monitor the industry, and was named best regulator in 2010 in a study commissioned by the Africa Forum for Utility Regulators. But the Authority was unable to fully enforce rules for competitive procurement and reported that TANESCO‘s cost of providing power was higher than it would have been without sole source selection. 12. The PRSC contributed to these outcomes to a limited extent. The focus of the dialogue was the financial recovery of TANESCO and improving its management. While TANESCO is not fully self-sufficient, it has made significant strides by winning approval to increase tariffs. Improved management has likely contributed to a reduction in the duration of outages, but expansion in capacity has been driven by the availability of government and development partner funds. 13. Policy support under the PRSC for the transport sector was intended to contribute to the objective of promoting sustainable and broad-based growth by improving the condition of the road network, and later, improvement in port efficiency. The triggers and prior actions that formed the basis of the PRSC dialogue are outlined in Table D.6. Four of the six triggers were fully completed within the stipulated timeframe and two others were delayed. Among those triggers converted to prior actions, the trigger for ports—to prepare a time-bound action plan—was diluted considerably, and the prior action was reworded to preparing a preliminary version of a time-bound action plan. 14. Because the two of the six triggers were essentially repeated after delays, prior actions covered four areas: enactment of the Roads Act; development of a Transport Sector Investment Plan (TSIP) with adequate provision for maintenance; adoption of a framework for public-private-partnership (PPP); and actions to reduce port congestion. The Roads Act, the TSIP, the framework for PPPs, and an action plan for the Tanzania Ports Authority were intended to strengthen the institutional and policy framework for the transport sector, and prioritize investment and maintenance spending. The reduction of port congestion was intended to reduce transport costs for the private sector directly. 67 While the drain on the budget has been reduced, hidden costs remain. For instance, a private investor was awarded $65 million in an arbitration case in 2010, and a corruption case is currently under court review, but neither the award nor a contingent liability appear on TANESCO‘s balance sheet, suggesting that the government will be covering relevant costs. 113 ANNEX D Table D.6. Triggers and Prior Action for Transportation Trigger Prior Action Status/status PRSC- 5 Roads Bill submitted to Parliament Preparation of draft Roads Bill, Completed. Prior including a road inventory and Action reflects Cabinet approval of the delays experienced submission of a new Roads Bill in submitting Roads to Parliament Bill to Parliament. Submission is now expected in April 2007. PRSC- 6 Roads Bill submitted to Parliament Roads Bill enacted Completed. Transport Sector Investment Plan Transport Sector Investment finalized with adequate provision Plan finalized with adequate PRSC-6 Completed. for maintenance and a framework provision for maintenance and a for PPP. framework for PPP. National PPP policy, including Partially Framework for PPP, including in the infrastructure sector, completed; the entire Transport Sector, PRSC- 7 prepared, discussed with Approval by finalized and approved by October stakeholders, and being readied Cabinet 2008. for Cabinet approval. forthcoming. PRSC- 8 Government improves port Government improved port Completed. congestion situation by three key congestion situation by three measures: (i) Tanzania Ports key measures: (i) Tanzania Ports Authority presents a time-bound Authority prepared a action plan for moving to Landlord preliminary version of a time- Status; (ii) Tanzania Ports Authority bound action plan for moving to to submit a solid proposal to Landlord Status; (ii) Based on SUMATRA to approve increase in Tanzania Ports Authority port storage tariffs and reduction in application, SUMATRA free time; and (iii) government approved increase in port establishes own area for impounded storage tariffs and reduction in containers. free time; and (iii) Government established own area for impounded containers. PRSC- 8 Draft bill and regulatory framework PPP bill approved by Completed. for Private-Public Partnership Parliament. submitted to Parliament. Source: Program document 5-8 (World Bank, 2007b, 2008b, 2009a, and 2010c). 15. The effort to strengthen the institutional and policy framework for the transport sector was spearheaded by the passage of the Roads Act in April 2007. It addressed three key issues: (i) replacing the out-dated 1932 ordinance and recognizing the existence of Tanzania National Roads Agency (TANROADS), the executive agency charged with developing and maintaining the national and regional roads network, as well as the Roads Fund intended for road maintenance; (ii) classifying public roads utilizing a Bank-financed inventory; and (iii) specifying the roles and responsibilities of relevant government ministries and authorities, including their clarifying their governance. 114 ANNEX D 16. The Ports Authority, which was to prepare an action plan to transform itself from an operator of the ports to a landlord in line with the Port Act of 2004, only managed a preliminary action plan while the actual transition to landlord status, beyond existing arrangements whereby some container terminals are already operated by a private party, was not expected to change until 2018, according to government officials. 17. To promote private investment in the transport sector, the Bank and government agreed that a strong institutional and policy framework for investment was required given the poor track record of PPPs in Tanzania. Initially, a framework for PPPs in the transport sector was to be developed as part of the TSIP (a trigger for PRSC-7), but the government instead decided to expand the policy beyond the transport sector, and so the TSIP included a ―roadmap‖, which outlined the timing in which a framework would be prepared and approved. As of May 2009, the Bank considered the proposed PPP framework as putting Tanzania ―on par with some of the best institutional and regulatory regimes in developing countries,‖ with a central PPP agency under the Ministry of Finance and Economic Affairs responsible for PPP promotion, packaging, negotiations, and vetting of fiscal impacts. But the final bill, which was ultimately approved by Parliament in July 2010, was considered by the Bank to be only on par with policies in other countries because of its lack of clarity on the division of government responsibility among the various implementing agencies (the Ministry of Finance and Economic Affairs, the Attorney General and the ministry responsible for investment in the relevant sector). 18. To prioritize investment and maintenance spending, the TSIP was prepared in 2006 as agree under PRSC-6. Prior to the introduction of the PRSC trigger, there was a doubling of the fuel levy to TSh 200 per liter and a concomitant increase in the size of the Roads Fund from TSh 101 billion in 2006/07 to TSh 208 billion in 2007/08. Because the Roads Fund finances road maintenance, the coverage of routine maintenance needs rose that year to over 115 percent, and the coverage of periodic maintenance needs stood at 54 percent, according to internal Roads Fund figures. Since the changes occurred prior to the introduction of the trigger/prior action, the PRSC trigger appears to have been meant to institutionalize the prioritization of maintenance. But the coverage of routine maintenance subsequently fell to 82 percent in 2008/09, 70 percent in 2009/10, and 64 percent in 2010/11, and the coverage of periodic maintenance fluctuated from 52 percent in 2008/09, to 64 percent in 2009/10, to 58 percent in 2010/11 (Road Fund internal sources). Nevertheless, the TSIP did make attempts to align itself with the actual budget allocations, but indicated that only 40 percent of the estimated cost of implementing the program had been committed by the government and development partners. The program expected the private sector to fill most of the gap, which did not happen. The government produced a transport sector medium-term expenditure framework and several rolling short-term TSIPs to try to match the required investment in the program with the available budget, but these were unsuccessful. Indeed, in 2009/10, roads contracts were signed by government agencies that amounted to twice the allocated funding, a problem that cut into the available funding for other development projects and rolled over into subsequent years and led to the suspension of the signing of any new contracts. 