Document of The World Bank Report No. : 87597 PROJECT PERFORMANCE ASSESSMENT REPORT THE REPUBLIC OF UGANDA SECOND LOCAL GOVERNMENT DEVELOPMENT PROJECT (IDA-37730, IDA-3773A AND IDA H0410) May 8, 2014 IEG Public Sector Evaluation Independent Evaluation Group Currency Equivalents (annual averages) Currency Unit = Uganda Shillings (UGX) 1997 US$1. 00 UGX 1083 1998 US$1. 00 UGX 1240 1999 US$1. 00 UGX 1455 2000 US$1. 00 UGX 1644 2001 US$1. 00 UGX 1756 2002 US$1. 00 UGX 1798 2003 US$1. 00 UGX 1964 2004 US$1. 00 UGX 1810 2005 US$1. 00 UGX 1781 2006 US$1. 00 UGX 1831 2007 US$1. 00 UGX 1723 2008 US$1. 00 UGX 1720 2009 US$1. 00 UGX 2030 2010 US$1. 00 UGX 2178 2011 US$1. 00 UGX 2523 2012 US$1. 00 UGX 2505 Abbreviations and Acronyms AAA Analytical and Advisory Assistance AD Administrative Units CAO Chief Administrative Officer CAS Country Assistance Strategy CFAA Country Financial Accountability Assessment CPAR Country Procurement Assessment Report CSO Civil Society Organization DFA Development Finance Agreement DMRP District Municipal Resource Pool DTB Development Transfer Budget ENR Energy and Natural Resources ESMF Environmental Assessment and Social Management Framework FDS Fiscal Decentralization Strategy FY Fiscal Year GDP Gross Domestic Product GNI Gross National Income GOU Government of Uganda HlPC Heavily Indebted Poor Countries HIV Human Immunodeficiency Virus HPPG Harmonized Participatory Planning Guide ICBP Institutional Capacity Building Project ICR Implementation Completion Report IDA International Development Association IEG Independent Evaluation Group IEGPS IEG Public Sector Evaluation IGG Inspectorate General of Government IMF International Monetary Fund JARD Joint Annual Review of Decentralization KCC Kampala City Council LDG Local Development Grant LG Local Government ii LGA Local Government Act LGDP Local Government Development Project LGFC Local Government Finance Commission LoGSIP Local Government Sector Investment Program LLG Lower Local Government LOGICS Local Government Information and Communication System HLG higher Local Government M&E Monitoring and Evaluation MDA Ministries Departments and Agencies MOFPED Ministry of Finance, Planning and Economic Development MoLG Ministry of Local Government MPS Ministry of Public Service MTEF Medium Term Expenditure Framework MTR Mid-Term Review NDP National Development Plan NGO Non-Governmental Organization NRM National Resistance Movement OPM Office of the Prime Minister PAD Project Appraisal Document PDO Project Development Objective PEAP Poverty Eradication Action Plan PEFA Public Expenditure and Financial Accountability PFM Public Finance Management PPAR Project Performance Assessment Report PRSC Poverty Reduction Support Credit PRSP Poverty Reduction Strategy Paper RTB Recurrent Transfer Budget SFR Strategic Framework for Reform TPC Technical Planning Committee TTL Task Team Leader UBOS Uganda Bureau of Statistics UGX Uganda Shilling UNDP United Nations Development Program UPE Universal Primary Education USE Universal Secondary Education VFM Value for Money Fiscal Year 1 July - 30 June 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 : Ms. Lourdes Pagaran iii Contents Principal Ratings ................................................................................................................. v Key Staff Responsible......................................................................................................... v Preface............................................................................................................................... vii Summary ............................................................................................................................ ix 1. Background and Context................................................................................................. 1 2. Objectives, Design, and Relevance ................................................................................. 4 Objectives ....................................................................................................................... 4 Relevance of Objectives ................................................................................................. 4 Design ............................................................................................................................. 5 Relevance of Design ....................................................................................................... 6 Design of Monitoring and Evaluation............................................................................. 6 3. Implementation ............................................................................................................... 7 4. Achievement of the Objectives ....................................................................................... 9 Improve Local Government institutional performance................................................... 9 Outputs ........................................................................................................................ 9 Outcomes .................................................................................................................. 11 Sustainable, decentralized service delivery .................................................................. 13 Outputs ...................................................................................................................... 13 Outcomes .................................................................................................................. 14 5. Efficiency ...................................................................................................................... 17 6. Ratings .......................................................................................................................... 18 Outcome ........................................................................................................................ 18 Risk to Development Outcome ..................................................................................... 18 Bank Performance ......................................................................................................... 19 Quality at Entry ......................................................................................................... 19 Quality of Supervision .............................................................................................. 20 Borrower Performance .................................................................................................. 20 Government Performance ......................................................................................... 20 Implementing Agency Performance ......................................................................... 20 Quality of M&E ............................................................................................................ 21 7. Lessons .......................................................................................................................... 21 References ......................................................................................................................... 23 Annex A. Basic Data Sheet ............................................................................................... 24 iv Annex B. Local Governments in Uganda – Overview and Timeline ............................... 26 Annex C. Evolution of Bank Portfolio ............................................................................. 41 Annex D. List of Persons Interviewed .............................................................................. 45 Annex E. Borrower Comments ......................................................................................... 46 Boxes Box 1. If you live in Wakitaka and are pregnant, remember to charge your phone every day ..................................................................................................................................... 17 Tables Table 1. Project Costs and Funding, Planned versus Actual (US$ million) ....................... 7 Table 2. Growth in Local Government (LG) and Administrative (AD) Units by Level .. 27 Table 3. Central Government Fiscal Operations (UGX billion, outturns) ........................ 27 Table 4. Functional Classification of Local Governments' Expenditure .......................... 27 Table 5. Consolidated Local Government Financial Operations (FY06-FY10, UGX Billion) .............................................................................................................................. 28 Table 6. Distribution of Districts and Municipal Councils by Performance as assessed by MoLG ................................................................................................................................ 29 Table 7. Composition of Grants to Local Governments (F06-FY12, UGX billion)......... 29 Table 8. Sub-National Administrative Units (Highest Level) for Sub-Saharan Africa countries with a population of more than twenty million, by number .............................. 30 Table 9. Evolution of Districts in Uganda (2001 – 2011)................................................. 31 Table 10. Allocation Parameters of Selected Conditional Grants .................................... 37 Figures Figure 1.Transfers to Local Governments as a share of public expenditure and GNI........ 8 Figure 2. Share of Local Governments Receiving a Penalty Score (%, 2005-2011)........ 12 Figure 3. Distribution of Local Development Grant Projects by Sector (n=31,832) ........ 14 Figure 4. Wakitaka Health Unit in the Sub-Council Mafubira, Jinja .............................. 16 Figure 5. Number of districts in Uganda (1962-2011) ..................................................... 26 This report was prepared by Stefano Migliorisi (Consultant), who assessed the operations in October 2012. The Task Manager is Lourdes Pagaran. The report was peer reviewed by Michael Lav and panel reviewed by Brett Libresco. Yezena Yimer provided administrative support. v Principal Ratings THE REPUBLIC OF UGANDA: Second Local Government Development Project – (IDA-37730, IDA-3773A AND IDA H0410) ICR* ICR Review* PPAR Outcome Satisfactory Satisfactory Moderately Unsatisfactory Risk to Moderate Moderate High Development Outcome Bank Performance Satisfactory Satisfactory Moderately Satisfactory Borrower Satisfactory Satisfactory Moderately Performance Unsatisfactory * The Implementation Completion Report (ICR) is a self-evaluation by the responsible Bank department. The ICR Review is an intermediate product that seeks to independently verify the findings of the ICR. Key Staff Responsible Division Chief/ Project Task Manager/Leader Sector Director Country Director Appraisal Lance Morrel Jaime M. Biderman Judy M. O’Connor Completion Naa Dei Nikoi Jaime M. Biderman John McIntire 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 the 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 stated 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 Project Performance Assessment Report (PPAR) covers an operation implemented over the period FY03-FY08. The project’s development objective was to improve the Local Government institutional performance for sustainable, decentralized service delivery. The report presents findings based on review of the Program Documents, the Implementation Completion and Results Reports, IEG’s Implementation Completion and Results Report Review, aide-memoires and supervision reports, and other relevant materials. Stefano Migliorisi visited Uganda for IEG on October 1 – 13, 2012 to interview government officials, the staff of non-governmental organizations, project staff, donor representatives, and other stakeholders. Additional interviews with Bank staff members, donor representatives, and other informants were carried out at headquarters or by conference call. The assessment aims, first, to serve an accountability purpose by verifying whether the operation achieved its intended outcomes. Second, the report draws lessons that are intended to inform future operations of this nature in Uganda and other low-income states. Following standard IEG procedures, the report was shared with the government for comment. The comments received are included as Annex F. ix Summary The objective of the Republic of Uganda’s Second Local Government Development Project (LGDP II) was “to improve the Local Government institutional performance for sustainable, decentralized service delivery.” At the time of appraisal, the first phase of decentralization—with the establishment of an enabling legislation, a conducive intergovernmental fiscal framework, and transfer of competencies to Local Governments—was well under way, and there was a need to support the central government and local authorities in its implementation. The focus on improved service delivery, particularly for the poor, was an important step towards achieving the Millennium Development goals on education, health, and access to water, given the predominant role of Local Governments in this respect. Project design and implementation was informed by the Mid- Term Review’s recommendations and the ICR findings and lessons learned from the previous operation, LGDP I. In particular, LGDP II was designed as an instrument to mobilize broad grant support to implement decentralization reform with a possible strong demonstration effect for the future in terms of development partner coordination and harmonization. Improved decentralized, sustainable service delivery would be achieved by (a) supporting decentralization through appropriate strategies and institutions (Component 1); (b) strengthening the capacity of Local Governments to provide better services, administer their resources, and raise more revenues (Components 3 and 4); and (c) providing direct financial support to Local Governments through non-sectoral conditional grants on a formula basis taking into account their needs, capacity, accountability and performance (Component 2). The project was approved on May 29, 2003, became effective on October 29, 2003 and closed on December 31, 2007. The total costs at appraisal were US$165 million: US$125 million from IDA (of which US$75 million as grant), US$15 million from the Netherlands, US$14.8 million from Government and Local Authorities, US$7.5 million from Ireland, US$1.8 million from the Danish International Development Assistance (DANIDA), and US$0.3 million from the Austrian Development Corporation. Actual costs were US$181.7 million and the additional costs were funded through the appreciation of the XDR against the US dollar, and increased contribution from Government, mostly in the form of Local Development Grants. The project was initially successful in building administrative capacity of Local Governments and supported over 30,000 sub-projects in several sectors. However, such improved administrative capacity has been underutilized due to substantial policy reversals since 2005. Institutional performance of Local Government started to deteriorate from 2007, as shown by the results of annual national assessments carried out by the Ministry of Local Government. The 2008 National Assessment, carried out immediately after project closing, showed that only 84 local governments out of the national total 1105 at the time earned rewards, while 931 were penalized, due to declining quality of record keeping, meager revenue collection, weak budgeting and planning, widespread use of force accounts against procurement regulations, and a more general lack of interest in the national assessment itself. The policy reversals started in 2005 have included near elimination of local revenue base, reduction of transfers to Local Governments, increased percentage of conditional grants, and creation of new districts, mostly for political patronage as shown by several studies on this x topic1. These reversals have been hindering service delivery and value for money, as noted in the latest Bank’s Country Assistance Strategy, leading to a weakening of Local Governments’ discretionary powers, a centralization of functions and resources, and a reduction of available financial and human resources at district level. Each new district increases the administrative costs, takes away staff of existing ones and reduces resources available for service delivery, creating