19. Complementing the efforts to strengthen the institutional framework and improve investment spending priorities, the Ports Authority was expected to reduce port congestion through several measures. An increase in port storage tariffs and a reduction in free time 115 ANNEX D were granted by the Surface and Maritime Regulatory Authority (SUMATRA), and the government established inland container depots to store containers outside the port area. These measures helped reduce some congestion, which may have also been affected by the global economic slowdown, but Bank and IFC staff and informants from the private sector still views delays in port clearance of cargo as a significant obstacle for investment. A constraining factor is that port clearance requires approval by many players in the government. Aside from logistics handled by Tanzania Ports Authority, cargo must be cleared by customs, the Tanzania Revenue Authority, the Government Chemist Laboratory Agency, the Tanzania Food and Drugs Authority, and like institutions. The government has created a single window system where all relevant institutions have an office at the port. But a number of agencies have not decentralized authority and require clearance from headquarters, generating additional delays. A 2009 study by the Revenue Authority concluded that processes for document submission needed streamlining; that many procedures still need to be automated and harmonized across agencies; and that infrastructure including access roads and cargo handling equipment remain insufficient (Tanzania Revenue Authority 2009). 20. Overall, the government‘s program in the transport sector was unable to achieve the key targets it set out for itself. Improvement in the condition of national and regional roads was negligible; more recent data indicate some improvement in the past year, although still well below the target. The shortcomings were likely accounted for by the shortfall in maintenance financing—after a burst in resources made available in 2007/08, maintenance funds covered less than two-thirds of routine maintenance and less than 60 percent of periodic maintenance—although severe weather and the apparent reclassification of district roads as national and regional roads may have contributed. There was real improvement in the efficiency of the Dar es Salaam port, with container dwell time falling from its peak in 2008 even as traffic increased. The PRSC likely contributed modestly to this improvement in performance by encouraging an increased tariff and alternative means for storage, but longer term issues regarding port efficiency do not appear likely to be addressed given the lack of strategic consensus between the Bank and the Ports Authority. 21. The third outcome area for the objective of growth and reduction of income poverty—improved livelihood in rural areas—covers the program‘s contributions to agriculture and rural development, natural resources, and transport sectors. Indicators for this outcome are summarized in Table D.7. 116 ANNEX D Table D.7. Indicators of progress on improved livelihood in rural areas Indicator Baseline Target 2010 (ICR) 2010 (IEG) Source Result 10% Agriculture sector 4.4% (MKUKUTA)/ 3.9% Not met growth (3-year average) 6% (PRSC) Revenues received from 25% (2008) Not met concession and license 4-15% N/A at 40% (2009) (data not in forestry, as % of their (2007) completion 50% (2010) available) estimated value Number of districts qualifying for top up 84 132 102 133 (2011) Bank team Met grants a/ % of passable rural 75 59 (2009/10, roads (in good and fair 50 (MKUKUTA)/ 56 Roads Fund Not met district roads) condition) 70 (PRSC) 8.1 to Number of customers 10.1 to 14.2%/ 11.3%/ 20% connected to national 849,000 Probably 586,000 (MKUKUTA)/ 14.5% TANESCO grid and off-grid households met households 12.9% (PRSC) sources a/ (2010) (2005) a/ The number of Local Government Authorities increased from 132 to 133 during the period. b/ Calculation of electricity access depends heavily on assumptions of number of persons per household. The range given is based on assumptions of 5-7 people per household. 22. The agricultural sector was expected to grow in excess of 5 percent per annum in order to make possible sustained economy-wide GDP growth of 6-8 percent, and help to reduce income poverty. The government‘s strategy to facilitate agricultural growth was a continuation of the policies in the previous three years in the context of the agricultural sector development strategy, which included (i) reform of crop boards; (ii) reform of local government taxation; and (iii) improvements in rural infrastructure. Other important areas of policy included reform related to improving the security of land rights to increase on-farm investment. Following amendments to the Land Acts (under PRSC 1-3), the government initiated a strategic plan for its implementation, which included land surveys and titling of land. 23. PRSC 4-8 supported the agricultural drive of MKUKUTA through the complementary IDA supported Agricultural Sector Development Project, approved on June 15, 2006, with a closing expected on June 30, 2013. The series focused on residual issues related to uncompleted reforms on the crop boards and land registration from the previous PRSC1-3 series. In addition, issues related to ―nuisance‖ local taxes on agriculture were also highlighted by the close of the series (Table D.8). 24. The trigger and prior actions related to crop boards under PRSC 4-6 were met against significant political pressure, although somewhat diluted. The important step of halting the price-setting function of crop boards was achieved with the issuance of a circular abolishing the use of levies by crop boards and requiring them to limit their functions to regulatory activities. On the other hand, the trigger under PRSC-5 requiring amendments of crop board legislation for at least two crop boards was not submitted to Parliament. Instead the government decided to conduct consultations as preparatory work to reform all four crop boards simultaneously, a more ambitious agenda that would take longer and to sign a memorandum of understanding, which became the binding prior action. In the end, only draft 117 ANNEX D amendments to crop board legislation, which was submitted for stakeholder consultations, were achieved. Follow-up strategic plans for all crop boards have not yet been finalized. Table D.8. Triggers and Prior Actions for Agriculture and Rural Development Trigger Prior Action Status/Comment PRSC-4 Progress in the implementation Progress in the implementation of the Completed. A of the Action Plan for the Action Plan for the Rationalization of circular abolishing Rationalization of Roles, Roles, Functions, Financing and levies by Crop Functions, Financing and Accountability of Crop Boards, consistent Boards issued. Crop Accountability of Crop Boards, with the Agriculture Sector Development Boards limited to consistent with the Agriculture Strategy, including issuing of ministerial regulatory Sector Development Strategy. circular for Crop Boards to cease charging functions. levies/cess. Progress in the implementation Progress in the implementation of the Completed, but it is of the strategic plan for strategic plan for operationalizing the not clear the target operationalizing the Land Act Land Act and Village Land Act, including for CCROs was and Village Land Act. issuing of 400 certificates of customary met. rights of occupancy (CCROs) during first half of FY06. PRSC-5 Amendments of Crop Board Consultations for four crop Boards Partially completed Legislation for at least two completed and MOUs signed as PA was modified. Crop Boards submitted to Parliament PRSC-6 Draft amendments to crop Draft amendments to crop board Diluted policy board legislation submitted for legislation submitted for stakeholder stance completed stakeholder consultations and consultations and strategic plans for all strategic plans for all crop crop boards finalized. boards finalized. PRSC-8 Update Facility for Investment Trigger dropped Climate Advisory Services because the study paying particular updating of Facility attention to regulations and for Investment local taxes affecting agriculture Climate Advisory and undertake consultations Services study was and agree to an action plan not completed with based on the findings and action plan and recommendations of the study. recommendation. PRSC-8 Progress in completion of land Progress in completion of land registry. Partially Completed registry. Source: Program document 4-8 (World Bank, 2006, 2007b, 2008b, 2009a, and 2010c). 25. The PRSC 4-8 series also supported the government‘s initiative to scale up decentralized land registration both in urban and rural areas. Some progress was realized with about 7,000 villages surveyed and registered out of about 11,000, but the issuance of certificates of customary rights of occupancy (CCROs) for individual customary landowners was delayed. In addition, the development of efficient land registries has been lagging, with limited progress in modernizing the central and zonal land registries, and limited capacity of local government authorities to develop land registry systems that match their responsibilities under the law. With limited access to titling and registration, management of land has remained problematic on the ground, with many land disputes and lack of transparency in land management practices, as well as lack of an instrument to track women‘s ownership of 118 ANNEX D land. The limitations are in part due to the inadequate implementation of the land laws, particularly titling and registration (PRSC-8) which has been fragmented through a series of small and somewhat disjointed pilots. As a consequence, it has received insufficient resources and little policy attention. The Strategic Plan for the Implementation of Land Legislation was finalized in 2005 but its implementation has been slow. To date, participatory village land-use plans have been prepared in 1,000 villages out of more than 11,000 villages that exist in mainland Tanzania. The National Land Use Framework Plan has been drafted and is in the process of being finalized for the Cabinet‘s approval. Implementing the Land Use Planning Act 2007, the Framework Plan is an indicative, basic, general plan that facilitates rational allocation of land resources, decision-making, and management at the national level while ensuring their sustainability. 26. The trigger introduced in PRSC-8 to begin tackling the issue of ―nuisance‖ taxes was dropped because the updating of the Facility for Investment Climate Advisory Services study was not completed. In general, government commitment with respect to the policy dialogue in agriculture decline appreciably under PRSC 4-8. It could in part be due to the political sensitivity of the nature of the dialogue and the impending presidential and parliamentary elections for the fall of 2010. 27. In terms of outcome, it is estimated that growth in agriculture averaged 4 percent per year during 2006-10, short of the PRSC program target of 6 percent and far short of the MKUKUTA target of 10 percent. Even this modest outcome is questionable as some doubt has been expressed about the integrity of the data on agricultural growth. Taking the data as accurate, agricultural growth is still only marginally faster than that of population growth, and appears to be, as under PRSCs 1-3, derived mainly from continued expansion of cultivated areas rather than increases in yields. Similarly, the impact of policy successes such as progress on land registration, and reform of the crop boards towards regulation and away from marketing are too early to yield perceptible outcomes in terms of improved access to credit and reduction of ―nuisance taxes.‖ As such, much of the outcomes under agriculture cannot be attributed to the PRSCs. 28. Natural resource issues were specially highlighted in PRSC-7 and -8, in response to the findings of the analytic work on growth contained in the Bank‘s Country Economic Memorandum of 2007. The natural resource sector, defined to include tourism, mining, forestry, and fisheries, was considered to be a significant income supplement to many Tanzanians, although much of it operated in the informal economy. The sector was viewed as a growing source of income, generating exports from traditional activities in forestry, fisheries, mining, and tourism. A key issue that needed attention was the failure of public oversight in the form of adequate regulatory framework and enforcement leading to production and marketing practices for these resources that undermined sustainable growth in the sector. The focus of the PRSC in the area of natural resources was thus ―to build a policy agenda to bring better transparency, to achieve poverty education and environmental sustainability‖ (World Bank, 2009a). 119 ANNEX D Table D.9. Triggers and Prior Actions for Natural Resource Sector Trigger Prior Action Status/Comment PRSC-7 Collect information and report on Information on transparency and Partially completed transparent and accountable systems accountability of as there was no of licenses/concession allocations licenses/concession allocations information available and how this meets market values in systems collected in the on market values. the Forestry, Fisheries, Wildlife, forestry, fisheries, wildlife, Prior action was Minerals and Oil & Gas sectors. minerals and oil & gas sectors. modified accordingly. PRSC-8 Approve and initiate Action plan to improve Completed, action implementation of an action plan to accountability and transparency plan approved for improve accountability and in allocating natural resource implementation. transparency in allocating natural licenses and concessions resource licenses and concessions approved and implementation initiated. Source: Program document 7-8 (World Bank, 2009a, and 2010c). 29. The initial trigger and prior action (Table D.9) had the objective of setting the stage for reform of the licensing and concession regimes in natural resource management. Meeting this conditionality was hindered by data availability and therefore the prior action had to be watered down. The follow-up trigger in PRSC 8 aimed at enhancing reform of the licensing and concession regimes in natural resources management. This was completed satisfactorily as the action plan to improve accountability and transparency in allocating natural resource licenses and concessions was approved and implementation initiated. 30. Overall, the objective to improve the natural resource institutional environment, especially in terms of transparency, to allow for a more sustainable and profitable exploitation of the resources for local communities was substantially attained and the PRSC played a substantial role in this effort.68 Using as a performance indicator, the level of revenues received from concession and license in forestry as percentage of their estimated values, the rise from 15 percent in 2007 to 50 percent by 2010 points to gains in the new framework. In addition, concerns over the mining sector, including the level of revenue generated by the sector and captured by the government (through royalties, taxes, and so on), prompted the government to issue a new Mining Act (the United Republic of Tanzania, 2009). On February 16, 2009, with Bank support, Tanzania joined the Extractive Industries Transparency Initiative as a Candidate Country and is currently at the validation stage. Following a request for extension, Tanzania has until February 15, 2013, to undertake validation, in which its compliance with the Initiative‘s principles and standards would be assessed. Successful validation pave the way for a more transparent framework in the sector, a situation widely supported by operators in the sector. 68 Natural resource functions are carried out by three agencies of the government: the Ministry of Natural Resources and Tourism (forestry and tourism), Ministry of Livestock Development and Fisheries (fisheries), and the Ministry of Energy and Minerals (mining). 120 ANNEX D Objective 2: Improvements in the quality of life and social well-being. 31. The three outcome areas serve roughly as a further articulation of the objective, and so progress in these areas would lead directly to achieving the objective. But the PRSC‘s role in contributing to progress was modest. It played a potentially important but supportive role, giving extra incentive to have a productive sector-level dialogue. 32. In total, four out of seven targets were met, and substantial progress was made toward reaching two others. The country met two of the three targets with regard to improving the health status of the population, on child vaccinations and HIV/AIDS prevalence, while limited progress was made on the third indicator related to births attended by skilled health workers. Similarly, two of three targets were met in the outcome area of education. The primary completion and tertiary education enrollment targets were met, and the primary enrollment target nearly so, but progress in improving the quality of education was slow and the program fell short of the target. In improving access to clean water, there was significant progress, but the target was missed. 33. The first outcome area covers the program‘s contributions to the health sector to improve the health status of the population. Indicators for this outcome are summarized in Table D.10, which shows that two of the three targets were met. Table D.10. Indicators for progress in improving the health status of the population. Indicator Baseline Target Latest Latest Source Result (ICR) (IEG) % of children <2 years receiving three doses of 71% 85% 88% 88% DHS, 2010 Met vaccines GBS % of births attended by a 41% 80% (MKUKUTA) 51% Annual 63 % Not met skilled health worker (2006) / 65% (PRSC) 2010 Review 2012 National HIV prevalence in 4% 2.4% THMIS, 4% 2.4% Met the 15 to 24 years age group (2003) (2007/08) 2008 Note: DHS – Demographic and Health Survey, National Bureau of Statistics; THMIS – Tanzania HIV/AIDS and Malaria Indicator Survey, National Bureau of Statistics 34. The triggers and prior actions for health in the second PRSC series focused on the annual sector reviews, with greater specificity in defining a satisfactory review toward the end of the series, as shown in Table D.11. The documentation of the PRSCs and of GBS generally made clear that implementation of the PRSCs in the health sector was expected to be a continuation of the process that had been underway in the health sector since 1999. This had both resulted in improved performance in the sector and in a sector dialogue, which was recognized as best approximating the ideals of donor harmonization under the Paris Declaration. The assessment of the annual sector reviews under both PRSC-4 and 5 were mixed pointing out both achievements and shortcomings in the sector; but overall the process was regarded as satisfactory with adequate progress made. PRSC-6, the first under a new CD and TTL, seemed to take a more critical tack. While acknowledging the continued achievements of the health sector, including the overall positive assessment of the just published Joint External Evaluation for 2000-2006, the evaluation stressed the continuing 121 ANNEX D shortfall in the share of the budget devoted to health, difficulties in deploying health care workers to underserved areas, delays in cash releases to spending units, especially local government authorities, poor infrastructure maintenance, performance data gaps, and weak management. On this basis it rated the annual sector review as ―fairly satisfactory.