a high administrative burden at the district level, with wages consuming a large and increasing share of total expenditures, leaving insufficient funding for non-wage needs. The combination of these factors has made the progress achieved under the project not sustainable, and affected the quality of service delivery. As the project’s development objective was to improve the Local Government institutional performance for sustainable, decentralized service delivery, this Review concludes that the achievement of objectives was Modest. The initial improvements in the institutional performance of Local Governments were not sustainable and are currently at a high risk of being reversed due to the new Government of Uganda’s attitude towards decentralization, although there have been some positive developments recently like the moratorium on the creation of new districts decided by Cabinet in March 2013. Lack of funds for operations and maintenance has subsequently led to the progressive erosion of the initial improvements in quality of management, and infrastructure. The project’s efficiency was nevertheless substantial. The sub-projects implemented under LGDP II were more cost effective compared to those under the LGDP I, even if comparable in size. Efficiency of sub-projects was relatively good, as the Value for Money Analysis, carried out by an independent firm, rated the efficiency of the subprojects it analyzed as essentially ‘good’ on a three-point scale. The relevance of objectives and the project’s efficiency were all substantial, while relevance of design and efficacy were modest, and the project’s outcome moderately unsatisfactory. The operation was aimed at supporting the building of capacity at Local Governments and the establishment of appropriate strategies for decentralization. These enhancements, combined with direct financial support for investments in social infrastructure through sub-projects, led to a visible improvement in basic service delivery. However, the GOU policy reversals, during and after the life of the project, made such results unsustainable in the medium term. The monitoring framework was sophisticated and included a dedicated Information and Communications System (LOGICS), designed and piloted during previous project; a financial information module (LOGICS Plus) allowing Local Governments to report on expenditures and revenues; annual Local Government performance assessments carried out by a team consisting of officials from the Ministry of Local Government, sector ministries and a private contractor for quality assurance; and a "Value for money" audit. However, the M&E design did not adequately address the multiplicity of monitoring systems, and did not include a systematic beneficiary feedback loop. There is a high risk to the limited institutional outcomes achieved. Political commitment has declined over time, leading to substantial policy reversals from 2005 onwards. It is likely that the current trend to centralize functions and increase the number of districts, as new ones are currently being considered, will over time reduce the progress made in building capacity and accountability of Local Governments. 1 See for example Green 2008 and Manyak and Katono 2010. xi Bank’s performance was moderately satisfactory. The design of the project was aligned with the Poverty Reduction Strategy Paper (PRSP) and the Bank's Country Assistance Strategy (CAS), and remains relevant to the latest CAS, and, to a lesser extent, to the latest Government’s National Development Plan. Project design drew on lessons from LGDP I (i. e. , more commitment to sustainability of investments, efficiency in utilization of the capacity building grants and the broadening of Local Government own revenue sources and collection methods), as well as from the Bank’s and other donors in Uganda and neighboring countries. The design also draws appropriately on the Bank’s Analytical and Advisory Assistance. The Bank selected an appropriate instrument, which was part of a long history of support to decentralization dating back for at least a decade. However, the Bank did not properly react to the GOU policy reversals that led to a growth in unconditional grants not consistent with the Ugandan Constitution. Borrower performance was moderately unsatisfactory. The Government’s performance in particular was unsatisfactory. Prior to the introduction of a multi-party system in 2005, the GOU had shown strong commitment to decentralization and the project. However, since 2005, commitment has been declining, leading to a substantial reduction in Local Governments’ mandates, funding, and autonomy. The performance of the Ministry of Local Government was instead satisfactory, as it led the process of reform and tried to adapt to an increasingly challenging environment. The lessons from this operation are that:  Policy reversals can cause serious damage to otherwise significant project outcomes, and are difficult to counter. District proliferation or reduction in un-earmarked funding or Local Governments’ rights to raise revenues need to be monitored closely as they could be an early signal of policy reversal.  Monitoring should be focused on outcome in addition to process indicators and unified across sectors.  Decentralization is not a sector, while it was treated as such in Uganda with a Sector Working Group, a Sector Investment Plan and specific donor support. Decentralization of service delivery affects all sectors of the economy and should be supported in a harmonized way across sectors and donor programs.  Many conditional grants to Local Governments are funded through donor programs. A fully decentralized sector allocation, supported through government budgets, requires a change in the way donors allocate funds across sectors within a given country as such allocations cannot be determined a priori any more if the choice of sectors is truly delegated to Local Governments and communities. This is often incompatible with the development cooperation frameworks of bilateral donors, or the sector-specific teams and earmarked funds of multilaterals, that work in tandem with line ministries in developing countries like Uganda. Caroline Heider Director-General Evaluation 1 1. Background and Context 1.1 Uganda was one of the first countries in Africa to embark on decentralization. When the project was designed, Uganda had already ‘one of the most far-reaching Local Government reform programs in the developing world’ (Francis and James 2003:325), as detailed in Annex B. Such program went through two distinct phases as discussed below. 1.2 Phase 1 - Establishment of an enabling legislation, a conducive intergovernmental fiscal framework, and transfer of competencies to Local Governments (1993-2004). In 1992, decentralization was formally launched through the decentralization policy pronouncement. This was followed by Constitutional Reform (1995), and the promulgation of the Local Government Act (1997). Local Governments were given responsibilities, inter alia, for primary education, agriculture, water, sanitation, primary health care, and district and feeder roads, and a system of conditional as well as unconditional government grants was put in place. Development partners assisted the central government in establishing a strong legal and fiscal framework, and supported Local Governments in the provision of essential services falling under their new mandate. 1.3 During this phase, the Bank provided support through the Institutional Capacity Building Project - ICBP (1995-2002), the District Health Project (1995-2002) and LGDP I (1999-2004). The Bank in particular supported the Government of Uganda in establishing a more conducive legal and institutional framework and in experimenting with more discretion over expenditure at Local Government level and a more rational grant allocation system. Other development partners (DANIDA, UNDP) provided support on similar decentralization issues during the same period. 1.4 When the project was designed (2002-2003), the Government had just started implementing its 2002 Fiscal Decentralization Strategy (FDS) which aimed at addressing the challenges within the existing inter-governmental fiscal transfer system and revenue mobilization. It had also established by Act of Parliament the Local Government Finance Commission (LGFC) to advise Government on fiscal decentralization issues, and launched a new Local Government capacity building framework. The Bank seized the opportunity provided by this technical entry point, and offered support for its implementation through the project. 1.5 As stated in the Project Appraisal Document-PAD (World Bank 2003a: 6), “through the FDS, the Government has made the strategic choice to mainstream the LGDP approach to fiscal decentralization, emphasizing improvement in decentralization of public service delivery through greater autonomy and downward accountability of Local Governments. LGDP II directly supports this strategic choice of Government. ” The FDS itself was considered by the Bank (World Bank 2003a:12) as “the strongest indicator of the Government’s commitment and ownership. ” During the project, the GoU was supposed to evaluate its pilot phase of the FDS and mainstream the resulting approach to all Local Governments, with follow-on support from the Bank through a future decentralization Poverty Reduction Support Credit -PRSC immediately after the closing of the project (World Bank 2003a: 10). 1.6 When FDS was approved, Local Government grants were above 30 percent of public expenditure. Support was mainly in the form of conditional grants (68 percent of the total) whose number had grown to 21 since the establishment of the Poverty Action Fund in 1998. As stated 2 in LGFC (2012), “the key concern, in 2002, was the impact of this growth in conditional grants on local governance and development, local service delivery and sustainable implementation of the poverty alleviations commitments under the Poverty Eradication Action Plan (PEAP). Local governments were experiencing a build-up of administrative costs arising from multiple procedures, bank accounts and lines of reporting due to the increasing number of conditional grants. ” 1.7 Phase 2- Transition to a multi-party system, recentralization and proliferation of Local Governments (2005-2012). In 2005, Uganda adopted a multi-party system, calling national and local elections for 2006. As noted in World Bank (2010), there have been substantial policy reversals since 2005 that are hindering service delivery and value for money, including near elimination of local revenue base, reduction of transfers to Local Governments, increased percentage of conditional grants, and creation of new districts. Five trends emerged, as discussed below. 1.8 The first trend was a proliferation of Local Governments. The number of districts, the basic unit of Local Government in Uganda, has almost doubled over the last ten years (from 56 in 2000 to 111 in 2011). A commission of inquiry in 1987 had recommended a reduction of the then 37 districts, as many were financially unviable. On the contrary, new districts were created in 1994, 1997, 2000, 2005, 2006, 2007, 2009 and 2010. By the time the project was designed, the number of districts had grown to 56 and when it closed there were already 80 districts. The total number of districts reached 111 in 2010, while the total Local Governments and administrative units grew from 50,800 in 2004 to 74,800 in 2011, an increase close to 50 percent. 1.9 According to Green (2008), most new districts were not created along ethnic or party lines. New districts were mostly used as a patronage tool, as they allowed the creation of new Parliamentary seats, and Local Government posts, each representing a patronage opportunity. “The creation of the districts did not follow any established parameters, neither was the process informed by administrative necessity or economic rationale. Instead, the President announced their creation via presidential decrees, often to reward politicians threatening to withdraw support for the NRM, or to punish those who had. ” (ARD 2005: 41). Oloka-Onyango (2007) remarks that “the strategy of district proliferation has been adopted by President Museveni as a means of dispensing patronage, and ultimately of splintering challenges to the central government hegemony and control ” 1.10 The second trend was a progressive recentralization of competencies and key appointments. Centralization of some functions and the reduction of discretion at local level were used to sterilize the effects of district proliferation, preventing the emergence of local politicians who could challenge the ruling elite. For example, since 2005, District Chief Administrative Officers have been appointed by central government instead of district service commissions with the aim of removing them from “pressure exerted by local politicians”. The 2010 Kampala City Bill placed Kampala under the administrative authority of the Central Government who appoints its executive director. Universal Secondary Education (USE) in Uganda was introduced in 2007 as a centrally managed service. In 2009, drug procurement was also centralized to "fight corruption." Finally, resource allocation for feeder roads has been managed by the central Uganda Road Fund since 2010. 3 1.11 The third was a progressive reduction of own revenues for Local Governments. In 2005 the Graduated Tax or G-Tax, the predominant source of local revenue, was suspended in the run up to 2006 national and local elections, as its abolition was proposed by opposition and government parties as part of their manifestos. The Graduated Tax provided Local Governments with 5 percent of their total revenues, was important for discretionary expenditures and has never been adequately replaced, as all taxes account now for less than 5 percent of Local Governments’ revenues2 whose collection is estimated to be at less than half of its potential (LGFC:2012). Local governments have become increasingly dependent on non-discretionary central government transfers, reducing accountability to local citizens. 1.12 The fourth trend was a decline in transfers from central to Local Governments, in relative and per capita terms. Transfers from central government to Local Governments fell from 4. 9 percent of GNI and 33 percent of public expenditure in FY01 to 3. 9 percent of GNI and 19 percent of public expenditure in FY10. The latest PER (World Bank: 2013, p. 20) found that transfers to local governments continued to decline as a share of central government revenues and expenditures in FY11 and FY12. In FY12, Local Governments spent almost 60 percent of their resources on non-discretionary expenditures like wages and salaries. In particular, unconditional and non-wage recurrent conditional grants, both of which are critical to routine supervision, oversight and management of services, declined in real terms over the period FY01-FY11 (see LGFC: 2012, p. 24). As discussed in the 2013 PER (World Bank:2013, p. 22), real per capita recurrent budgets of Local Governments have trended downwards since FY04 from a peak of USh 49,213 to USh 38,838 in FY12 (2011/12 shillings, GDP deflated), representing a drop of about 21 percent. 1.13 The growth in unconditional grants has not been consistent with the Constitutional provision (Seventh schedule) requiring that “for a given fiscal year the unconditional grant shall be equal to the amount paid to local governments in the preceding fiscal year for the same items adjusted for general price changes plus or minus the budgeted cost of running added or subtracted services”. If the constitutional formula had been followed, the allocation would have been at least one third higher in FY10, according to estimates by the Local Government Finance Commission. 