‖ This resulted in the trigger being regarded as only partially completed, although these problems had been recognized in earlier annual reviews. Table D.11. Triggers and Prior Actions for the Health Sector Trigger Prior Action Status PRSC-4 Satisfactory health sector review. Completed PRSC-5 Satisfactory health sector review carried out Satisfactory health sector review Completed carried out PRSC-6 Satisfactory health sector review Fairly satisfactory health sector Partially review Completed PRSC-7 Updating the 2006/07 Health Sector Health Sector Performance Report Completed Performance for 2006/07 Report by January 31, 2008 and Health updated and Health Sector Sector Performance Report for 2007/08 Performance Report produced and disseminated by October 2008. for 2007/08 produced and disseminated. PRSC-8 Satisfactory health sector review (based on Satisfactory health sector review Completed an assessment of progress on agreed outcome carried out targets, adequacy of resource allocation, achievement of sector milestones, and adequacy of stakeholder consultations) Source: Program documents for PRSCs 4-8 (World Bank, 2006, 2007b, 2008b, 2009a, and 2010c). 35. PRSC-6 and 7 had been intended to mark the transition from the health sector basket fund to general budget support funding for the health sector. Bank staff clarified that this transition would be conditional on commensurate increases in the allocation of resources to the health sector by the Ministry of Finance, and on progressive improvements in procurement systems and the satisfactory implementation of a health care waste management plan. Critically, health budget allocations continued a steady negative trend (in terms of share of total expenditure minus consolidated fund services in order to exclude non- discretionary overheads such as debt service) begun in 2005 when the peak of 14 percent had been reached. As a result the Ministry of Health opposed shifting funds from the health sector basket fund to general budget support and reportedly told the development partners that those who did not participate in direct health financing (such as the basket fund) would not have a seat at the table with respect to sector dialogue. This was in sharp contrast to the position of a few years earlier in which the Ministry of Health appeared ready to de-link policy dialogue from direct sector financing. In an extreme example the MOH cited its experience when DFID had abruptly shifted its entire health financing to general budget support. The Ministry of Health claimed that it received no compensatory allocation from the Ministry of Finance and consequently lost about USD 70 million from its annual budget. While a more gradual transition with strengthened dialogue between the Ministry of Finance and the Ministry of Health should have mitigated such a negative outcome for the Ministry of Health, it seems that the general momentum toward general budget support away from sector pooling had also begun to dissipate on the part of many development partners including the Bank. In 2009, the Bank prepared a second installment of additional financing for the second 122 ANNEX D Health Sector Development Program II clearly signaling that the PRSC was no longer expected to take over for sector financing. 36. Over the PRSC 4-8 period, health outcomes continued to improve. In spite of the declining budget share, public expenditure per-capita increased (due to the substantial increase in overall public expenditure) and was anticipated to reach USD 14 per person by 2010/11. The 2010 Demographic and Health Survey documents continued marked declines in infant and under-five mortality (of 30 percent and 22 percent respectively), putting Tanzania on track to achieve its Millennium Development Goals in these areas. Targeted vaccination rates and AIDS prevalence among 15-24 year olds which were the specific results listed in the PRSC4-8 results matrix had already been achieved by 2008. Child nutrition indicators also continued to improve and there was some evidence of a beginning decline in maternal mortality (although large margins of estimation error preclude a definitive trend). Significant disparities across districts remain and, as highlighted by a review of the General Auditor‘s Office, little progress has been made in matching workload to staff deployment. Little progress is evident on other issues raised by a 2007 joint evaluation of the health sector in Tanzania (COWI and others 2007), including, for example, poor hospital management, and a weak Public-Private Partnership framework. 37. The second outcome area covers the program‘s contributions to the education sector to expand enrollment and enhance quality in all levels of education. Indicators for this outcome are summarized in Table D.12. Table D.12. Indicators for progress in expanding access to education and enhancing quality. Indicator Baseline Target Latest (ICR) Latest (IEG) Source Result Net primary enrollment All 73% All 99% All 95.4% All 94% Nearly met, ESDP, Boys 71% Boys 99% Boys NA Boys 93.7% but decreasing 2011 Girls 75% Girls 99% Girls NA Girls 94.2% trend Percentage of cohort Met in 2010, ESDP, completing primary 66.9% 69% 69.3% 64.4% but decreasing 2011 education trend in 2011 Qualified teacher / pupil ESDP, ratio in primary 1:73 1:45 1:54 1:48 Nearly met 2011 education Tertiary education 0.27% ESDP, 6% 5.3% 8.0% Met enrollment ratio (2002) 2011 Note: ESDP – Education Sector Development Program 2011, Education Sector Development Committee, government of Tanzania 38. Bank Support for education through the PRSCs was underpinned by the first ever Annual Sector Review in the sector prepared by the government with inputs from technical working groups, including civil society and development partners. The series included specific triggers in the education sector, which focused on the completion of a satisfactory annual sector reviews dealing with performance issues in the sector (Table D.13). 123 ANNEX D Table D.13. Triggers and Prior Actions on Education Trigger Prior Action Status PRSC-5 Satisfactory education sector review Satisfactory education sector Completed carried out review carried out PRSC-6 Satisfactory education sector review Fairly satisfactory education Partially sector review completed PRSC-7 By October 2008, agree on education Education sector performance Completed sector performance report on the basis report for 2007/08 prepared on of an agreed sector monitoring tool. the basis of an agreed sector monitoring tool. PRSC-8 Satisfactory education sector review Satisfactory education sector Completed (based on an assessment of progress on review carried out agreed outcome targets, adequacy of resource allocation, achievement of sector milestones, and adequacy of stakeholder consultations) Source: Program documents for PRSCs 5-8 (World Bank, 2007b, 2008b, 2009a, and 2010c). 39. PRSCs 5-8 focused on strengthening of the sector policy dialogue in education. The process of annual sector reviews had only just begun in education, in contrast to the health sector in which annual reviews were well established; but the performance of the Education Sector Development Program (ESDP) and especially the Primary Education Development Program (PEDP) had been widely regarded as successful (including by IEG in its Project Performance Assessment Report of Bank support to education in Tanzania). As a result education was regarded as one of the sectors most likely to mark the transition to GBS as the primary aid modality. The satisfactory rating for the first education sector review in PRSC-5 seemed largely based on the successful initiation of the annual review and not on the substance of the review. Recognizing the need to improve the annual review process, Bank staff presented a proposed set of criteria to define a satisfactory sector review as part of the PRSC-5 documentation.69 The education sector review carried out as a prior action for PRSC-6, although more substantive than the previous year, fell short of these criteria in a number of aspects, primarily related to an inadequate integration of progress on results with reporting on outputs and financial allocations. The rushed timing and limited consultative process may have also played a role. As a result, the PRSC-6 prior action was revised to a ―fairly satisfactory sector review.‖ This downgrading was across the board for similar triggers in the health and water sectors. While no specific reductions were made from the credit amount related to the diminished actions, the size of the PRSC-6 was USD30million less than PRSC-5 marking a clear reversal from the previous upward trend in PRSC contributions to GBS. 69 These included: :( i) timely inputs to the budget process;(ii) involvement of a wide range of stakeholders; (iii) assessment of performance against agreed benchmarks and targets;( iv) use of reliable information stemming from sector MISs, Government of Tanzania statistics or MDA implemented projects; and (v) a review of both retrospective and prospective actions. The definition of ―reliable information‖ subsequently proved problematic as it could be interpreted to exclude third party monitoring and data collection. 124 ANNEX D 40. By the mid-point in the series (PRSC6), the initial intention to gradually shift the bulk of the Bank‘s education support from investment operations to the PRSC had changed and preparation for a second secondary education development program as a sector adjustable program loan (APL) supporting specific measures to address problems arising from the spectacular expansion of secondary enrollment was initiated. The triggers for PRSC-7 and PRSC-8 progressively made more explicit the requirement for the annual sector review to focus on performance and on the linkage between results and sector funding and policies. The quality of the sector reviews did improve and were deemed satisfactory by the development partners. However other developments made the determination of sector review quality more difficult. Following the ―fairly satisfactory‖ determination in 2008, the government clarified that it would only respond to a binary determination – satisfactory or not satisfactory. In addition the development partner sector working group developed a weighting system which gave equal weights to outcomes, resources and dialogue in determining the overall rating for the annual performance reviews. In the view of Bank staff and some development partners, the combination of these factors reduced the leverage of the annual sector review on GBS and the PRSCs by narrowing the assessment range. 41. For the sector as a whole, the remarkable progress in increasing access to both primary and secondary education leveled off during the second PRSC series. Primary net enrolment ratios peaked at 97.2 percent in 2007/08, falling back to 94 percent in 2010/11. Thus, the MKUKUTA target of 100 percent NER by 2010 was not achieved. However, the MKUKUTA target of a 50 percent transition rate to secondary school was exceeded during all but one year of this period. In addition, the number of qualified primary school teachers increased and there was a slight improvement in pupil/teacher and pupil/classroom ratios. Independent quality assessments were mixed. Uwezo, a regional NGO which conducted surveys of literacy and numeracy across 38 randomly selected districts in Tanzania found severe shortcomings in the competencies of primary school leavers: results which suggested that Tanzania learning achievements were well below those in Uganda and Kenya. However a regional initiative supported by 16 Ministries of Education, the Southern and Eastern Africa Consortium for Measuring Educational Quality (SACMEQ) which conducted surveys in 2003 and 2007 found that Tanzanian students performed well above regional averages in reading and arithmetic in both years and were one of the few countries to have significantly improved performance between the two surveys. The SACMEQ and Uwezo assessments are not comparable. Uwezo had a much larger sample and included household characteristics; SACMEQ used more traditional school-based testing which may have been biased in the selection of students. Analytic work undertaken for the 2010 PER found that education developments in Tanzania represented good value for money overall; but that there was still scope to improve value for money significantly by shifting resources to under-served districts. Major differences in the efficiency of resources use across districts also suggested that district efficiency be factored in to district resource allocations. 42. The secondary education net enrollment ratio continued to rise but at a slower rate than earlier in the decade growing from 24.4 percent in 2007 to 34.5 percent in 2010. Even this slowing growth continued to put great strains on the overall education system. Independent evidence from the NGO Twaweza suggested that the capitation grant of TSh 25,000, which was intended to contribute to learning materials and other non-salary support to secondary schools, was actually being paid out at a small fraction of the mandated amount. 125 ANNEX D (The same shortfalls were found in the payout of the capitation grant to primary schools.) After an initial deterioration in the ratio of secondary school pupils to qualified teacher, the ratio did improve slightly toward the end of the period. Overall it appeared that aggregate resources for secondary education were being squeezed between primary and higher education student loans in addition to being spread across a much larger cohort of students. The consequence was seen in a plummeting Form IV pass rate down from 26.7 percent in 2007 to 11.4 percent in 2009. In this context the Bank‘s focus in SEDPII for tackling the problems that emerged with the rapid expansion of secondary enrollment seemed highly appropriate. The MKUKUTA goal for higher education was met as the gross enrollment ratio jumped from 0.27 percent in 2005/06 to 8 percent in 2010/11. However this remarkable growth threatens the fiscal sustainability of the whole sector through the commensurate growth of a student loan scheme which lacks adequate repayment mechanisms and is inherently regressive. Overall, there were significant achievements in maintaining high levels of enrollment at the primary and secondary level, and mixed evidence on quality. 43. Actions in the water sector were intended to contribute to the objective of improving quality of life and social well-being through increased access to clean, affordable and safe water and sanitation. With the exception of PRSC-6 when the water sector review was considered fairly satisfactory as in health and education, the triggers and prior actions were all met (Table D.14). Initially, the focus on jump-starting the annual sector review process was useful, particularly when standards were introduced for what constitutes a ―satisfactory‖ sector review. But over time, annual review meetings became excessively focused on the rating rather than a discussion of policy actions critical to improve development outcomes. Ultimately, all parties faced strong incentives to rate programs satisfactory, given their interest in maintaining dedicated funding to the sector. 44. To evaluate performance, the percentage of the population that had access to clean and safe water, distinguished by rural and urban populations, was the only indicator specified. This indicator was, however, subject to two significant limitations. First, access was poorly defined and was sensitive to the assumption made on how many people had access to household connection or water outlet. Second, the data provided from government sources assumed that all connections or outlets were functional, given the absence of a mechanism to identify functionality, but informants suggested there are a significant number of non-functioning water points, particularly in rural areas.70 70 A one-time water point mapping exercise, currently underway, will survey the existence and functionality of rural water points. But given this will only be an end-of-period single observation, household survey data is the most useful source to measure progress. 126 ANNEX D Table D.14. Triggers and Prior Action for Water Sector Trigger Prior Action Status/comment PRSC 5 Satisfactory water sector review carried Satisfactory water sector review Completed out. carried out. PRSC 6 Satisfactory water sector review Fairly satisfactory water sector Partially completed review. PRSC 7 Satisfactory Water Sector Review Satisfactory Water Sector Review Completed carried out. carried out. PRSC 8 Satisfactory water sector review (based Satisfactory water sector review Completed on an assessment of progress on agreed carried out. outcome targets, adequacy of resource allocation, achievement of sector milestones, and adequacy of stakeholder consultations) Source: Program documents for PRSCs 5-8 (World Bank, 2007b, 2008b, 2009a, and 2010c). 45. Despite this limitation, available survey data71 suggested modest improvement in the share of households using protected water sources, from 45 percent 57.8 percent in rural areas and 73 percent to 86 percent in urban areas during 2007 and 2010, as shown in Table D.15. This slightly fell short of the targets envisaged in the PRSC. There was also some improvement by 2010 in the share of the rural population which could access clean and safe water within 30 minutes (arise of nearly 20 percentage points over the baseline for 2007), but this constituted less than half the rural population. Similar analysis for the urban population was subject to data limitations, but there appears to be no improvement in the share of urban population with access of clean and safe water within 30 minutes.72 For sanitation, the use of pour flush toilets rose from 1 percent to 1.6 percent in rural areas and from 6 percent to 15.3 percent in urban areas during 2007 and 2010. The surveys also indicate significant drops in the use of latrines in rural areas, and thus suggesting reduced access to basic sanitation facilities. 46. This link between the satisfactory completion of PRSC triggers and progress made in meeting expected outcomes is weak and very indirect. The PRSCs succeeded in upgrading and providing standards for the dialogue among the government and various development partners in the water sector. This dialogue gave direction to the sector, and contributed to improvements in access to water and sanitation. 71 Refer to the Water Sector Status Report 2011, Household Budget Survey 2007, and the Demographic and Household Survey 2010. 72 Urban figures for access in 2007 do not include Dar es Salaam. Its inclusion would have raised the baseline for 2007and decreased (or even make negative) the percentage change for 2010. 127 ANNEX D Table D.15. Indicators for Improving Well-Being through Better Access to Clean Water Indicator Baseline Target Latest (ICR) Latest (IEG) Source Result Percentage of Rural 45% Rural 57.8% Rural 57.8% population with Rural 65% Urban 73% Urban 86% Urban 86% Not met access to clean and Urban 90% (2007) (2010) (2010) safe water Household Percentage of Budget Survey Rural 40.4% Rural 47.9% households using 2007, Urban 77% Urban 80% Not met protected sources of Demographic (2007) (2010) water and Health Survey 2010 Italicized indicators were not identified in program documents but collected for this evaluation. Objective 3: Governance and accountability. 