1.14 The fifth trend was an increase in earmarking of funds provided by donors and government. Tied sector conditional grants have increased as a percentage of total grants, from 72 percent in FY97 to over 86 percent in FY11 and a projected 88 percent in FY12, making it difficult for Local Governments to plan for and respond to local demand and needs. Donors contributed to the increase of conditional grants by increasing sector funding. Conditional grants were first introduced in the late 1990’s through the Poverty Action Fund, growing steadily to 26 in 2002 and 38 today, each with its own formula, and eligibility rules, as discussed in Annex B. In 2005, after the launch of the Local Government Sector Investment Plan (LoGSIP), donors 2 See: (1) LGFC:2012, p. 6. “LG own source revenues have plummeted to below 3% contribution to district budgets following the scrapping of graduate tax in the 2005/06 FY.” (2) Ministry of Finance, Planning and Economic Development, Background to the Budget FY2012/13, Consolidated Local Government Financial Operations6, 2005/06 - 2009/10 (billion shillings), page A:37. (3) World Bank:2013, p. 8:” In the early years of decentralization, own-source revenues accounted for 8 to 10 percent of total local government revenues, whereas in recent years this figure has dipped to 5 percent or less.” 4 established a common basket fund to support LoGSIP, while in other sectors (e. g. , Education) vertical planning tools were established, reducing Local Governments’ discretionary powers. 2. Objectives, Design, and Relevance Objectives 2.1 The PAD identified the objective of the Second Local Government Development Project (LGDP II) as “to improve the Local Government institutional performance for sustainable, decentralized service delivery” (World Bank, 2003a: 2). The formulation of objectives in the Development Financing Agreement (DFA) was identical (World Bank, 2003b: 24). 2.2 The project targeted both higher Local Governments (District Councils) and lower Local Governments (Municipalities, Sub-counties, and Town Councils). The PAD (World Bank, 2003a: 19) clarified that the final goal was “sustainable and decentralized service delivery resulting from improved LG institutional performance”, with the latter being means to an end. Performance is defined as fiscal performance (as measured by the percentage of own revenues over total revenues), and budget performance, both in terms of mid-term planning (measured through the use of three-year rolling Development Plans), and reporting (measured as timely submission of accounts to the Auditor General’s Office). 2.3 The development objective was not revised over the life of the project. Relevance of Objectives 2.4 The relevance of objectives was substantial. At the time of appraisal, the first phase of decentralization, with the establishment of an enabling legislation, a conducive intergovernmental fiscal framework, and transfer of competencies to Local Governments, was well under way, and there was a need to support the central government and local authorities in its implementation. The focus on improved service delivery, particularly for the poor, was an important step towards achieving the millennium development goals on education, health, and access to water, given the predominant role of Local Governments in this respect. The project was therefore highly relevant and its objectives reflected the Government’s ones spelled out in the Poverty Eradication Action Plan (PEAP) 2004/5 – 2007/8, namely ensuring good governance, including improvement in public service delivery and decentralization. 2.5 These objectives are less consistent with the Government’s National Development Plan- NDP 2010/11-2014/15 (launched in 2010), whose theme is “Growth, Employment and Socio- Economic Transformation for Prosperity”. This theme has eight objectives, including three that were addressed by the project: improving stock and quality of economic infrastructure, increasing access to quality social services, and strengthening good governance, defense and security. However, the new NDP is focusing on the final outcome (improved service delivery) giving much less emphasis to strengthening decentralization per se (e. g., the word “decentralization” was mentioned 21 times in PEAP and only twice in NDP). Finally, the project’s objectives are consistent with the Bank’s Country Assistance Strategy FY2011 -2015 whose second objective is to enhance public infrastructure (including roads, water and sanitation, 5 and delivery of urban services), and third objective is to promote human capital development through improved access to and quality of education, and strengthened health care delivery. 2.6 However, the GOU de facto policy reversals have not influenced in any way the Ugandan Local Governments’ responsibilities in service delivery, enshrined in the Constitution. The Ugandan Constitution (Art 189, 3) in fact states that “district councils shall have responsibility for any functions and services not specified in the Sixth Schedule to this Constitution. ” Based on the Constitution, Uganda Local Governments are responsible for primary education, agriculture, water, sanitation, primary health care, and district and feeder roads. As noted in the Bank’s latest CAS (World Bank 2010:6), “basic social services are delivered by Local Governments; thus, weaknesses in decentralization affect service delivery. ” Design 2.7 The project had the following five components:  Component 1: Support for the Decentralization Process (appraisal estimate US$23. 8 million, actual cost US$33. 6 million). This component supported the following sub- activities: (i) Fiscal Decentralization Strategy (FDS) implementation and tools to support decentralization of the development budget; (ii) institutional structure surrounding decentralization; (iii) sector issues related to decentralization, (iv) formulating and implementing a national Local Government capacity building strategy; (v) strengthening Local Government human resource development function, (vi) audit of the accounts of lower Local Governments (LLGs)3; (vii) procurement capacity building at the higher Local Government (HLG)4, and (viii) local financial management.  Component 2: Local Development Service Delivery (appraisal estimate US$107. 5 million, actual cost US$118. 8 million). This component comprised a non-sectoral conditional grant (the Local Development Grant - LDG) distributed on a formula basis to all local authorities in Uganda for investment in local infrastructure in accordance with local needs as determined through local planning and budgeting processes. Access to such funding was determined by capacity, accountability, and performance conditionalities which were designed to incentivize improvements in sustainable service delivery at the local level.  Component 3: Local Government Capacity Building (appraisal estimate US$15 million, actual cost US$14. 7 million). This component provided capacity building grants to Local Governments aimed at increasing the quality of training and to help establish a national system of standardized training modules recognized by all stakeholders.  Component 4: Local Government Revenue Enhancement (appraisal estimate US$11. 9 million, actual cost US$8. 1 million). This component provided support to enhance the capacity of Local Governments to administer their own sources of revenue in order to 3 Municipalities, Sub-counties, and Town Councils. 4 District Councils. 6 achieve and sustain acceptable levels of service delivery. It had five sub-components namely: (i) strengthening local revenue policies and legislation, (ii) training of politicians and officials, (iii) strengthening local revenue systems, (iv) extension of property tax system, and (v) monitoring Local Government revenues.  Component 5: Support to Project Implementation (appraisal estimate US$6. 8 million, actual cost US$6. 5 million). This component supported the overall coordination, implementation, monitoring and evaluation and auditing of the project. 2.8 The project focused on financial and fiscal management priorities, but tried also to ensure a balanced sector distribution of investments, by assessing performance through an 85 percent minimum threshold for investments in Poverty Eradication Action Plan (PEAP) focal areas. Sector specific grants (outside the project) remained the predominant form of Local Government funding, affecting the discretion of Local Governments and their horizontal accountability. 2.9 Capacity building grants were linked to a pre-set menu of training activities, for non- sectoral generic training and skills development, and tried to ensure that sufficient resources were spent to train sub-county staff. There was no link between the transfer of financial resources and the role realignment of central ministries or the transfer of their staff to Local Governments. The project legitimately excluded any effort to build sectoral capacity at the local level. The project focused instead on non-sectoral development and capacity building grants. 2.10 The components were not revised over the life of the project. Relevance of Design 2.11 The relevance of design was modest. The project had a relatively weak Results Framework. While the causal chain between inputs (e. g. , technical assistance, training, unconditional and capacity building grants), and outputs (e. g. , improved Local Government capacity, improved social infrastructure) was clear, the outcomes identified were mostly process related (e. g. Local Governments with 3-year rolling plans, final accounts submitted on time to the Auditor General), and referred to outputs rather than to the project’s PDO (e. g. , improved Local Government institutional performance for sustainable decentralized service delivery). At the same time, the demand-driven nature of sub-projects did not allow the definition of credible service delivery indicators at the appraisal stage. Design of Monitoring and Evaluation 2.12 The monitoring framework was very sophisticated and captured well all data concerning the project including those relating to improved service delivery at the local level. The monitoring framework included: (a) the Local Government Information and Communications System (LOGICS) which was designed and piloted during the previous project; (b) LOGICS Plus, a financial information module which allowed Local Governments to report on expenditures and revenues; (c) annual Local Government performance assessments carried out by a team consisting of officials from the Ministry of Local Government, sector ministries and a private contractor for quality assurance; and (d) a "Value for money" audit. The lack of a systematic feedback loop from beneficiaries was a weakness in the M&E design. Feedback from 7 beneficiaries was in fact introduced late in the project through a survey to assess Beneficiary Participation and Accountability under LGDP II and to inform the follow-on project. 3. Implementation 3.1 The project was approved on May 29, 2003, became effective on October 29, 2003, and had its mid-term review in June 2005. LGDP II's initial closing date of June 30, 2007 was extended once, to December 31, 2007, mainly because of the long-standing constitutional requirement of obtaining parliamentary approval for any new credit. 3.2 Project’s costs and funding are summarized in Table 1 below. Project costs were 10 percent higher than planned, and the extra costs were funded through an appreciation of the XDR versus the US dollar, which increase the US dollar value of the IDA credit denominated in XDRs, and a contribution from Government, mostly in the form of local development grants, that was 44 percent higher than planned. The total costs at appraisal were US$165 million: US$125 million from IDA (of which US$75 million as grant), US$15 million from the Netherlands, US$14. 8 million from Government and Local Authorities, US$7. 5 million from Ireland, US$1. 8 million from the Danish International Development Assistance–DANIDA, and US$0. 3 million from the Austria Development Corporation. Actual costs were US$181. 7 million, of which US$135. 21 million were provided by IDA, US$21. 33 million by Government and Local Authorities, US$15 million by the Netherlands, US$7. 5 million by Ireland, US$2. 4 million by DANIDA, and US$0. 3 million by Austria. Table 1. Project Costs and Funding, Planned versus Actual (US$ million) Project Costs Actual/ Component Planned Actual Planned Component 1: Support for the Decentralization Process 22. 90 33. 61 1. 47 Component 2: Local Development Service Delivery 107. 50 118. 80 1. 11 Component 3: Local Government Capacity Building 15. 00 14. 66 0. 98 Component 4: Local Government Revenue Enhancement 11. 40 8. 15 0. 71 Component 5: Support to Project Implementation 6. 40 6. 52 1. 02 Price contingencies 1. 80 Total 165. 00 181. 74 1. 10 Funding Actual/ Financier Planned Actual Planned IDA 125. 00 135. 21 1. 08 Borrower 14. 80 21. 33 1. 44 Netherlands 15. 00 15. 00 1. 00 Ireland 7. 50 7. 50 1. 00 Denmark 2. 40 2. 40 1. 00 Austria 0. 30 0. 30 1. 00 Total 165. 00 181. 74 1. 10 Source: World Bank 2008 3.3 Implementation was negatively affected by the reduction in the discretionary powers of Local Governments, their weakened ability to collect local taxes and user fees, and the 8 proliferation of districts. Transfers to Local Governments (see Figure 1) and unconditional grants per capita, and as a share of public expenditures and GNI, declined over time while local revenues were being reduced. Almost all allocation decisions were made at the center and the improved planning capacity of Local Governments could therefore be used only in a limited way. The proliferation of districts, carried out by splitting existing districts into smaller units due mostly to political patronage, had the dual effect of reducing capacity in these districts, and increasing costs, as each new district would add capital and operating costs and take staff away from existing ones. The pool of unconditional grants declined as a share of total grants from 32 percent in FY01 to 10. 5 percent in FY11 and to less than 5 percent in FY13, according to the latest Public Expenditure Review (World Bank: 2013), and had to be divided among a greater number of districts, making the funds available to reward performance relatively less significant over time, while non-wage recurrent conditional grants declined in real terms over the period FY01-FY11. New districts were initially dramatically understaffed with often less than 10 percent of strategic posts filled (Table 9 in Annex B). 3.4 The effect of the splitting of districts on the sustainability of decentralization is well exemplified by the case of Bushenyi, one of the best performing districts in Uganda, with chairmen from NRM both before and after the splitting (see Annex B, Table 9). Before Bushenyi was split into five separate districts in 2010, the district had a population of 723,427 receiving government grants of UGX 42,621 per capita, and was 10th in the FY09 Health District League and 11th in the FY08 Education District League (out of 80 districts) with 91percent of strategic posts filled. The splitting of the district into five in 2010 led to a reduction of per capita grants even in nominal terms for 80 percent of the combined population of the five new districts, a reduction of staffing to 81 percent for what remained as Bushenyi and to 8 percent for the other four districts. While Bushenyi itself improved its FY11 Health District League position from 10th to 2nd, the remaining four new districts fell to 61st, 68th, 93rd, and 104th position out of 112 districts, notwithstanding their sanitation rank well above the national average. Figure 1.Transfers to Local Governments as a share of public expenditure and GNI (%, FY01-FY10) 35.0% 6.0% 30.0% 5.0% 25.0% 4.0% 20.0% 3.0% 15.0% 10.0% 2.0% 5.0% 1.0% 0.0% 0.0% FY01 FY03 FY04 FY05 FY07 FY08 FY09 FY10 Share of Public Expenditure (left axis) Share of GNI (right axis) Source: Background to the Budget, various years 9 3.5 M&E Implementation included: (a) monitoring of sub-project performance that allowed a comparison of sub-projects implemented under LGDP I and LGDP II; (b) data collection on physical benefits achieved in the three key services of health, education and water; (c) a Communities Beneficiaries’ Assessment carried out by the Uganda Bureau of Statistics (UBoS) with a quantitative module covering a sample of 1,500 Households from 150 communities, and a qualitative one covering key informant interviews with 30 higher Local Governments, 30 lower Local Governments, 30 private firms and 30 civil society organizations selected randomly from within the quantitative sample; and (d) a Value for Money (VFM) Analysis, covering a sample of 1,083 sub-projects out of the total of 8,204 subprojects financed under the project. 