47. The three outcome areas were used to better define the overall objective, but there was little improvement in any of them. In total, six out of eight targets were met, but several of these were flawed in their construction and focused on outputs rather than outcomes. On enhancing the accountability of the state, the two output targets focused on prosecuting corruption were met, but perception data on corruption cases suggest that the outcome has not been achieved. On improving the effectiveness of public administration, the two flawed output targets were met. But one indicator—resources paid to public workers—did not help to achieve the desired outcome, as enhanced pay was not successfully targeted to positions key to delivering public services. Similarly, the other indicator—share of local government expenditures allocated on a formula basis—provides no information about whether or not those resources were used effectively, and there were no major actions taken to ensure that they were. On improving public financial management, two of the four targets were met, with progress in auditing functions, but deterioration in budget execution and limited improvement in compliance with procurement rules. 48. The prior actions in the series, highlighted in Table D.16, covered 5 aspects of public sector reform: anti-corruption (PRSC 6), public service pay (PRSC 4, 7 and 8), procurement (PRSC 4, 7 and 8), external audit (PRSC 4, 5 and 6), and internal audit (PRSC 7). PFM/budget and decentralization were identified in the objectives although there were no associated triggers or prior actions. 49. The Bank had investment and TA credits covering several areas of public sector reform—Public Finance Management Reform Program, Public Service Reform Program, Tax Administration Reform, FY05 Local Government Support Project (LGSP), FY06 Accountability, Transparency and Integrity Project (ATIP), and a FY08 Performance, Results and Accountability Project (PRAP). The government counterparts for the first two were informed of and involved in the PRSC process. The tax (Tanzania Revenue Authority) and local government counterparts (even at the center) were less involved in the PRSC process, and there were no prior actions linked with them. The Regional and Local Government (RALG) office in the Prime Minister‘s Office was somewhat involved with PRSC discussions, but not local government officials. 128 ANNEX D Table D.16. Triggers and Prior Actions for Public Sector Reform in the PRSCs 4-8 Trigger Prior Action Status PRSC-4 President‘s Office - Public Service Management Organization Structure of PPRA Revised prior establishes a Procurement Cadre.
President‘s approved and additional budgetary action for PPRA Office - Public Service Management establishes resources allocated to PPRA completed. Organization structures and staffing levels of Procurement Management Units (PMUs). National Audit Office staff has been trained, and National Audit Office staff has been Completed procurement and installation of computer trained, and procurement and equipment at the National Audit Office installation of computer equipment at completed, allowing audits through Integrated the National Audit Office completed, Financial Management Information System in allowing audits through Integrated future periods Financial Management Information System in future periods Pay enhancement consistent with the approved Public service salaries increased by Completed budget for FY06, and the overall thrust of the pay 8.34% for the lowest paid, while reform Process agreed to reform the public sector salaries in other grades increased on allowances regime, based on the ongoing review. average by 12.5% duringFY06. PRSC-5 National Audit Office General Report for FY05 Same. Completed issued by April 2006. PRSC- 6 Submission of Anti-corruption Act to Parliament Anti-Corruption Bill enacted. Completed Audit Bill Submitted to Parliament Audit Bill enacted, geared to Afrosai 3 Completed requirements. PRSC- 7 A revised Medium-Term Pay Policy is adopted by Concrete steps taken to prepare a Partially government by February 2009 as the basis for revised Medium-Term Pay Policy by achieved; draft reform of pay and allowances. Draft available by government background Dec. 2008. as the basis for reform of pay and report produced allowances. by March 2009. PRSC- 7 Approval granted by October 2008 to create a new Approval granted to create a new Partially Department of Internal Audit in the Ministry of Department of Internal Audit in the achieved. Finance and Economic Affairs, and regulations Ministry of Finance and Economic Procurement under the Procurement and Supplies Professional Affairs, and regulations under the regulations Board Act prepared and issued (gazetted). Procurement and Supplies Professional prepared but not Board Act prepared. yet gazetted. PRSC- 8 Government approves Targets related to the Dropped. Not met adopted Medium-Term Pay Policy, incorporating strategies to address (inter-district) inequity in human resource distribution. Presentation to Parliament of new Public A new public procurement bill was Completed Procurement Act that increases autonomy of presented to Parliament to give PPRA PPRA and its power to enforce procurement rules. more autonomy and power to enforce procurement regulations ex-ante. Source: Program document 4-8 (World Bank, 2006, 2007b, 2008b, 2009a, and 2010c). 50. The first outcome area covers the PRSC‘s contribution to enhancing accountability of the state through control of corruption. Outcomes in this area are summarized in Table D.17. 129 ANNEX D Table D.17. Indicators for progress in enhancing accountability of the state. Indicator Baseline Target Latest (ICR) Latest (IEG) Source Number of grand corruption 20% 50% NA at completion 59% Director of cases prosecuted as a % of *30 grand corruption (2011) Public investigated cases cases prosecuted in Prosecution 2010, compared to 1 in 2007, 14 in 2008, and 17 in 2009 % of court cases outstanding 70% 40% 13% NA for 2 years or more 51. Control of corruption was intended as an important area for the PRSC 4-8 series, but the results seem poorer than in the first series. Even as anti-corruption legislation was enacted, perceptions of corruption increased, as reflected in ratings by the International Country Risk Guide, the Bank‘s Country Policy and Institutional Assessment, and WGI rankings for Tanzania, shown in Table D.18, and described above. In retrospect, the passage of a new anti-corruption bill seems to have been a suboptimal choice as a prior action for PRSC-6 because there was not much implementation and it may have detracted attention from other actions that may have combated corruption, including through increasing transparency and improving financial management. Table D.18 Tanzania-ICRG and CPIA Ratings 2005 2006 2007 2008 2009 2010 2011 ICRG (1-4 scale with 4 as best) Bureaucracy Quality 1 1 1 1 1 1 1 Corruption 2 2.5 3.5 3 3 2.5 2.5 CPIA (1-5 scale with 5 as best) Budgetary and financial management 4.5 4.5 4.0 3.5 3.5 3.5 Revenue mobilization efficiency 4.0 3.0 4.0 4.0 4.0 4.0 Public administration 3.5 3.5 3.5 3.5 3.5 3.0 Public sector transparency, 3.5 3.5 3.5 3.0 3.0 2.5 accountability, and corruption Sources: ICRG website; Bank CPIA ratings. Note: The International Country Risk Guide (ICRG), produced by the PRS Group, monitors political, economic, and financial risk in 140 countries. The World Bank’s Country Policy and Institutional Assessment (CPIA), conducted annually for all borrowing countries, measures the quality of each countries policies and institutions, with separate ratings for each of 16 criteria. 52. The second outcome area covers the PRSC‘s contributions to improving the effectiveness of public administration by improving incentives for public servants and reforms to local government. Outcomes in this area are summarized in Table D.19. Table D.19. Indicators for progress in improving the effectiveness of public administration. Indicator Baseline Target Latest (ICR) Latest (IEG) Source Current pay as a proportion 8 6% 100% 90% of the government‘s pay (MKUKUTA) target 90% (PRSC) % of LGA expenditures 16% 25% 19.4% 25% GBS Annual calculated on a formula (2006/07) (2010/11) Review 2010 basis Final Report 130 ANNEX D 53. With respect to the issue of improved quality of civil service, the PRSC series focused on pay reform, which again failed to meet its objectives over the PRSC period. The repetition of public service pay reform triggers, with a vague one in PRSC 4 and unmet triggers in PRSCs 7 and 8 (Table 15) points to problems in design. The pay strategy remained across- the-board, continuing the failed approach from the first series. Table D.19 shows the results according to the flawed indicators associated with this design. The intermediate output target was current pay as a percent of the government‘s pay target rather than targeting pay increases to positions with the greatest need (for example, in remote areas) and in agencies critical for delivery of prioritized services, despite the empirical evidence that such attempts are usually unsuccessful (IEG 1999 and 2008; Evans 2008). Even on this indicator, progress was modest, meeting the PRSC target but not the MKUKUTA target. The program document for PRSC-9 notes the persistence of problems related to ―(i) the excessive use of allowances to augment pay by public servants, which distorts incentives to service delivery; (ii) pay and incentives for staff working in underserved and remote areas; as well as (iii) decompression of the salary structure that has in recent years faced the pressure of increased employment of public servants, particularly in the social sector.‖ Indicators suggest that outcomes were not good: the WGI on government effectiveness deteriorated from -0.35 (2005) to -0.45 (2008), in contrast to the program target of 0.0 for 2010.73 This is consistent with the ICRG rating for Bureaucratic Efficiency, which remained steady at 1 on a 0-4 scale, and the CPIA rating for Quality of Public Administration, which declined from 3.5 in 2005-09 to 3.0 in 2010. 54. Local government reform in Tanzania has had a mixed record since 2005— disbursing more money but taking few actions to improve the effectiveness of that funding. The share of the budget transferred directly to LGAs increased, following the de- concentration of the secondary education budget. A system of intergovernmental transfers, which provides performance-based discretionary grants (Council Development Grants: CDGs; and Capacity Building Grants: CBGs) to LGAs, was disbursing by 2010. The development partner group on local government reform gave the program an unsatisfactory rating because of limited progress made in providing autonomy in human resources management at the LGA level; transparency and accountability of use of funds undermined by multi-channels of funding at the local level; and continued large inequity in the allocation of funds among districts (World Bank 2010c). Although the PRSC series had the intention of improving the effectiveness of decentralization, none of the five PRSCs had prior actions to promote this—a flaw of project design. On the other hand, decentralization efforts seem to have declined as an activity area for public sector reform. Although more money in the aggregate is going to local governments, the targeting and accountability have weakened. There were no prior actions for this issue in any of the PSRCs 4-8. No Public Expenditure and Financial Accountability review has been done at the local level since 2006, although the funding going through those governments has increased greatly. 55. The third outcome area covers the PRSC‘s contributions to improving public financial management through reforms in procurement, external and internal audit, and 73 Fundamental problems with the WGI as a measure of change over time imply that it was a flaw of program design to pick it as an indicator. 131 ANNEX D budget and PFM issues. Outcomes in this area are summarized in Table D.20. Budget management is deteriorating, while auditing and procurement are becoming more thorough. Table D.20. Indicators for progress in improving public financial management. Indicator Baseline Target Latest (ICR) Latest (IEG) Source Recurrent budget deviation 5% 10% 12.7% 13.7% Program (2005/06) Document (PER 2011) PRSC-9 National Audit Office Fully 9 out of 10 AFROSAI National starting to introduce compliant with AFROSAI 3 and ISA Audit Office INTOSAI and ISA AFROSAI 3 conditions standards Report, met used in all FY2011 audits Delay in submission of 6 months 0 months 0 months 0 months National National Audit Office audit Audit Office report to Parliament Report, FY2011 Number of procuring 39% 80% 73% 64% National entities complying with the (2006/07) Audit Office Public Procurement Act Report, 2004 FY2011 56. For procurement reform, the strategy was to implement the new law passed during the first PRSC series—to stop using a central tender board, to allow Ministries and local governments to do their own procurement, and to create a regulatory body to set and enforce standards for efficient and non-corrupt procurement. This is international best practice and aimed to expedite procurement with independent checks, appeals and audits. The number of agencies participating in the Public Procurement Regulatory Authority evaluations grew from 20 in 2006/7 to 100 in 2007/8 and to 174 in 2010/11. The percent of agencies found in compliance with PPRA regulations rose to 50 in 2007/8 and reached a plateau around 70 percent in 2009/10 and 2010/11 (PPRA Annual Performance Evaluation Report, 2011). The 2009 Public Expenditure and Financial Accountability review and the 2011 DfID Risk Assessment, nonetheless, noted the problem of low compliance with procurement laws and regulations, stating that ―anecdotal evidence and the national corruption survey still question the integrity of the actual [procurement] selection process at both central and local government level‖ (DfID 2011). 57. External Audit was strengthened with the passage of the Public Audit Act in 2008, increasing the independence, powers and responsibilities of the Controller and Auditor General. To reform internal audit the government moved in 2008 the internal audit function from the office of the Accountant General to make it independent, with direct reporting to the Permanent Secretary of the Treasury. Auditing showed progress in terms of increased coverage capacity, and independence of the National Audit Office. The Audit Office is well respected in the region and through 2014 is the chair of AFROSAI-East. Their annual reports were on time since 2005 and contained substantive analysis and frank critiques. Outcomes reported by the audits (a measure of the quality of PFM) showed progress in increasing the proportion of ―unqualified opinion‖ until 2009, which reversed in FY 2010, but recovered for FY 2011. Some important ministries, like Finance and Education, were still not getting ―unqualified opinions‖. Also ―adverse opinions‖ in audits of local governments continued to 132 ANNEX D be prevalent (TWAWEZA, 2010). These reflected PFM problems in the audited MDAs, and a plus for the auditors in identifying them. While problems in the audits of some Bank projects showed room for improvement in quality of audits, the timeliness and independence of the National Audit Office met expectations. The main problem recently has been the lack of government responsiveness to National Audit Office recommendations.74 58. There was little progress made with respect to Budget/PFM issues, as the government umbrella PFM reform program slowed down in 2006-7. The third phase of the PFM Reform Program started in July 2008, with focus on improvements to cash management, payment systems, coverage and integration of Integrated Financial Management System, and budget classification. Despite this, the CPIA and Public Expenditure and Financial Accountability ratings for PFM deteriorated over the period, and the Open Budget ratings did not make net improvement. Recurrent budget deviation at vote level improved, but remained far above the target of 10 percent, as shown in Table D.21. The main recipients of reallocated funds were those Ministries with large staff complements—the Education, President‘s Office and Cabinet, Defense, and Police—because most of the funds were for salary arrears and adjustments. Table D.21. Trends in budget execution Indicator 2002/03 2003/04 2004/05 2005/06 2006/07 2007/08 2008/09 Source Recurrent budget 22.1% 23.9% 20.4% 17.1% 16.0% 16.7% 18.3% PEFA 2005, deviation at vote (22.6%) (13.1%) PEFA 2009 level Note: Figures in parentheses exclude force majeur items. PEFA – Public Expenditure and Financial Accountability review 59. The early years of the second series of PRSCs saw sound implementation of the public expenditure management agenda, including adoption of budgets in line with MKUKUTA priorities, as well as satisfactory budget execution. The situation changed toward the final operations, with varying outcomes. Two triggers on budget execution and alignment of expenditures to MKUKUTA were missed in PRSC7 and PRSC8, respectively, although the situation seems to have improved in 2010/11 after deteriorating the year before (Public Expenditure Review 2011). The PERs also found major decline in allocations for capital spending, very high and growing share of wages in total budget, and a very large increase in discretionary spending in the 2009 budget, all quite separate from the fiscal stimulus needed for mitigating the global financial crisis of 2008. In terms of the adequacy of monitoring indicator and its target level, the PER stated that the 13 percent deviation of actual expenditure from the budget (the actual indicator) was high, and suggested the need to aim for less that 10 percent. However, the PRSC target remained 18 percent throughout the second series of PRSC operations. 74 The National Audit Office report on FY2010/11 said the following: ―Most of the recommendations issued in the previous year‘s General Report of the Central Government were not responded to, which is an indicator of lack of seriousness on the part of the Government in implementing those recommendations.