3.6 Fiduciary Issues. There were no financial management or procurement issues during the project’s implementation. The project contributed to a substantial enhancement of financial management and procurement capacity in Local Governments. 3.7 Safeguards Compliance. The project had an environmental category of B (partial assessment) which required the preparation of an Environmental Assessment and Social Management Framework (ESMF). The ICR (World Bank 2008:15) found that Local Governments followed adequate environmental screening, but did not make adequate provisions for implementing environmental mitigation measures during sub-project preparation. MoLG guidelines now require Local Governments to include funding for environmental mitigation measures in implementation contracts. The project established an environment and natural resource (ENR) management checklists for Local Governments and trained technical staff of higher Local Government in ENR management. 4. Achievement of the Objectives 4.1 The achievement of objectives will be measured first by considering achievement of PDOs, and then by considering additional evidence. The logic of the intervention was relatively simple. Improved decentralized, sustainable service delivery would be achieved by  Supporting appropriate strategies and institutions for decentralization (Component 1) and strengthening the capacity of Local Governments to provide better services, administer their resources, and raise more revenues (Components 3 and 4); and  providing direct financial support to Local Governments through non-sectoral conditional grants on a formula basis (see Annex B for details) taking into account their needs, capacity, accountability and performance (Component 2) Improve Local Government institutional performance OUTPUTS 4.2 The project’s support for the Decentralization Process achieved several outputs.  A Harmonized Participatory Planning Guide (HPPG) was provided to all levels of Local Governments (56 districts, 13 municipalities and all lower Local Governments). 10  An Urban Planning Guide was printed and distributed to 93 districts and 96 urban councils and a total of 189 (95 percent) of district and urban council officials were trained in the use of the guide.  The project supported the preparation of the principles for the revision of the Town and Country Planning Act.  Local Government Public Procurement and Disposal of Assets Regulations were developed, published and distributed to all Local Governments.  370 Local Government technical staff (Chief Administrative Officers, Chief Finance Officers, and heads of departments) and 314 members of higher Local Government contracts committees were trained.  A national Local Government assessment manual was developed, published, distributed and explained to all Local Government staff; and annual assessments were carried out with feedback to Local Governments.  Three joint annual reviews of decentralization (JARD) were carried out leading to the development of the Decentralization Policy Strategic Framework (DPSF) - a one stop reference document on Uganda decentralization policy and a ten year Local Government Sector Investment Plan (LoGSIP).  A total of 896 lower Local Governments were audited for FY2002/3 and FY2003/4.  A national Local Government capacity building strategy was developed, published and launched.  32 standardized generic training modules (GTM) were developed, published and distributed to all Local Governments and private training providers.  A total of 260,089 participants5 were trained under the project. The most relevant training topics were: (i) investment appraisal and project planning (19,853 participants); (ii) management and leadership skills (16,167 participants); (iii) development planning (12,896 participants); (iv) participatory planning (12,611 participants); (v) gender mainstreaming (12,621 participants); and (vi) financial management (10,734 participants).  District/Municipal resource pool (D/MRP) teams were institutionalized and became operational in all district/municipal Local Governments to assist technical planning committees (TPCs).  Over 100 private training service providers for Local Governments were registered by MoLG and oriented to decentralization and training of Ugandan Local Governments in various fields. 4.3 In addition, the project provided support for revenue raising capacity.  All higher Local Governments (districts and municipalities) and Town councils prepared and implemented annual local revenue enhancement plans.  Trade licensing reform to simplify and shorten the trade licensing procedure was rolled- out to 10 Local Governments.  A draft Business and Levies Bill was produced to rationalize, consolidate and repeal the existing six business and levies acts (trading licensing act, markets act; shoppers act, 5 Local Governments’ staff (137,609 participants), councilors (63,288 participants), private sector (4,620 participants), CSOs (3,238 participants), women and youth (2,654 participants) and farmers (2,519 participants). 11 liquor licensing act, enguli licensing act, and sugar sell act) into one simplified act which would promote business start-ups, growth and revenue generation.  734 Local Government officials were trained in local revenue enhancement and sensitized about the constitutional mandate and civic duties for citizens to pay taxes.  A revenue handbook and a cost/benefit assessment tool were prepared to assist Local Governments in appraising local revenue enhancement opportunities  Seed capital for local revenue enhancement was provided to 170 Local Governments (74 districts, 13 municipalities and 83 towns).  25 technical staff from weaker Local Governments were twined with strong Local Governments’ staff in revenue generation to have peer-to-peer learning on revenue mobilization.  The Local Government (Rating) Act 2005; Local Government (Rating) Regulations, 2006; a Property Rating Handbook, and a training module for property rates were produced and published.  Officials from 92 urban councils and 13 municipalities were trained in property rates taxation.  UGX 3. 8 billion were used to assist 60 districts, 13 municipalities and 85 towns to carry out valuation of properties and update or develop new valuation rolls which would be valid for 5 years after approval. OUTCOMES 4.4 Supporting appropriate strategies and institutions for decentralization. Given the numerous reversals since 2005, this outcome was not fully achieved, as key elements of the Fiscal Decentralization Strategy were not implemented6. By the end of LGDP II (December 2007), the project had financed the first three Joint Annual Reviews of Decentralization which have resulted in a number of policy reforms - revision of the Local Government Act, the Local Government Finance and Accounting Regulation, the Town and Country Planning Act, and the Local Government (Rating) Act). Also, these annual reviews led to the establishment of a Decentralization Sector Working Group and the development partners pooling their resources into a basket to support the implementation of the Local Government Sector Investment Plan. A Local Government Capacity Building Policy Committee chaired by the Ministry of Local Government was established to guide the process of Local Government capacity building. The Ministry of Local Government established an in-house Capacity Building Unit aimed at providing support to all Local Governments in the formulation of their own Capacity Building Needs Assessments and Capacity Building Plans. However, as discussed later in this section, the Fiscal Decentralization Strategy, supported under component 1, was not successful. 4.5 Strengthening the capacity of Local Governments to provide better services, administer their resources, and raise more revenues. The capacity of Local Governments to better administer their resources was marginally improved, while the capacity to raise more revenues weakened, due to changes in legislation. Their capacity to provide better services was instead legitimately not addressed by the project as its capacity building was non-sectoral by 6 The first sub-component of Component 1 focused on the “Support to the FDS Implementation and Tools to Support Decentralization of the Development Budget.” 12 design. By the end of the project (December 2007), all higher Local Governments had good quality three year development plans linked to the budget, even though the improvements achieved up to 2006 started to deteriorate during the last year of the project, due to the policy reversals started in 2005, including the abolition of two major sources of revenue for Local Government (i. e. , Graduated Tax and Rates by owner-occupiers). Without the introduction by law of a new source of revenue to replace the Graduated Tax, the support provided by the project through the revenue handbook and related training could not achieve tangible results. 4.6 The annual national assessment of Local Governments - with an inbuilt incentives and sanctions mechanism - analyzed the level of compliance of Local Governments with laws and regulations governing their operations. After improving up to 2006, when several policies were reversed as discussed earlier, Local Governments’ performance deteriorated dramatically in 2007 and 2008 (the LGDP project closed at the end of 2007), due to declining quality of record keeping, meager revenue collection, weak budgeting and planning, widespread use of force accounts against procurement regulations, and a more general lack of interest in the national assessment itself. For 2008, the assessment of minimum conditions and performance found that out of the then national total of 1105 Local Governments, only 84 earned rewards of which only 19 were districts. A total of 46 Local Governments were static while 931 were penalized on the basis of failure to meet minimum conditions. 4.7 Subsequent annual assessments (see Figure 2 for details) showed a marked improvement in 2009 and 2010, followed by deterioration in 2011. However, independence of the assessments had declined after having been fully internalized by MoLG, instead of being carried out by a team consisting of officials from MoLG, sector ministries and a private contractor for quality assurance. Figure 2. Share of Local Governments Receiving a Penalty Score (%, 2005-2011) Source: MOLG - Annual Assessment of Minimum Conditions and Performance Measures for Local Governments, Synthesis Report, various years 4.8 The project improved the degree of participation of beneficiaries in decision making before the policy reversals mentioned above reduced the discretionary resources to be allocated at the Local Government level. The Communities Beneficiaries’ Assessment carried out by the 13 Uganda Bureau of Statistics through a survey of 1,500 households in 150 communities found that 36 percent of the respondents had participated in identifying development activities under the project. Of those who did participate, 63 percent reported that their priorities did appear in the Local Government plans and thereby demonstrate the responsiveness of the LGDP II planning and prioritization processes. Furthermore, 82 percent of respondents who did participate go on to report that the projects they selected were actually delivered. 4.9 Such improvements, however, have not withstood the test of time, as the reduced local discretion over expenditure has reduced the incentive for participation, as highlighted in key Government studies7. As noted in particular in World Bank: 2013 (p. 9), there has been a lack of attention in the reform process to building downward accountability and civil society capacity to interact with local governments, among others. Local revenues declined dramatically after 2005. The 2006 abolition of the graduated tax or G-Tax, that represented 85 percent of the revenues of Local Governments, stifled own revenues. While efforts to increase local participation in the budgeting process at the lower levels of government were better synchronized and integrated with the national budget planning process as part of the FDS, the substantially reduced discretion of Local Governments over expenditures and lower funding levels have constrained participation8. The combination of the centralization of the appointment of the Chief Administrative Officer (CAO), the re-centralization of district procurement, and the suspension and later abolition of the G-Tax weakened local accountability and discretionary powers over expenditure. 4.10 The efficacy of the first PDO is therefore rated as Modest, as the performance of Local Governments was initially improved, before the significant drop in LG performance measured by the 2007 and 2008 National Assessments. Sustainable, decentralized service delivery OUTPUTS 4.11 The project provided direct financial support to Local Governments through non-sectoral conditional grants on a formula basis taking into account their needs, capacity, accountability and performance. A total of 31,832 sub-projects were implemented - the Local Development Grant were used by Local Governments to invest in roads, education , agriculture extension, health, water and sanitation, construction of public administration buildings , and solid waste management . Figure 3 below presents the distribution of sub-projects by sector. 7 See for example LGFC (2012), p. 6: “There are factors that continue to constrain participation among them: (a) the low level of financing for this process; LGs at all levels often do not have the financing to enable the levels of consultations on investment activities and to report back to communities; (b) the limited discretion means that LGs are constrained to implement priorities and objectives provided by sectors. This provides very limited opportunity for dialogues at LGs; and (c) delays in issuing budget guidelines and IPFs have often led to a reduction in the amount of time available for participation and dialogues for LGs.” 8 LGFC (2012) p. c. “Community Participation in planning and budgeting is low.” 14 Figure 3. Distribution of Local Development Grant Projects by Sector (n=31,832) Source: Project Monitoring Reports 4.12 The Value for Money Analysis showed an Overall Value for Money score for all sampled projects of 4. 99 out of 9, which is very close to rating Good (5 to 7 points = Good)9. Projects in the education sector were rated to have achieved the highest value for money (5. 68), followed by health with an average VFM score of 5. 14; water and sanitation with a score of 5. 12; administration with a score of 4. 90; roads with a score of 4. 59 and lastly production10 with a score of 4. 51. Infrastructure sub-projects received also the lowest rating in the Community Beneficiaries Assessment. The better ratings of health and education investments is due to the fact that under the project, Local Governments were asked to ensure completeness of investments in the education and health sector, in terms of other complementary infrastructure, to make them more functional. OUTCOMES 4.13 The project did achieve important improvements in service delivery, but attribution is difficult. Based on data collected by MoLG Implementation Unit, towards the end of the project’s period (2006), residents had to travel an average of around 4. 5 Km to access health centers (in 1999 they had to travel at least 5. 0 Km). Primary school students had to travel an 9 The Value for Money rating was on a 0-9 scale and equal to the sum of the ratings for economy, efficiency and effectiveness which were on a 0-3 scale each. 10 For example, construction of valley dams and animal water troughs, seed multiplication like cassava, vanilla and colonial coffee, exotic goats, fish farming, and beekeeping. 15 average distance of around 1. 2 Km to reach their schools (in 1999 the distance was 1. 4Km) and persons fetching water from safe water points had to travel around 0. 8Km (in 1999 the average distance was 1. 78Km). Such reductions in distance traveled increased access to vital basic services and impact poverty reduction. However, there were several projects funded through sector conditional grants running in the same districts at the same time, and it is therefore difficult to isolate the impact of the project. 4.14 However, the Fiscal Decentralization Strategy (FDS), supported by the project, did not achieve its main objectives as sustainability remained low. Sustainability, an explicit element of LDGP2’s PDO, was low as the inability of Local Governments to increase own revenues, and the subsequent shift from unconditional to conditional grants for almost all of the Local Government funding, reduced the ability of Local Governments to adapt the allocation of funds to meet the most urgent demands of the people they serve. 4.15 The 2002 FDS had set two broad objectives: a) to increase Local Government autonomy and widening participation in decision making, and (b) to improve the effectiveness of the Local Government programs through increased effectiveness, transparency and accountability in expenditures. The main strategy was to streamline the funds transfer mechanisms channeling grants through two systems; the Recurrent Transfer Budget (RTB) system and the Development Transfer Budget (DTB) system, to improve the balance between discretion and non-discretionary financing of Local Governments. The FDS was to lead to an overall reduction in the number of conditional grants while at the same time providing increased flexibility and the participation of Local Governments in decisions of allocations of these resources. The Development Transfer Budget system would have used the LGDP methodology. 4.16 As stated in a study of the Local Government Finance Commission (2012), the Fiscal Decentralization Strategy failed to achieve its twin objectives. Local government autonomy was reduced. The grant system was in fact not streamlined. The RTB/DTB mechanism was never adopted, while the relative importance and number of conditional grants (up to 89 percent and 38, respectively) grew substantially rather than declining. An increased flexibility in sector allocations was introduced in FY07 and abolished by Cabinet in FY09. There was substantial resistance from line ministries that limited progress in the early years of the project. For example the Ministry of Education, which accounted for the largest share of conditional grants, had become a net loser, falling back on its targets, and the Ministry of Water and Environment had witnessed a similar trend. 4.17 LGDP grants represented about 36 percent of development transfers to Local Governments between FY01 and FY07. The project effectively provided sector budget support. Funds were co-mingled with Government resources and transferred. LGDP grants were based on independent performance assessments that received strong attention at the local level. Such assessments were focused on compliance with the Local Government framework and the improved capabilities benefitted the management of all grants, including conditional ones. 4.18 In 2006 Sector Budget Support was abandoned by donors. Such interruption came at a time of declining own revenues, and had a strong impact on Local Government funding. While LGDP grants continued after the end of the project with Government funding, the assessment of performance was internalized by the Ministry of Local Government, and the relative importance 16 of unconditional grants declined. Alignment with the country’s intergovernmental fiscal system was only partial, as the LGDP formula was applied only for unconditional grants. 4.19 The project’s assumption that more capable and better organized Local Governments will improve basic service delivery, was not proven correct, as little attention was paid to local accountability systems, particularly at the lower Local Government level, where most of the funds are spent. There was also evidence that funds could be used locally as a patronage tool. Reinikka and Svenson (2004), for example, found that the decentralization of education funds in Uganda allowed rural elites to capture the funds to the detriment of local schools and used them instead to support local patronage networks. 4.20 Today, funding for service delivery is generally low and current policies do not allow local revenue generation and contributions from local communities, as exemplified by one of the health centers supported by the project – see Box 1. Financing gaps vary by district. LGFC (2012) estimated that districts’ revenues including grants would need to triple to meet national standards for local government services, given the funding gap currently estimated at 64 percent (World Bank:2013). The 2013 Public Expenditure Review also found that Local Governments’ real per capita expenditure on education falling from USh 26,000 before FY07 to about USh 20,560 in recent years. Health expenditures have been more stable, but also trending downwards. The PER (p. 35) concluded that district governments are less able to fund local services than in the past with a negative net fiscal impact between 2 and 10 percent of district per capita budgets. Figure 4. Wakitaka Health Unit in the Sub-Council Mafubira, Jinja Source: IEG Mission 17 Box 1. If you live in Wakitaka and are pregnant, remember to charge your phone every day Wakitaka is a village in the sub-county of Mafubira, part of the district of Jinja. The Wakitaka Health Unit, the only such unit in the sub-county, received project support to build a house for health staff working at the center (on the right in the above picture). The unit is headed by a young and energetic nurse and serves about 20-30 patients a day. The main problem is the sub-county council has not paid electricity bills for a total of UGX1 million and the power company has cut off power supply several months ago. The head nurse was able to mobilize support from NGOs to buy solar panels, but could not afford batteries. As a consequence, there is no power at night or when the sky is cloudy. As a result, pregnant women delivering at night are assisted under the light generated by mobile phones, and using tools sterilized during the day. Vaccination for children is usually offered once a week. As the refrigerator cannot be powered and the staff does not want to turn away mothers that might show up, doses are picked up in Jinja in the morning and unused ones are returned late in the afternoon, with a trip of about 25 km back and forth. Source: PPAR TEAM 4.21 The efficacy of the second PDO is therefore rated as Modest, compared to substantial in the ICR. The achievements of the project in service delivery were not sustainable and are being reversed due to the new Government attitude towards decentralization. The entire intervention logic rested on the assumption that service delivery would improve by moving decision making closer to the beneficiaries. This did not happen, as the discretion over expenditure was reduced rather than increased. Lack of funds for operations and maintenance has subsequently led to the progressive erosion of the initial improvements in quality of management, and infrastructure. According to LGFC (2012), there is currently a 54 percent funding gap for the management functions of planning, supervision and monitoring of service delivery as well as auditing and infrastructure maintenance that does not allow Councils to execute their role in this respect. 5. Efficiency 5.1 Neither the PAD nor the ICR calculated an ERR/FRR on the grounds that benefits would be difficult to quantify. There are, however, a number of observations that can contribute to an overall assessment of efficiency. 5.2 Cost Effectiveness. The ICR, however, noted that the sub-projects implemented under LGDP II were more cost effective compared to those under the LGDP, even if comparable in size. It states that the average cost of the sub-projects financed under the project was 88 percent of the average cost of sub-projects implemented under the previous project (even without accounting for inflation) except for education and health sub-projects which were higher (110 percent). This is because under the project, Local Governments were asked to ensure completeness of investments in the education and health sector, in terms of other complementary infrastructure, to make them more functional. 5.3 Operational Efficiency. Economy11 and Efficiency12 of LGDP II’s sub-projects, accounting for two thirds of actual project costs, were also relatively good, as the Value for 11 Economy in the VFM analysis referred to minimizing the cost of resources used for an activity, with due regard for appropriate quality - and is audited in terms of deviations between budgeted cost, contract price; actual cost or market cost and defect cost. “Economy” was rated as: Best (3); Good (2); Fair (1); or Poor (0). 18 Money Analysis, carried out by an independent consultancy firm, rated the economy of subprojects it analyzed at 1. 7 and their efficiency at 1. 62 out of 3 (the minimum rating for good is 1. 67). The project was carried out within the timeframe envisaged at design, as the revision of the closing date was due to the time needed for Parliamentary approval, which was relatively fast. Costs for project management and coordination were low at 4 percent of total project costs, and there was a relatively small cost overrun (6. 25 percent). 5.4 The above findings lead IEG to rate the project’s efficiency as Substantial. 6. Ratings Outcome 6.1 Several ratings (i. e. , relevance of objectives, and efficiency) were substantial, but the project’s efficacy and the relevance of its design were both modest. The operation supported the building of capacity at Local Governments and contributed with direct financial support for investments in social infrastructure through sub-projects, to visible improvements in basic service delivery that were not sustainable, due to GOU policy reversals during and after the life of the project. Due to its modest efficacy, IEG rates the outcome of the project as Moderately Unsatisfactory. Risk to Development Outcome 6.2 The risk to the institutional outcomes achieved is rated high. Political commitment has declined over time, leading to substantial policy reversals from 2005 onwards. It is likely that the current trend to centralize functions will over time reduce the progress made in building capacity and accountability of Local Governments. A positive development in this respect is the Cabinet’s decision (made in March 2013) to place a moratorium on the creation of districts and to withdraw from parliamentary consideration the pre-existing proposal for the creation of 25 new districts. 6.3 Patronage was the political economy factor that most influenced project performance. The proliferation of districts, as discussed earlier, was driven by patronage, and sterilized by centralization of control and reduction in local discretionary powers, after the introduction of multi-party politics. The swing in the ruling party attitude towards decentralization of powers and functions was difficult to foresee in 2003. The 2013 PER (World Bank: 2013, p. 37) estimated that the impact of the creation of new districts on public expenditures has been low so far because the number of vacant positions in district governments has been increasing along with the number of districts, and the real wage bill of district governments has been eroded by a recent outburst of inflation. However, it concluded that such impact is likely to increase in the future, making district proliferation a potential “time bomb.” 12 Efficiency in the VFM analysis referred to the investment processes and was defined in five stages: planning and approval process; procurement process; implementation process; financial management; and commissioning /handover process. Efficient execution of a sub-project (“efficiency”) was rated as: Best (3); Good (2); Fair (1); or Poor (0). 19 6.4 At the same time, a political economy factor to be considered is represented by the preferences and modes of operations of donors. Sector approaches can often collide with decentralization. Bank Performance QUALITY AT ENTRY 6.5 The design was aligned with the PRSP and the Bank's CAS, and remains relevant to the latest CAS. The design drew on lessons from LGDP I as well as from the Bank and other donors in Uganda and neighboring countries. The design also draws appropriately on the Bank’s Analytical and Advisory Assistance, although the PAD included no reference to any study made outside the Bank. The Bank selected an appropriate instrument, which was part of a long history of support to decentralization dating back for at least a decade. However, the results framework was focused on process (e. g. , development of 3-year rolling development plans, final accounts submitted on time to the office of the Auditor General) rather than outcome indicators and the risk of policy reversal was set as moderate, and therefore partly underestimated, representing moderate shortcomings in the project design, while the M&E system was strong. 6.6 As noted in the ICR (World Bank 2008:10), project design and implementation were informed by the MTR recommendations and the ICR findings and lessons learned for LGDP I. In particular, in line with the MTR recommendations, the project was designed as an instrument to support the Government’s Fiscal Decentralization Strategy, leading the way towards a more permanent intergovernmental fiscal transfer system, and to mobilize broad grant support to implement decentralization reform with a possible strong demonstration effect for the future in terms of development partner coordination and harmonization. 6.7 The entry point was provided by the approval by Cabinet of the Fiscal Decentralization Strategy in June 2002. As the project supported a revamping of the intergovernmental fiscal transfer system, it was not selective in terms of jurisdictions or communities it sought to support. The project covered all higher Local Governments, and tried to address the issue of Local Government funding in a systematic way, trying to replace conditional grants funded through Uganda’s Poverty Action Fund with the LGDP approach, that was supposed to be mainstreamed for development transfers to Local Governments. 6.8 The Bank sought to use IDA resources to help develop a rules-based, transparent intergovernmental system (i. e., beyond simply financing local investments)13, but failed to achieve this objective, even though most local investments were completed successfully. 6.9 Considering the relevant but ambitious design, Quality at Entry is rated Moderately Satisfactory. 13 See PAD (p.6): “Through the FDS, the Government has made the strategic choice to mainstream the LGDP approach to fiscal decentralization, emphasizing improvement in decentralization of public service delivery through greater autonomy and downward accountability of local governments. LGDPII directly supports this strategic choice of Government, and other donors are increasingly supportive of this approach.” 20 QUALITY OF SUPERVISION 6.10 There were ten supervision missions with an input of 152 staff weeks over a 5-year period. The project had one Task Team Leader for its duration, from preparation to closing with a second one appointed only at the time of the ICR. ISRs always rated project performance satisfactory, even after the policy reversals of 2005-2006, a sign that their relevance might have been underestimated by the Bank team. There was good donor coordination, with several donors funding LGDP II directly. However, the importance of direct funding of Local Governments by donors through conditional grants grew over time and undermined some of the project’s achievements in terms of donor coordination and harmonization. 6.11 The Bank was quick to adapt to the Government’s policy reversal and tried to react to the abolition of the Graduated Tax by (a) highlighting the need for adequate compensation of Local Governments by Central Government for the loss of the Graduated Tax revenue; (b) preparing a note which assisted government to identify alternative own revenue sources to replace the Graduated Tax; and (c) supporting the Government request to use part of the credit to facilitate Local Governments to improve their revenue collection efficiency from their own revenue sources. However, a stronger reaction to the policy reversals and a growth of unconditional grants below constitutional requirements (as discussed in para 1. 13) could have been possible considering also the related PRSC operations running at the same time. Overall, the Bank Supervision Performance was Moderately Satisfactory. 