‖ (p. xxiv) 133 ANNEX E Annex E. List of Persons Consulted Government of Tanzania Name Title Organization 1 Ndulu, Benno Governor Bank of Tanzania 2 Lyimo, Peniel M. Permanent Secretary Prime Minister‘s Office 3 Khijjah, Ramadhari Permanent Secretary Ministry of Finance 4 Mpango, Philip Isdor Executive Secretary President‘s Office, Planning Commission 5 Mgalula, Happiness N. Dep. Executive Secretary President‘s Office Planning Commission 6 Mwanri, Florene Dep. Executive Secretary President‘s Office Planning Commission 7 Madete, Loran Dep. Executive Secretary President‘s Office Planning Commission 8 Mafuru, Patroba Dep. Executive Secretary President‘s Office Planning Commission Asst. Commissioner for 9 Buretta, Jerome Ministry of Finance Multilateral Aid 10 Mavura, John External Finance Officer Ministry of Finance Commissioner, Policy Analysis 11 Shallanda, Bedason A Ministry of Finance Department Acting Commissioner of Public- 12 Mhilu, Frank M. H. Ministry of Finance, PPP-Unit Private Partnerships Ministry of Agriculture, Food Security, 13 Reg, Mbongo, Futakamba Dep. Permanent Secretary & Cooperatives Ministry of Agriculture, Food Security 14 Achayo, Emmanuel M., Director of Policy & Planning and Cooperatives Ministry of Agriculture, Food Security 15 Mtambo, Karum Director of Food Security and Cooperatives Assistant Director of Crop Ministry of Agriculture, Food Security 16 Lemwele, Ombaeli Monitoring & Early Warning and Cooperatives Assistant Director Agricultural Ministry of Agriculture, Food Security 17 Msolla, M. M. Inputs and Cooperatives Ministry of Agriculture, Food Security 18 Tabwene, Didas Statistician and Cooperatives Ministry of Livestock Development & 19 Budeba, Yohana L. Dep. Permanent Secretary Fisheries Ministry of Natural Resources and 20 Kimwaga, Zahoro A. Assistant Director Tourism Assistant Director of Forestry & Ministry of Natural Resources & 21 Mkamba, Gladness A. Beekeeping Tourism Ministry of Natural Resources & 22 Mfunda, Iddi Principal Economist Tourism Acting Director of Policy Ministry of Lands, Housing and Human 23 Lazaro, Simon L Planning Settlement Ministry of Lands, Housing and Human 24 Mnyanga, Christ Principal Land Surveyor Settlement 25 Mwinjaka, Shabani R. Dep. Permanent Secretary Ministry of Industry and Trade Director of Trade Promotion & 26 Majengo, Odilo Ministry of Industry and Trade Marketing 27 Nyagiro, Obadiah M. Economist Ministry of Industry and Trade 28 Mmasi, Primi Trade Officer Ministry of Industry and Trade Director of Administration and 29 Kayuni, Emmanuel A. Ministry of Industry and Trade Human Resource Management 30 Maswi, Eliakim C. Permanent Secretary Ministry of Energy and Minerals Assistant Commissioner for 31 Magayane, Alex A. Small Scale Mining Ministry of Energy and Minerals Development 32 Massawe, Getty Ministry of Energy and Minerals 134 ANNEX E Tanzania Electric Supply Company 33 Kassanga, Lusekelo J. Chief Financial Officer (TANESCO) Principal Officer Small Power 34 Shayo, Charles E. TANESCO Projects Manager Projects (Generation) 35 Jilim, Simon M. TANESCO Investment 36 Lutegauya, Rwabanji TANESCO Energy and Water Utilities Regulatory 37 Haruna Masebu Director General Authority Energy and Water Utilities Regulatory 38 Omujuni, Charles Stephen Director of Natural Gas Authority Energy and Water Utilities Regulatory 39 Mwambo, Mathew Technical Manager, Electricity Authority 40 Ndunguru, John S. Dep. Permanent Secretary Ministry of Works Director of Safety of 41 Kipande, Madeni Ministry of Works Environment 42 Makala, Stanford K. Senior Economist Ministry of Works Acting Director Policy and 43 Migire, Gabriel J. Ministry of Transport Planning Ecoomist/Former Director of 44 Temba, Abisai Ministry of Transport/DFID Policy and Planning 45 Kaira, Charles K. Transport Specialist Ministry of Transport/DFID 46 Hine, John L. Senior Transport Specialist Ministry of Transport/DFID Tanzania National Roads Agency 47 Mfugale, P.A.L. Chief Executive (TANROADS) 48 Rwiza, Jason Director of Planning TANROADS 49 Seff, Victor Projects Manager TANROADS 50 Mgesha, Doto J. Transport Economist TANROADS 51 Masaki, Arnold Engineer, Planning TANROADS 52 Mgawe, Ephraim Ngoza Director General Tanzania Ports Authority Director of Planning and 53 Nkya, Florence P. Tanzania Ports Authority Investment 54 Kimaro, Theophil Head, PMU Tanzania Ports Authority 55 Kitillya, Harry M. Commissioner General Tanzania Revenue Authority Planning and Modernization 56 Maganga, Mary Ngelela Tanzania Revenue Authority Programme Manager 57 Mwamunyange, Monica Commissioner for Budget Ministry of Finance Principal Economist, Poverty 58 Cheyo, Mudith B. Eradication & Economic Ministry of Finance Empowerment Public Procurement Regulatory Authority 59 Mlinga, Ramadhan Chief Executive Officer (PPRA) Prevention and Control of Corruption 60 Hoseah, Edward Director General Bureau (PCCB) 61 Masaka, Lilian Dep. Director General PCCB 62 Paul, Victor Asst. Accountant General Office of Accountant General 63 Mwakapalila, Francis Acting Chief Auditor General National Audit Office President‘s Office of Public Service 64 Kwizum, Hab M. Dep. Permanent Secretary Management (PO-PSM) 65 Kafanabo, Joyce Program Coordinator PSRP PO-PSM 66 Kabunouguru, Mathias PO-PSM 67 Mkony, Selina Head Public Financial Management Secretariat Prime Minister‘s Office – Regional 68 Kattango, Hussein Permanent Secretary Administration and Local Government (PMO-RALG) 69 Kidata, Alphayo Dep. Permanent Secretary PMO-RALG 70 Mallya, Joseph Lead Consultant PMO-RALG 135 ANNEX E 71 Chuwa, Albina A. Director General National Bureau of Statistics 72 Maro, Radegunda H. Director of Statistical Operations National Bureau of Statistics 73 Oyuke, Morrice N. Economic Statistics Director National Bureau of Statistics Industrial & Construction 74 Sawe, Joy E. National Bureau of Statistics Statistics Manager 75 Matowa, Anna Asst. Director for Policy Ministry of Health & Social Welfare 76 Kenga Asst. Director for Budget Ministry of Health & Social Welfare 77 Joseph Rubona Asst. Director for M&E Ministry of Health & Social Welfare 78 Gabriel A. Saelie Director of Policy & Planning Ministry of Water Acting Director, Water 79 Justus Rwetabula Ministry of Water Programme Coordination Development Partners Name Title Organization 1 Sebregondi, Filiberto Ceriani Ambassador, Head of Delegation European Union 2 Strampeli, Enrico First Counsellor European Union U.K. Department for International 3 Elliot, Marshall Head of Delegation Development (DFID) 4 Moberly, Richard Senior Economic Adviser DFID 5 Forster, Stuart Senior Governance Adviser DFID 6 Mwakyebe, Atu Programme Officer DFID 7 Nelson, Kerry Value for Money Adviser DFID 8 Mapunda-Kihunrwa, Gertrude Social Policy Adviser DFID 9 Andersen , Steen Sonne Dep. Head of Mission Danish Embassy 10 Noejgaard, Kirstine Thyge Counsellor, Health Danish Embassy Canadian International Development 11 Huntington, Corey First Secretary Agency Canadian International Development 12 Armstrong, Christopher Counsellor, HIV & AIDS Agency 13 Peter Allum Division Chief International Monetary Fund 14 Aiko Mineshima Economist International Monetary Fund 15 Pauschert, Dirk Advisor, Economic Regulation GtZ First Secretary Sustainable 16 Al Fayadh, Samer Embassy of Sweden Services 17 Mwakifwamba, Stephen Programme Officer Energy Embassy of Sweden Program Management Specialist 18 Knudsen, Niels Vestergaard United Nations Development Programme (Aid Effectiveness) Private Sector, Academics, and NGOs Name Title Organization 1 Kamote, Hussein S. Director of Policy and Advocacy Confederation of Tanzania Industries 2 Green, Kenneth M. President SusDev Tech, LLC Institute of Management and 3 Olomi, Donath Chief Executive Officer Entrepreneurship Development Chairman (Former Permanent 4 Mgonja, Gray Equator Consulting Co. Ltd Secretary, Ministry of Finance) 5 Rajani, Rakesh R. Head Twaweza ni sisi 6 Wangwe, Samuel Executive Director Research on Poverty Alleviation 7 Krolikowski, Aaron Graduate Research Student University of Oxford World Bank/IFC Staff Name Title Role/Sector 1 Dongier, Phillippe Country Director (2012-) Guidance and dialogue 136 ANNEX E Former Country Director (2007- 2 McIntire, John Guidance and dialogue 2011) Former Country Director (2002- 3 O‘Connor, Judy Guidance and dialogue 2007) 4 Krumm, Kathie Former Sector Manager (2004-2010) Guidance and dialogue 5 Morriset, Jacques Lead Economist Guidance and dialogue 6 Utz, Robert Lead Economist Task Team Leader (PRSC 2-5) 7 Zacchia, Paolo Lead Economist Task Team Leader (PRSC 6-8) 8 Yoshino, Yutaka Senior Economist Task Team Leader (PRSC 8-9) 9 Kweka, Josephat Senior Economist PRSC team (PREM) 10 Mungunasi, Emmanuel Economist PRSC team (PREM) 11 Schlotterer, Robert Senior Financial Analyst Energy 12 Lewi, Negede Senior Highway Engineer Transport 13 Mchomvu, Yonas E. Transport Specialist Transport 14 Semgalawe, Zainab Zitta Senior Rural Development Specialist Agriculture & Rural Development Regional Extractive Industries 15 Omar, Abdul Transparency Initiative Coordinator Natural Resources (SEGOM) 16 Sheldon, Christopher Lead Mining Specialist Natural Resources 17 Kasirye, Dan Senior Investment Officer IFC Country Representative 18 Lyniewska, Agnieskia E T Consultant Private Sector Development 19 Haazen, Dominic Lead Health Policy Specialist Health 20 Malangalila, Emmanuel Senior Health Specialist/ Consultant Health 21 Joshi, Arun Senior Education Specialist Education 22 Ueda, Satoru Lead Water Resources Specialist Water 23 Morin, Denyse Senior Operations Officer Public Sector Governance 24 Biseko, Denis Senior Public Sector Specialist Public Sector Governance 25 Lee, Stevan Senior Economist Public Sector Governance Lead Financial Management 26 Brar, Parminder Public Financial Management Specialist 27 Ahluwalia, Sanjeev Senior Public Sector Specialist Public Financial Management 28 Mehta, Barjor Senior Urban Specialist Decentralization 29 Glasser, Matthew Adviser Decentralization 137 ANNEX F Annex F. Borrower Comments 138 ANNEX F 139 ANNEX F 140