6.12 Taking account of both Quality at Entry and Supervision performance, the Bank‘s Overall Performance was also Moderately Satisfactory. Borrower Performance GOVERNMENT PERFORMANCE 6.13 The Government’s performance was Unsatisfactory. Prior to the introduction of a multi- party system in 2005, the GOU had shown strong commitment to decentralization and the project. However, since 2005, commitment has been declining, leading to a substantial reduction in Local Governments’ mandates, funding, and autonomy, even though the recent Cabinet decision to declare a moratorium on the creation of new districts is an encouraging sign of a more conducive environment for decentralization in the near future. IMPLEMENTING AGENCY PERFORMANCE 6.14 The MoLG’s performance was Satisfactory. It led the process of reform and tried to adapt to an increasingly challenging environment. The Ministry ensured that the grants were transferred on a timely basis to Local Governments, funds were not diverted and the necessary policy reforms such as the restructuring of Local Governments, identification of new taxes for Local Governments, support for urban planning, and support to update Local Governments’ valuation rolls were achieved. 6.15 Considering the importance of the Government’s policy reversals and the moderately unsatisfactory outcome rating, IEG rates Borrower Performance as Moderately Unsatisfactory. 21 Quality of M&E 6.16 M&E Design. As discussed in the section on relevance of design, the monitoring framework was very sophisticated and captured well all data concerning the project including those relating to improved service delivery at the local level. However, the M&E design did not adequately address the multiplicity of monitoring systems14, the need for independent impact assessments and a systematic beneficiary feedback loop. 6.17 M&E Implementation. As discussed in the section on implementation, there was substantial monitoring of sub-project performance, data collection on physical benefits achieved in the three key services of health, education and access to water; a Communities Beneficiaries’ Assessment (although only late in the project); and a Value for Money (VFM) Analysis, covering a sample of sub-projects. 6.18 M&E Utilization. M&E information was used to analyze the performance of Local Governments and determine their eligibility for grant funding under Component 2. It also allowed a real time monitoring of project performance. LOGICS, however, did not replace existing monitoring systems at the sector level but drew from them, leading at times to duplication of efforts and poor reporting. LOGICS is not effectively used by Local Governments, and suffers from, inter alia, inadequate funding, poor ICT infrastructure, unreliable power supply, limited computer literacy, and multiplicity of information systems in Local Governments. 15 6.19 Considering its sophisticated design, and good implementation and use, partially counterbalanced by the lack of integration of LOGICS with other monitoring systems, and limited beneficiary feedback loop, the overall quality of the project’s M&E system is rated as Significant. 7. Lessons 7.1 The first lesson is that policy reversals can cause serious damage to otherwise significant project outcomes, and are difficult to identify and counter. Although sometimes harmful, policy reversals are not uncommon and fall under the sovereign right of nations to determine their own policies. With local government development projects early warning signals should include any significant increase in the number of districts, or reduction in un-earmarked funding or Local Governments’ rights to raise revenues. When early warning signals of major policy reversals appear, particularly if motivated by political patronage, the Bank should consider whether withdrawal of support for decentralization could lead to better results than trying to continue to work with Local Governments under the new policy environment. 7.2 The second lesson is that monitoring should be focused on outcome in addition to process indicators, and it should be unified across sectors. The project achieved an impressive series of 14 This issue was highlighted in the Mid-Term Review (p. 13):” There is a need to consider merging MIS systems that were operating in LGs e.g. LOGICS, IFMS, LOGFIAS, NIMES, FDS.” 15 See for example http://makir.mak.ac.ug/handle/10570/2284 . 22 outputs at the local level, but was not sustainable. As the vision pursued under the Fiscal Decentralization Strategy was abandoned, the project continued producing outputs that could not be adequately maintained over time, rather than being suspended or restructured. The parallel functioning of the project’s M&E systems and those of other sector programs within the same Local Governments should be avoided as it duplicates efforts and makes attribution more difficult. 7.3 The third lesson is that decentralization is not a sector, while it was treated as such in Uganda with a Sector Working Group, a Sector Investment Plan and specific donor support. Decentralization of service delivery affects all sub-national units and all sectors of the economy and should be supported in a harmonized way across sectors and donor programs. Resistance by line ministries and donors alike, worried about having control of the achievement of key results, undermined Local Governments leading to a significant missed opportunity. Had decentralization shown stronger results by 2006, would it have been so easily reversed? 7.4 The fourth lesson is that a fully decentralized sector allocation, supported through government budgets, requires a change in the way donors allocate funds across sectors within a given country, often using conditional grants to reach Local Governments. If the choice of sectors is truly delegated to Local Governments and communities, such allocation cannot be determined a priori. This is often incompatible with the development cooperation frameworks of bilateral donors, or the sector-specific teams and earmarked funds of multilaterals, that work in tandem with line ministries in developing countries like Uganda. 23 References Azfar, Omar, Livingston, Jeffrey and Meagher, Patrick. 2002. Decentralization in Uganda. ECO Consult. 2009. Country Level Evaluation Uganda, Brussels: European Commission. Green, Elliott. 2008. District Creation and Decentralization in Uganda. Development Studies Institute. London: London School of Economics IEG. 2005. Capacity Building in Africa. An OED Evaluation of World Bank Support. Washington, DC: World Bank. _____. 2008. Decentralization in Client Countries. An Evaluation of World Bank Support, 1990–2007. Washington, DC: World Bank. _____ and OPEV. 2009. Uganda – Joint IEG/OPEV Country Evaluation Assistance 2001-2007. LGFC - Local Government Finance Commission. 2012. Review of Local Government Financing in Uganda, Financing Management and Accountability for Decentralized Service Delivery, Draft. Manyak, Terrell G. , and Katono, Isaac W. 2010. “Decentralization and Conflict in Uganda: Governance Adrift”. African Studies Quarterly , Volume 11, Issue 4, Summer 2010. Ministry of Local Government. 2010. Ministerial Policy Statement for Financial Year 2011/2012. Kampala, Uganda. _____. 2012. Ministerial Policy Statement for Financial Year 2012/2013 . Kampala, Uganda. ODI. 2005. Republic of Uganda. Local Government Public Financial Management Assessment. Okidi, John and Guloba, Madina. 2006. Decentralization and Development: Emerging Issues from Uganda’s Experience. Economic Policy Research Center, Kampala, Uganda: Makerere University. Reinikka, Ritva and Svensson, J. 2004. “Local Capture: Evidence from a Central Government Transfer Program in Uganda. ” Quarterly Journal of Economics, 119(2), 678-704. Singiza, Douglas K. , and DeVisser, Jaap. 2010, Chewing more than one can swallow: the creation of new districts in Uganda. Steffensen, Jesper. 2010. Sector Budget Support in Practice, Desk Study, Local Government Sector in Uganda. Mokoro for ODI. _____, Tidemand, Per, Naitore, Harriet, Ssewankambo, Emmanuel and Mwaipopo, Eke. 2004. A Comparative Analysis of Decentralisation in Kenya, Tanzania and Uganda . USAID. 2010. Comparative Assessment of Decentralization in Africa: Uganda Desk Study. Washington, DC: World Bank. World Bank. 2003a. Second Local Government Development Project, Project Appraisal Document. Washington, DC: World Bank. _____. 2003b. Development Financing Agreement, Second Local Government Development Project. Washington, DC: World Bank. _____. 2005. Public Financial Management Performance Report and Update of CIFA Action Plan 2005 _____. 2008. Second Local Government Development Project, Implementation Completion and Results Report. Washington, DC: World Bank. _____. 2010. Country Assistance Strategy for the Republic of Uganda for the Period FY 2011-2015. Washington, DC: World Bank. _____. 2013. Service Delivery with More Districts in Uganda: Fiscal Challenges and Opportunities for Reforms. Public Expenditure Review. Washington, DC: World Bank. 24 ANNEX A Annex A. Basic Data Sheet THE REPUBLIC OF UGANDA: SECOND LOCAL GOVERNMENT DEVELOPMENT PROJECT (LOAN IDA – 3770, AND 3773A, AND GRANT IDA H0410) Key Project Data (amounts in US$ million) Appraisal Actual or Actual as % of estimate current estimate appraisal estimate Total project costs 165. 0 181. 7 110% Loan and grant amount 125. 0 135. 2 108% Cofinancing 40. 0 46. 5 116% Cancellation - - - Cumulative Estimated and Actual Disbursements (amounts in US$ million) FY04 FY05 FY06 FY07 FY08 Appraisal estimate (US$M) 34. 1 73. 9 114. 0 125. 0 125. 0 Actual (US$M) 31. 6 80. 4 108. 0 135. 0 135. 2 Actual as % of appraisal 93% 109% 95% 108% 108% Project Dates Original Actual Initiating memorandum 06/10/2002 06/10/2002 Negotiations 04/07/2003 04/10/2003 Board approval 05/29/2003 05/29/2003 Signing 06/20/2003 Effectiveness 08/31/2003 10/29/2003 Closing date 06/30/2007 12/31/2007 ANNEX B 25 Staff Inputs (staff weeks) Staff Time and Cost (Bank Budget Only) Stage of Project Cycle US$ thousands (including No. of staff weeks travel and consultant costs) LENDING FY00 FY01 FY02 1 25. 74 FY03 45 252. 75 Total 46 279. 49 SUPERVISION/ICR FY04 33 181. 36 FY05 36 227. 45 FY06 35 133. 15 FY07 26 84. 07 FY08 22 66. 49 Total: 152 692. 52 Mission Data Staff Date No. of days Specializations Performance Rating Types of (month/year) persons in field represented rating trend problems Identification/ Preparation Appraisal Supervision 7/2003- S S 11/2007 Completion S S Other Project Data Borrower/Executing Agency: Follow-on Operations Operation Credit no. Amount Board date (US$ million) Local Government Management and Services IDA-43720 55 Dec. 8, 2007 Delivery Project 26 ANNEX B Annex B. Local Governments in Uganda – Overview and Timeline The system of Local Government in Uganda is a five-tiered one, consisting of (i) districts, (ii) counties, (iii) sub-counties, (iv) parishes, and (v) villages (rural) or wards (urban). Only districts, cities, municipalities and sub-counties are legal persons with capacity to sue and be sued, and vested with executive and legislative powers. The Constitution of Uganda states that the district is the basis of the Local Government system and that it is a unit under which other lower Local Government and administrative structures operate. The autonomy of Local Government, in other words, finds expression in the powers and functions of districts as the main unit of local governance. Districts are centers of political and administrative power. The Constitution provides for a district chairperson as the political head of a district. The Local Government Act (LGA) confers the title of “mayor” on the district chairperson. The chairperson is directly elected by universal adult suffrage. Under Part I of the Second Schedule to the LGA districts are mandated to perform a range of functions. They provide education services as well as medical and health services by running hospitals. They are furthermore tasked with the provision of maternity and child welfare services and the control of communicable diseases such as HIV/AIDS, leprosy and tuberculosis. Districts must control the spread of diseases and provide rural ambulance and primary health care services. Hence districts are important structures in the public health sector. In addition, districts provide water services, including the maintenance of water supplies in liaison with the relevant line ministry. Finally, districts are required to provide road services (Singiza 2010). Figure 5. Number of districts in Uganda (1962-2011) Growth of the number of districts in Uganda 111 80 56 33 38 17 19 1962 1971 1979 1991 2000 2007 2011 Source: PPAR TEAM Figure 5 shows that the number of districts increased by almost 50% during the life of the project, and grew by over one third after the project closed. Table 2 shows instead that all types of Local Government grew by almost 50 percent between 2004 and 2011. ANNEX B 27 Table 2. Growth in Local Government (LG) and Administrative (AD) Units by Level % Level 2004 2011 change District Councils (including Kampala City Council) (LC 5) LG 56 112 100% County Councils (LC4) AD 151 164 9% Municipal Councils 13 22 69% City Divisions 5 5 0% Sub-county Councils (LC3) LG 857 1,116 30% Municipal Divisions 34 56 65% Town Councils 69 174 152% Parishes (including city wards) (LC2) AD 5,225 7,138 37% Villages (LC1) AD 44,402 66,036 49% Total 50,812 74,823 47% Source: Local Government Commission (2011) Table 3 shows the distribution of resources among wages, recurrent and development expenditures. Local Governments are defined here as Districts, Municipalities and Town Councils, as the Administrative Units receive resources from the respective councils or divisions. It is clear that Local Governments account for a large share of the wage bill but have proportionally less resources for development and recurrent costs. Table 3. Central Government Fiscal Operations (UGX billion, outturns) FY12 % UGX billion FY09 FY10 FY11 (Projected) cumulative Government Budget 4,499 5,898 9,863 7,744 100% Wage bill 1,185 1,308 1,659 1,910 22% Recurrent 2,107 2,999 4,314 3,695 47% Development 1,207 1,591 3,890 2,139 32% Local Government Budget 1,150 1,299 1,506 1,602 100% Wage bill 664 707 914 940 58% Recurrent 268 276 237 295 19% Development 218 316 355 367 23% Share of LG Wage bill 56% 54% 55% 49% Recurrent 13% 9% 5% 8% Development 18% 20% 9% 17% Source: The Background to the Budget - 2012/13 Fiscal Year Table 4 shows the functional distribution of Local Government expenditure with a preponderance of education (44 percent), health (14 percent) and general administration (23 percent) accounting for 91 percent of the total. Table 4. Functional Classification of Local Governments' Expenditure UGX billion FY09 FY10 FY11 FY12 Share 28 ANNEX B (Approved) General Public Administration 267 302 284 257 23% Public Order and Safety Affairs 4 3 3 3 0% Education 455 487 539 651 44% Health 136 201 167 175 14% Community and Social Services 25 47 33 21 3% Water 21 22 30 24 2% Economic Affairs and Services 0% Agriculture 84 129 118 80 9% Roads 39 66 52 32 4% Other economic services 7 8 19 14 1% Total 1,038 1,265 1,245 1,257 100% Source: Uganda Bureau of Statistics The consolidated financial operations by Local Governments, as shown in table, show a preponderance of grants. Such grants, as shown in Table 5, are mostly conditional, provided by various line ministries on the basis of different formulas, some extremely complicated. There are currently 38 conditional grants given to each Local Government on an annual basis. Table 5. Consolidated Local Government Financial Operations (FY06-FY10, UGX Billion) UGX billion FY06 FY07 FY08 FY09 FY10 Revenue 1,018. 1 1,075. 7 1,143. 2 1,265. 3 1,567. 0 Taxes 49. 8 35. 7 42. 7 16. 3 25. 7 Grants 931. 9 1011 1,064. 2 1,211. 2 1,496. 4 Other revenues 36. 4 29 36. 3 37. 8 44. 9 Expenses 932. 1 987. 3 1,029. 8 1,177. 5 1,440. 5 Compensation of employees 478. 5 567. 3 619. 8 644. 5 709. 9 Use of goods and services 396. 1 319 238. 5 289. 7 295. 1 Consumption of fixed capital 1. 1 0. 4 0. 4 0 0. 5 Interest 0 0. 4 0 5. 5 0. 1 Subsidies 0 0 Grants 55 96. 7 163. 5 228. 7 426. 5 Social benefits 1. 4 2. 2 2. 7 5. 3 6. 1 Other expense 0 1. 3 4. 9 3. 8 2. 3 Net Operating balance 86. 0 88. 4 113. 4 87. 8 126. 5 Grants/Revenue 92% 94% 93% 96% 95% Grants/Expense 100% 102% 103% 103% 104% Source: The Background to the Budget - 2012/13 Fiscal Year Districts are heavily dependent on government grants representing 95 percent of revenues. District performance, assessed annually to determine allocations of unconditional grants (a small fraction of the total), seems to be extremely variable as show in Table 6. The National Assessment Exercise is carried out once a year to measure the extent to which a Local ANNEX B 29 Government meets a series of set minimum conditions and performance measures which are derived from applicable policies, laws, regulations and guidelines. Among others, financial management, procurement and tendering are extensively examined during the assessment. Local Governments not meeting such minimum conditions are penalized with a smaller allocation of unconditional grants, while those exceeding them are rewarded. Table 6. Distribution of Districts and Municipal Councils by Performance as assessed by MoLG Status 2005 2006 2008 2009 2010 2011 Penalty 26 19 59 63 17 22 Static 16 19 18 7 67 30 Reward 14 27 16 22 50 81 None 1 1 1 Total 56 65 93 93 134 134 of which Districts (including Kampala City Council) 80 80 112 112 Municipal Councils 13 13 22 22 Source: National Assessment 2008 and MoLG Policy Statements 2010, 2011 and 2012 Table 7. Composition of Grants to Local Governments (F06-FY12, UGX billion) Approved estimates of revenues and expenditures Outturn Budget Grants FY06 FY07 FY08 FY09 FY10 FY11 FY12 Share Unconditional Grants 98. 40 94. 30 132. 00 133. 60 144. 60 157. 20 192. 32 11% Conditional 735. 1,004. 1,145. 1,212. 1,459. Grants 80 840. 60 914. 50 50 90 31 50 86% Equalization Grants 3. 50 3. 50 3. 50 3. 50 3. 50 3. 14 3. 49 0% Graduated Tax Compensation 34. 80 45. 00 12. 00 32. 00 45. 00 44. 46 - 3% 872. 1,173. 1,339. 1,417. 1,655. Total Grants 50 983. 40 1,062. 00 60 00 11 31 100% Source: The Background to the Budget – 2012/13 Fiscal Year Notwithstanding the efforts of the project of increasing unconditional grants, conditional (earmarked) grants still represent the vast majority of the decreasing grants to Local Governments. The combination of declining Local Governments’ tax revenues and the increase in earmarked funding from the center has effectively reduced Local Governments’ autonomy in the country. The progressive fractioning of districts through the constant creation of new ones make historical comparison of financial flows and performance at the district level difficult. Table 9 has been assembled as part of this PPAR by combining available information on the territorial origin of new districts with financial data. Districts and municipal councils are the 30 ANNEX B highest tier of sub-national governments in Uganda, and, as shown in Table 8 from Green (2008), among the smallest in Sub-Saharan Africa. It should be noted that since 2008, the average population per district in Uganda has fallen from 383,071 to 216,315. Table 8. Sub-National Administrative Units (Highest Level) for Sub-Saharan Africa countries with a population of more than twenty million, by number Source: CIA World Factbook, US Census Bureau - from Green (2008) How are grants determined? Unconditional Grants. Between 1993 and 1998, unconditional grants were distributed based on a formula that considered child mortality (40 percent), school age population (40 percent), general population (10 percent), and area (10 percent). In 1998 the formula, criticized for being too complicated was replaced on a formula giving an 85 percent weight to population and 15 percent to land area, after a constant of UGX 160 million per district had been taken into consideration to pay for the administrative structures of districts that were essentially the same. 85 percent of the difference between the total available for unconditional grants and the constant times the number of district was then allocated based on population and 15 percent on land area. The constant was later removed from the formula. 31 ANNEX B Table 9. Evolution of Districts in Uganda (2001 – 2011) 2001 2007 2011 Posts Filled Posts Filled Population- Population- Population- % Strategic % Strategic Table FY09 Table FY11 FY09 grant FY11 grant Sanitation Per capita Per capita Census Census Census League League District District District No. Health Health (UGX) (UGX) Rank 2002 2002 2002 No. No. 51. 1 Adjumani 202,290 1 Adjumani 202,290 34,075 27% 1 Adjumani 202,290 41,811 42. 70 38% 52 40 52. 2 Apac 415,578 44,974 64% 2 Apac 249,656 49,668 55. 30 62% 56 80 2 Apac 683,993 3 Oyam 268,415 47,993 56. 50 62% 68 60. 3 Oyam 268,415 44,644 27% 00 4 Kole 165,922 43,991 40. 40 8% 33 57. 4 Arua 559,075 32,296 82% 5 Arua 559,075 44,216 52. 70 77% 47 60 3 Arua 833,928 5 Koboko 129,148 54,519 45% 6 Koboko 129,148 48,557 49. 40 46% 26 42. 6 Nyadri 145,705 91,198 7 Nyadri 145,705 54,322 52. 20 8% 35 50 59. 8 Bugiri 266,944 41,949 52. 80 77% 50 4 Bugiri 412,395 7 Bugiri 412,395 37,850 73% 10 9 Namayingo 145,451 38,858 39. 70 8% 41 57. 10 Bundibugyo 158,909 51,929 52. 70 46% 45 5 Bundibugyo 209,978 8 Bundibugyo 209,978 43,867 45% 60 11 Ntoroko 51,069 50,070 22. 90 8% 35 12 Bushenyi 205,671 41,839 74. 20 85% 64 13 Buhweju 82,881 30,670 40. 10 8% 39 75. 6 Bushenyi 731,392 9 Bushenyi 731,392 42,157 91% 14 Mitoma 160,802 41,753 44. 70 8% 46 20 15 Rubirizi 101,804 34,507 53. 70 8% 50 16 Sheema 180,234 50,124 52. 50 8% 52 68. 7 Busia 225,008 10 Busia 225,008 48,038 45% 17 Busia 225,008 41,785 44. 70 54% 49 30 76. 11 Gulu 298,527 42,361 55% 18 Gulu 298,527 53,684 69. 00 77% 48 20 8 Gulu 475,260 19 Amuru 135,723 47,958 57. 60 38% 45 61. 12 Amuru 176,733 48,146 36% 80 20 Nwoya 41,010 92,343 55. 20 8% 43 ANNEX B 32 2001 2007 2011 Posts Filled Posts Filled Population- Population- Population- % Strategic % Strategic Table FY09 Table FY11 FY09 grant FY11 grant Sanitation Per capita Per capita Census Census Census League League District District District No. Health Health (UGX) (UGX) Rank 2002 2002 2002 No. No. 54. 9 Hoima 343,618 13 Hoima 343,618 36,893 64% 21 Hoima 343,618 25,555 48. 90 85% 43 90 55. 14 Iganga 540,999 41,492 73% 22 Iganga 355,473 40,020 59. 90 69% 51 50 10 Iganga 708,690 23 Namutumba 167,691 41,284 59. 40 38% 43 60. 15 Namutumba 167,691 41,272 27% 50 24 Luuka 185,526 31,225 41. 60 8% 36 77. 11 Jinja 387,573 16 Jinja 387,573 47,434 91% 25 Jinja 387,573 32,260 66. 70 92% 51 50 68. 12 Kabale 458,318 17 Kabale 458,318 42,724 55% 26 Kabale 458,318 46,188 59. 90 62% 48 80 62. 13 Kabarole 356,914 18 Kabarole 356,914 34,653 73% 27 Kabarole 356,914 30,921 73. 10 69% 51 60 51. 14 Kaberamaido 131,650 19 Kaberamaido 131,650 51,736 36% 28 Kaberamaido 131,650 52,039 65. 20 38% 41 40 62. 15 Kalangala 34,766 20 Kalangala 34,766 112,006 82% 29 Kalangala 34,766 91,497 51. 50 69% 53 50 82. 16 Kampala 1,189,142 21 Kampala 1,189,142 31,728 30 Kampala 1,189,142 2,123 77. 50 20 56. 22 Kamuli 361,399 50,534 45% 31 Kamuli 361,399 36,292 60. 90 46% 76 50 17 Kamuli 516,066 52. 23 Kaliro 154,667 43,771 27% 32 Kaliro 154,667 47,631 48. 00 38% 59 60 62. 33 Buyende 191,266 26,685 54. 80 8% 45 18 Kamwenge 454,996 24 Kamwenge 454,996 31,219 73% 70 34 Kamwenge 263,730 32,636 61. 20 77% 58 65. 19 Kanungu 204,732 25 Kanungu 204,732 47,550 73% 35 Kanungu 204,732 49,743 58. 80 69% 57 70 59. 26 Kapchorwa 141,439 66,043 73% 36 Kapchorwa 74,268 87,117 63. 60 69% 36 40 20 Kapchorwa 190,391 37 Bukwo 48,952 106,635 58. 90 46% 42 57. 27 Bukwo 48,952 97,810 36% 10 38 Kween 67,171 46,731 17. 90 8% 20 58. 21 Kasese 523,033 28 Kasese 523,033 43,299 82% 39 Kasese 523,033 33,185 57. 20 85% 46 60 ANNEX B 33 2001 2007 2011 Posts Filled Posts Filled Population- Population- Population- % Strategic % Strategic Table FY09 Table FY11 FY09 grant FY11 grant Sanitation Per capita Per capita Census Census Census League League District District District No. Health Health (UGX) (UGX) Rank 2002 2002 2002 No. No. 70. 29 Katakwi 118,928 65,073 91% 40 Katakwi 118,928 58,918 66. 40 85% 48 80 22 Katakwi 298,950 53. 30 Amuria 180,022 48,244 64% 41 Amuria 180,022 46,661 47. 00 62% 65 60 64. 23 Kayunga 294,613 31 Kayunga 294,613 41,712 73% 42 Kayunga 294,613 41,454 58. 30 85% 25 00 49. 24 Kibaale 405,882 32 Kibaale 405,882 37,353 91% 43 Kibaale 405,882 37,686 50. 90 85% 46 70 66. 44 Kiboga 108,897 54,529 51. 50 85% 59 25 Kiboga 229,472 33 Kiboga 229,472 45,130 91% 90 45 Kyankwanzi 120,575 36,989 46. 10 8% 49 64. 26 Kisoro 220,312 34 Kisoro 220,312 50,206 91% 46 Kisoro 220,312 51,686 63. 90 69% 40 30 67. 47 Kitgum 167,030 69,568 57. 20 69% 61 27 Kitgum 282,375 35 Kitgum 282,375 57,792 73% 60 48 Lamwo 115,345 56,067 47. 70 8% 58 57. 36 Kotido 122,442 43,890 64% 49 Kotido 122,442 52,539 49. 50 69% 23 20 41. 28 Kotido 377,102 37 Kaabong 202,757 41,187 55% 50 Kaabong 202,757 38,218 37. 00 54% 43 20 72. 38 Abim 51,903 112,884 36% 51 Abim 51,903 92,365 66. 40 38% 50 00 71. 39 Kumi 267,232 54,780 45% 52 Kumi 165,365 51,577 61. 80 46% 46 20 29 Kumi 389,665 53 Bukedea 122,433 48,688 56. 00 46% 49 65. 40 Bukedea 122,433 50,787 27% 50 54 Ngora 101,867 53,462 50. 80 8% 38 58. 55 Kyenjojo 266,246 32,744 53. 20 8% 46 30 Kyenjojo 377,171 41 Kyenjojo 377,171 35,286 64% 90 56 Kyegegwa 110,925 35,988 56. 20 8% 66 76. 42 Lira 515,666 39,937 64% 57 Lira 290,601 48,107 64. 20 77% 69 90 31 Lira 741,240 69. 43 Amolatar 96,189 60,256 9% 58 Amolatar 96,189 62,315 51. 30 15% 55 60 ANNEX B 34 2001 2007 2011 Posts Filled Posts Filled Population- Population- Population- % Strategic % Strategic Table FY09 Table FY11 FY09 grant FY11 grant Sanitation Per capita Per capita Census Census Census League League District District District No. Health Health (UGX) (UGX) Rank 2002 2002 2002 No. No. 59 Dokolo 129,385 54,419 58. 30 69% 48 65. 44 Dokolo 129,385 51,791 64% 60 Otuke 62,018 57,693 45. 80 8% 56 40 61 Alebtong 163,047 38,529 10. 00 8% 47 62. 45 Luwero 341,317 49,054 73% 62 Luwero 341,317 49,514 49. 60 77% 71 70 32 Luwero 478,595 68. 46 Nakaseke 137,278 56,710 91% 63 Nakaseke 137,278 53,905 56. 20 85% 55 20 64 Masaka 228,170 38,160 63. 00 85% 62 75. 65 Bukomansimbi 139,556 34,144 50. 20 8% 52 33 Masaka 770,662 47 Masaka 770,662 31,575 91% 40 66 Kalungu 160,684 40,415 53. 80 8% 69 67 Lwengo 242,252 30,910 41. 80 8% 63 62. 48 Masindi 396,127 39,354 73% 68 Masindi 208,420 37,679 54. 40 54% 54 60 34 Masindi 459,490 69 Buliisa 63,363 56,516 67. 00 62% 53 48. 49 Buliisa 63,363 65,433 45% 80 70 Kiryandongo 187,707 34,953 42. 90 8% 38 51. 35 Mayuge 324,674 50 Mayuge 324,674 35,534 45% 71 Mayuge 324,674 34,416 56. 10 62% 30 40 69. 51 Mbale 332,571 41,744 55% 72 Mbale 332,571 42,722 64. 40 62% 43 30 63. 36 Mbale 718,240 52 Manafwa 262,566 50,094 55% 73 Manafwa 262,566 53,244 59. 10 85% 41 70 62. 53 Bududa 123,103 61,063 73% 74 Bududa 123,103 54,881 62. 70 69% 48 60 75. 54 Mbarara 361,477 43,026 91% 75 Mbarara 361,477 38,478 70. 30 85% 72 50 71. 37 Mbarara 1,088,356 55 Ibanda 198,635 49,654 73% 76 Ibanda 198,635 43,341 54. 20 31% 65 40 67. 56 Isingiro 316,025 37,842 73% 77 Isingiro 316,025 32,580 60. 20 69% 46 60 ANNEX B 35 2001 2007 2011 Posts Filled Posts Filled Population- Population- Population- % Strategic % Strategic Table FY09 Table FY11 FY09 grant FY11 grant Sanitation Per capita Per capita Census Census Census League League District District District No. Health Health (UGX) (UGX) Rank 2002 2002 2002 No. No. 58. 57 Kiruhuura 212,219 42,056 36% 78 Kiruhuura 212,219 40,783 48. 90 31% 44 60 54. 79 Moroto 77,243 60,666 43. 00 54% 20 38 Moroto 189,940 58 Moroto 189,940 40,744 55% 00 80 Napak 112,697 29,708 41. 60 8% 25 51. 39 Moyo 194,778 59 Moyo 194,778 45,642 55% 81 Moyo 194,778 46,032 40. 70 62% 46 60 82 Mpigi 187,771 46,370 62. 10 92% 50 70. 40 Mpigi 407,790 60 Mpigi 407,790 45,239 91% 83 Butambala 86,755 57,795 55. 80 8% 58 60 84 Gomba 133,264 39,703 41. 20 8% 51 54. 61 Mubende 423,422 33,706 100% 85 Mubende 423,422 32,792 47. 10 100% 45 50 41 Mubende 689,530 77. 62 Mityana 266,108 42,438 45% 86 Mityana 266,108 40,314 58. 80 62% 74 30 87 Mukono 423,052 35,315 65. 30 62% 57 42 Mukono 795,393 63 Mukono 795,393 37,789 73% 88 Buikwe 329,858 27,506 46. 20 31% 41 89 Buvuma 42,483 52,868 50. 70 8% 35 45. 90 Nakapiripirit 90,922 61,866 49. 10 62% 16 43 Nakapiripirit 154,494 64 Nakapiripirit 154,494 50,830 64% 10 91 Amudat 63,572 33,678 23. 10 8% 11 65. 44 Nakasongola 127,064 65 Nakasongola 127,064 59,293 73% 92 Nakasongola 127,064 58,915 57. 00 77% 48 80 72. 93 Nebbi 266,312 46,168 62. 60 54% 53 45 Nebbi 435,360 66 Nebbi 435,360 41,106 55% 10 94 Zombo 169,048 39,847 54. 80 8% 40 59. 46 Ntungamo 379,987 67 Ntungamo 379,987 42,294 73% 95 Ntungamo 379,987 37,888 59. 40 54% 53 30 64. 96 Pader 142,320 68,817 53. 80 38% 56 47 Pader 326,338 68 Pader 326,338 46,173 36% 90 97 Agago 184,018 38,474 54. 20 8% 48 69. 48 Pallisa 520,578 69 Pallisa 384,089 42,076 73% 98 Pallisa 255,870 49,760 61. 50 77% 37 40 ANNEX B 36 2001 2007 2011 Posts Filled Posts Filled Population- Population- Population- % Strategic % Strategic Table FY09 Table FY11 FY09 grant FY11 grant Sanitation Per capita Per capita Census Census Census League League District District District No. Health Health (UGX) (UGX) Rank 2002 2002 2002 No. No. 59. 99 Budaka 136,489 50,253 61. 30 69% 57 70 Budaka 136,489 52,620 73% 80 100 Kibuku 128,219 34,106 51. 50 8% 42 71. 71 Rakai 404,326 48,508 91% 101 Rakai 404,326 47,454 64. 00 85% 64 30 49 Rakai 470,365 79. 72 Lyantonde 66,039 74,517 9% 102 Lyantonde 66,039 54,619 67. 70 23% 68 10 73. 50 Rukungiri 275,162 73 Rukungiri 275,162 50,777 73% 103 Rukungiri 275,162 41,408 65. 80 69% 63 80 57. 51 Sembabule 180,045 74 Sembabule 180,045 53,764 73% 104 Sembabule 180,045 55,842 51. 50 46% 45 80 59. 105 Sironko 185,819 47,724 65. 10 46% 34 52 Sironko 283,092 75 Sironko 283,092 45,328 45% 10 106 Bulambuli 97,273 55,216 17. 80 8% 28 56. 107 Soroti 193,310 40,707 53. 70 77% 37 53 Soroti 369,789 76 Soroti 369,789 38,987 82% 30 108 Serere 176,479 39,432 30. 10 8% 26 77. 77 Tororo 379,399 42,667 64% 109 Tororo 379,399 44,122 60. 90 69% 51 20 54 Tororo 536,888 73. 78 Butaleja 157,489 53,331 36% 110 Butaleja 157,489 52,270 68. 00 38% 46 00 65. 55 Wakiso 907,988 79 Wakiso 907,988 29,425 45% 111 Wakiso 907,988 25,688 53. 00 92% 66 70 49. 56 Yumbe 251,784 80 Yumbe 251,784 44,780 45% 112 Yumbe 251,784 48,808 47. 90 46% 56 30 65. 24,227,297 24,227,297 42,558 43% 24,227,297 40,541 58. 40 47% 30 9 Sources: Statoids. com for district boundaries, MoLG various Policy Statements for posts and grants, Annual Health Sector Performance Reports for FY09 and FY11, Water and Environment Sector Performance Report 2011, and Background to the budget on population. All grants are included in per capita grant calculations. League tables: minimum 1, maximum 100. 37 ANNEX B Conditional Grants. Conditional grants have grown in share (from 62 to 89 percent of all grants) and in number (from 16 in FY02 to 38 in FY11), and have become more restrictive over time. Each of the 38 conditional grants has its own allocation parameter, as shown by examples in Table 10 taken from LGFC (2012). Table 10. Allocation Parameters of Selected Conditional Grants Grant Present allocation parameters Primary Education (UPE)  Enrolment Urban Water O&M  Tariff subsidy  System specific allocation  Connection subsidy allocation Primary Health Care (PHC)  Population Non-Wage  Infant mortality Roads Maintenance (URF)  District roads: population and area  Community access roads: population National Agricultural  Rural Land Area Advisory Services (NAADS)  Rural Population  Poverty Head Count Rural Water and Sanitation  Basic minimum allocation for overheads/operations  Per capita cost for delivery of water and sanitation services  Population at sub-county  Safe water coverage at sub-county Rural Feeder Roads  Land area  Population Source: LGFC (2012) The initial LGDP allocation (for FY04) was determined as follows:  LGDP funds would be divided between rural and urban Local Governments on a per capita basis: US$2 per capita for urban authorities (including town councils) and US$1 per capita for rural authorities, in real terms.  The amounts allocated to rural areas would then be distributed among districts and sub-counties based on population (85 percent) and land area (15 percent), and to parishes based on population alone (100 percent). Within each district, funds would be further allocated to districts (35 percent) and sub-counties (65 percent). Sub- counties would then allocate 30 percent to parishes.  The amounts allocated to urban areas would then be distributed based on population alone (100 percent). Within each City or Municipal Council, funds would be further allocated to city or municipal councils (50 percent) and division councils (50 percent). Division councils would then allocate 30 percent to parishes.  As part of the performance evaluation system, incentives and penalties would be applied wherein those Local Governments that perform adequately would be rewarded with access to 20 percent additional funds and those who do not perform well would have their allocation reduced by 20 percent. Access to LGDP grants was open to all Ugandan Local Governments that would meet minimum conditions and were willing to co-fund 10 percent of the projects supported by LGDP grants through their own resources. ANNEX B 38 Decentralization Timeline • 1986: NRM gain power in Uganda • Local Resistance Councils (RCs) are very influential in the NRMs rise to power. • The value of local mobilization and organization is strongly appreciated. • RCs become de facto Local Governments. Building political support • Decentralization as a politically driven process is established for decentralization • Formal adoption of decentralization policy in 1992 1986-1992 • 1993-97: Staff posted from line ministries and introduction of district managed votes and grants • 1994: Development Partners (Danida) start supporting decentralization. Others follow. Focus is on community based grants and technical support to the Government. Community based development (away from centre) is a major theme. • 1997: LG Act passed . It assigns political, administrative, fiscal and service delivery powers to LGs in line with the Constitution • 2000: LGDP1 starts, Establishing the Local • 2002: The Fiscal Decentralization Strategy Paper is approved. It aims to enhance the Governance System discretionary local government funding and local revenue in support of their mandates as 1993-2004 autonomous institutions, whilst at the same time improving planning, reporting and accountability processes. • 2003: Local Government Finance Commission established by Act of Parliament to advise on Fiscal Decentralization issues and LGDP II starts • 2005: Adoption of a multi-party system with elections scheduled for 2006 • 2005: launch of the Local Government Sector Investment Plan (LoGSIP) . Donors establish a common basket fund to support LoGSIP. Other sectors (e.g Education) establish vertical planning tools, reducing LG discretionary expenditures. It will be supported by DPs till 2010. • 2005: District Chief Administrative Officers will be appointed by central government instead of district councils with the aim of removing them from “pressure exerted by local politicians”. IEG (2008) found that this fact increased independence and did not necessarily derail devolution. • 2005: Suspension of Graduation Tax, the predominant source of local revenue, in the run up to 2006 national and local elections. G Tax has never been adequately replaced. Recentralization and This political strategy was proposed by opposition and government parties as part of Proliferation of LGs their manifestos. • 2006-2011: the number of districts almost doubles (from 56 in 2000 to 111 in 2011) - 2005-2012 see Figure 2. • 2007: introduction of Universal Secondary Education (USE) in Uganda as a centrally managed service • 2009: drug procurement is centralized to "fight corruption" • 2010: centralization of Feeder Roads resource allocation by the Uganda Road Fund. • 2010: Parliament approves The Kampala City Bill (which places Kampala under the administrative authority of the Central Government who appoints its executive director) Source: PPAR Team 39 ANNEX B Source: PPAR Team ANNEX B 40 Source: PPAR Team 41 ANNEX C Annex C. Evolution of Bank Portfolio The First Urban Project (approved 1991, closed 2000) supported the government's efforts to decentralize the responsibility for the planning, design and management of urban services to local authorities while assisting with reconstruction of the country after a period of political instability, social strife and physical destruction. The project objectives were to: (a) improve living conditions and alleviate poverty in Kampala by restoring key infrastructure services; (b) support the development of decentralized local urban management by strengthening the revenue base, financial management and technical capacities of the Kampala City Council (KCC) and by improving the ability of central government to assist local authorities to increase their revenue base and strengthen financial management; (c) strengthen the country's capacity to manage the process of urban land development; and (d) promote sound cost recovery policies and practices. The project included support for: (a) the rehabilitation of urban markets, streets and drains, refuse collection and disposal; (b) the servicing of land for residential development; (c) the preparation of up-to-date mapping and a strategic urban development plan for Kampala; and (d) technical assistance and training for KCC, the Ministry of Local Housing and the Physical Planning Department staff. The Small Towns Water and Sanitation Project (approved 1994, closed 2003) supported the government's economic recovery program by extending the rehabilitation and upgrading of water supply and sanitation services to towns that had not been covered before. It aimed at: (a) improving health conditions through better water supply, excreta disposal, waste water management, and public hygiene; (b) alleviating poverty and improving the lot of women; and (c) reducing environmental degradation through better waste management. The project provided 12 small towns and Jinja with improved and sustainable water supply and sanitation through: (a) the rehabilitation and/or expansion of water supply and sanitation facilities; (b) hygiene education related to water supply and sanitation; (c) community participation in planning, implementation, operation, and maintenance of water supply and sanitation facilities; and (d) institutional strengthening, technical assistance and training for the organizations in the sector. The Institutional Capacity Building Project (approved 1995, closed 2002) aimed to establish local institutional and human capacity to develop and implement public policy and support the growth of the private sector through: (a) support for the continuation of the Civil Service Reform; (b) assistance to the Government of Uganda in its decentralization program; and (c) strengthening the legal and financial accountability framework and institutions. The project consisted of five components: (i) central government capacity building; (ii) Local Government capacity building; (iii) legal sector reform; (iv) accountancy profession; and (v) training funds. The District Health Project (approved 1995, closed 2002) aimed at pilot-testing and demonstrating the feasibility of delivering an essential health services package to district populations, within a prudent financial policy framework for the sector in order to improve the efficiency and equity in the provision of health services. The project was designed to support the Government's strategy of decentralizing health services. It planned to increase the ANNEX C 42 efficiency of the existing health infrastructure and institutions through consolidation and improved management with increased local accountability, while simultaneously supporting cost recovery and budgetary policies that will enable the health care system to move toward long term sustainability. It supported the Government's efforts to reorder priorities within the existing health care system through efficiency improvements and by reallocating financial and human resources toward ensuring the provision of a package of essential health services for all Uganda's citizens. Accordingly, the project included activities to: 1) pilot and test new sector policies and strategies which will facilitate the implementation of essential health services; 2) strengthen management and planning capacity at district levels so that they are prepared to provide essential health services; and 3) restructure the Ministry of Health so as to build its capacity to provide health policy leadership and to support the Government's decentralization policy. The Nakivubo Channel Rehabilitation Project (approved 1999, closed 2004) was aimed at alleviating the frequent flood incidence on the road network, which affected traffic flows, with an adverse impact on the economic activity and living conditions in Kampala. Institutionally, the Project tried to enhance the Kampala City Council (KCC) management capacity, helping establish the infrastructure investment policies, and ultimately, support KCC's reform program. The project components included civil works regarding the rehabilitation of the main channel, construction of auxiliary drains, and, the rehabilitation of priority drainage. Consulting services for construction supervision were to be financed by the project, as well as the required program and policy studies, namely the Kampala Drainage Master Plan Study and the Kampala Urban Transportation Improvement Program Study. Institutional strengthening was planned for both the KCC reform program under the strategic framework for reform, and the revenue enhancement program. To facilitate outsourcing of basic delivery functions, seed funds were made available and training and implementation support were further provided. The Second Economic and Financial Management Project (approved 1999, closed 2006) was part of the assistance strategy for supporting improvements in the public expenditure management process in Uganda, and the project was aimed at improving the efficiency of Government planning/budgeting, financial management, and monitoring/evaluation processes. The components were expected to: 1) support strengthening of central, and Local Government planning, and budgeting processes, through improved transparency, and accountability of budget, and public expenditure practices. Fiscal decentralization was to be expanded, and effective linkages between national, and Local Government planning were to be strengthened, together with recurrent, and development budget integration; 2) improve financial and accounting practices, through human resources capacity building, to improve financial management systems, and information, and enhance external auditing functions; 3) support capacity building to monitor the performance, and delivery service of public expenditure on living standards among various groups. This included improved capacity of the Uganda Bureau of Statistics, and implementation of the National Service Delivery Survey; 4) test the effectiveness of distance learning, through the construction of a learning center, equipment supply, and operational technical assistance. The Local Government Development Program (approved 1999, closed 2004) initiated a long- term effort, to assist Uganda in the decentralization of basic public services, and in the ANNEX C 43 implementation of Local Governments, and tested alternative services delivery mechanisms, through the private sector, beneficiary communities, or stakeholders. Given the cross-sectoral nature of this project, the strategic framework was widely supported, namely, through the promotion of labor-intensive macroeconomic growth, and increased private sector investments. The components did: 1) provide technical assistance to the Ministry of Local Government (MoLG), and the Local Government Finance Commission Secretariat, for capacity building and policy development, as both institutions strive to become operational. Under the decentralization policy, MoLG was expected to mentor Local Governments, in line with the Constitution, and the Local Governments Act of 1997. Computerized databases provided the required linkages for adequate information communications; 2) finance basic services delivery investments, and capacity building, through grants from the Central Government to the Local Governments; 3) support and test, alternative basic services delivery for the Kampala City Council, implemented through reorganization, private financing, and improved management practices; and, 4) support program management, monitoring/ evaluation, and, future program formulation. The Kampala Institutional and Infrastructure Development Project (approved 2007, it is scheduled to close by end December 2012) aims to improve institutional efficiency of the Kampala City Council (KCC) through implementation of the Strategic Framework for Reform (SFR). The project includes the following components: Component 1 -- support to KCC and its stakeholders to refine and expand the SFR into a comprehensive approach to municipal development, consonant with Kampala's central role in the nation's economic and political life. Component 2 -- provide city wide infrastructure and services improvements. Component 3 -- support to KCC on project implementation and the establishment and implementation of a comprehensive monitoring and evaluation system. The Local Government Management and Services Delivery Project (approved 2007, it is scheduled to close by end December 2012) aims to strengthen the ability of the Ministries, Departments and Agencies (MDAs) and Local Governments (LGs) to plan and manage resources in collaboration with communities for service delivery. The project had three components: Component 1 - Support to the Public Financial Management (PFM) system reform program - intended to strengthen PFM at central and Local Government levels and to ensure efficient, effective, transparent and accountable use of public resources; Component 2 - Support to the Local Government Sector Investment plans (LoGSIP) - support Local Governments infrastructure development through the (a) GoU local development grant (LDG), (b) professionalization of Local Government staff, and (c) support to Local Governments in Northern Uganda which were weak and emerging from situation of civil war. The Transforming the Settlements of the Urban Poor in Uganda - A Secondary Cities Support Programme (TSUPU) (approved 2010, it is scheduled to close by end December 2012) aims to strengthen the ability of Local Government (LG's) to plan and manage resources in collaboration with communities for service delivery. Negative measures include: air pollution, noise pollution, waste management, water pollution, soil erosion, ecology, drainage, and natural resources. Mitigation measures include: a) traffic emissions should be monitored and legally permitted levels should not be exceeded; b) during construction, un- paved roads should be water sprayed and doused to reduce dust levels; c) employers should ANNEX C 44 provide protective equipment e. g. dust masks and construct well-ventilated workshops as necessary; d) movement of vehicles and operation of construction machinery should be confined to daytime; and e) construction should be monitored by local and district public health officials notably in the siting of these items. 45 ANNEX D Annex D. List of Persons Interviewed Mr. Roland White, Senior Institutional Development Specialist Mr. Gaiv Tata, former TTL ICBP Mr. Moustapha Ndiaye, Country Manager for Uganda Eng. Paul Kasule-Mukasa, Project Coordinator, LGMSD, MoLG Mr. David Kiggundu, Programme Coordinator, Kampala Institutional & Infrastructure Development Project Mr. Andrew Musoke, Asst. Commissioner, Policy & Planning Department, (MoLG) Mr. Martin Onyach-Olaa, ICR Author and former MoLG staff Mr. Swizin Kinga Mugyema, Commissioner, Local Council Devt, MoLG Mr. Patrick Mutabwire, Ag. Permanent Secretary, Ministry of Local Govt Mr. Abbey Iga, Commissioner, Urban Inspection, MoLG Mr. Richard Sewakiryanga, Executive Secretary, Uganda National NGO-Forum Mr. Pius Bigirimana, Permanent Secretary, Office of the Prime Minister Mr. Paul Okot-Okello, Commissioner, District Admin, MoLG Mr. Lawrence Banyoya, Commission Secretary, LGFC Eng. Gumusiriza Birantana, Chairperson, Uganda National Association of Building and Civil Engineering Contractors Mr. Samuel Amule, Commissioner, District Inspection, MoLG Mr. David Kigenyi Naluwayiro, CAO, Wakiso District Ms. Sylvia Keera, M&E, MoLG Mr. Byamugisha Albert, Commissioner, Monitoring & Evaluation, Prime Minister Office Ms Hope Alice Nakyanzi, CAO Jinja District Mr. Keith Muhakanizi, DST, Ministry of Finance, Planning & Economic Development Mr. Swizin Kinga Mugyema, Commissioner, Local Council Devt, MoLG Mr. John Muwanga, Auditor General 46 ANNEX E Annex E. Borrower Comments ANNEX E 47 ANNEX E 48