Report No. 40153 Local Government Discretion and Accountability: A Local Governance Framework Social Development Department in collaboration with the Finance, Economics and Urban Department (FEU) and the Social Protection Team (HDNSP) THE WORLD BANK June 2007 ©2007 The International Bank for Reconstruction and Development / The World Bank 1818 H Street NW Washington DC 20433 Telephone: 202-473-1000 Internet: www.worldbank.org E-mail: feedback@worldbank.org All rights reserved This volume is a product of the staff of the International Bank for Reconstruction and Development/ The World Bank. The findings, interpretations, and conclusions expressed in this paper do not necessarily reflect the views of the Executive Directors of The World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Rights and Permissions The material in this publication is copyrighted. Copying and/or transmitting portions or all of this work without permission may be a violation of applicable law. The International Bank for Reconstruction and Development/ The World Bank encourages dissemination of its work and will normally grant permission to reproduce portions of the work promptly. For permission to photocopy or reprint any part of this work, please send a request with complete information to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA, telephone 978-750-8400, fax 978-750-4470, http://www.copyright.com/. All other queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Publisher, The World Bank, 1818 H Street NW, Washington, DC 20433, USA, fax 202-522-2422, e-mail pubrights@worldbank.org. Table of Contents Acknowledgements.................................................................................................................................................. iv Abbreviations and Acronyms.................................................................................................................................. v Executive Summary.................................................................................................................................................. vi 1. Objective and Target Audience....................................................................................................................... vi 2. A Framework for Local Government Discretion and Accountability......................................................vii 3. Strengthening the Bank's Work in Local Governance................................................................................. ix Chapter 1: Introduction..............................................................................................................................................1 1.1. Background.......................................................................................................................................................1 1.2. Objective and Target Audience......................................................................................................................2 1.3. Analytical Content...........................................................................................................................................3 1.4 A Framework for Local Government Accountability................................................................................11 1.5 Local Governance Categorizations...............................................................................................................13 1.6 Summary and Outline of the Study .............................................................................................................13 Chapter 2: Decentralization and Local Discretion ...............................................................................................15 2.1 Introduction.....................................................................................................................................................15 2.2 Political Dimension.........................................................................................................................................15 2.3 Administrative Dimension............................................................................................................................19 2.4 Fiscal Dimension.............................................................................................................................................24 2.5 Conclusion.......................................................................................................................................................33 Chapter 3: Decentralization and Local Accountability........................................................................................35 3.1. The Multiple Dimensions of Accountability..............................................................................................35 3.2. Political Accountability in Decentralization ..............................................................................................38 3.3. Administrative Accountability in Decentralization..................................................................................44 3.4. Financial Accountability in Decentralization.............................................................................................50 3.5 Conclusion.......................................................................................................................................................57 Chapter 4: Reconciling Accountability and Discretion in Decentralization....................................................58 4.1 Policy Recommendations ..............................................................................................................................59 4.2 Strengthening The Bank's Work in Local Governance..............................................................................65 4.3 Further Critical Issues to Address for Achieving Good Local Governance ..........................................71 4.4 Conclusion.......................................................................................................................................................72 References..................................................................................................................................................................74 i Annex 1: Study on Uganda .....................................................................................................................................81 Local Government Discretion and Accountability...............................................................................................81 Abbreviations And Acronyms............................................................................................................................81 1.1 Background......................................................................................................................................................82 1.2 Political setting and Accountability in Decentralization..........................................................................86 1.3 Administrative Discretion and Accountability in Decentralization.......................................................92 1.4 fiscal discretion and accountability in decentralization ............................................................................94 References............................................................................................................................................................101 Annex 2: Study on Kerala, India Local Government Discretion and Accountability ...................................103 Abbreviations And Acronyms..........................................................................................................................103 1.1 Background....................................................................................................................................................103 1.2 Political setting and Accountability in Decentralization........................................................................105 1.3 Administrative Discretion and Accountability in Decentralization.....................................................108 1.4 fiscal discretion and accountability in decentralization ..........................................................................112 References............................................................................................................................................................118 Annex 3: Study on Philippines Local Government Discretion and Accountability....................................120 Abbreviations And Acronyms..........................................................................................................................120 1.1 Background....................................................................................................................................................121 1.2 Political setting and Accountability in Decentralization.........................................................................122 1.3 Administrative Discretion and Accountability in Decentralization.....................................................128 1.4 fiscal discretion and accountability in decentralization ..........................................................................132 References............................................................................................................................................................139 Boxes and Tables Box 2.1: Relationships between the political dimension of decentralization and local political settings....16 Box 2.2: Uganda's "no-party" system....................................................................................................................18 Box 2.3: Strengthening political parties in Africa.................................................................................................19 Box 2.4: The relationship between administrative decentralization and local discretion...............................20 Box 2.5: Mexico's health sector human resource management and decentralization reforms ......................22 Box 2.6: Administrative discretion over local staff and the tendering process in Uganda.............................23 Box 2.7: Relationship between fiscal decentralization and local discretion......................................................24 Box 2.8 Criteria for assigning responsibilities between central and local governments.................................27 Box 2.9 A simple model for analyzing intergovernmental transfers.................................................................29 Box 2.10: Municipal borrowing in Brazil...............................................................................................................33 Box 3.1. Impact of reserved seats in India and Uganda.......................................................................................39 Box 3.2. Citizen Oversight Bodies in Bolivia and Japan......................................................................................42 Box 3.3. Building a culture of citizen oversight: a vignette from Indonesia's Kecamatan Development Project.........................................................................................................................................................................44 Box 3.4. Overseeing the local government tendering process............................................................................49 ii Box 3.5. The citizens' charter in the municipality of Mumbai, India.................................................................50 Box 3.6. The Public Expenditure and Financial Accountability Initiative defines core dimensions of an effective PFM system:...............................................................................................................................................52 Table 3.1: Local governments and internal controls in selected countries........................................................54 Table 4.1: Policy Recommendations to Strengthen Discretion and Accountability.........................................62 Table 4.1: Policy Recommendations to Strengthen Discretion and Accountability (continued).....................63 Table 4.1: Policy Recommendations to Strengthen Discretion and Accountability (continued).....................64 Box 4.1. Examples of operational entry points: local governance support project in Bangladesh and demand for good governance project in Cambodia ............................................................................................70 Annex 3: Box 1 CSO Representation.....................................................................................................................................131 Figures Figure 1.1: Framework for local government discretion and accountability......................................................5 Figure 1.2: Actors of Accountability..........................................................................................7 Figure 1.3: Good local governance and accountability........................................................................................12 Figure 3.1: The public financial management cycle.............................................................................................52 Figure 4.1: Local governance turnarounds--trajectories ....................................................................................60 Annex 1: Figure 1 Local Government Units in Uganda.......................................................................................................85 Figure 2 Administrative Structure at the District Level ......................................................................................87 Figure 3 Local Council Role and Oversight over Administration .....................................................................90 Annex 2: Figure 1 Indian Government Structure................................................................................................................104 Figure 2 Organization Structure of the Gram Panchayat, Kerala ....................................................................106 Figure 3 Source of Finances for Gram Panchayats in Kerala............................................................................115 Annex 3: Figure 1 Municipal/City Government Structure.............................................................................................123 Figure 2: Provincial Government Structure................................................................................124 iii Acknowledgements This report was prepared by a team led by Serdar Yilmaz (SDV). The report is a collaborative effort with Finance, Economics, and Urban Department (Balakrishna Menon Parameswaran) and HD Network (Rodrigo Serrano-Berthet). Core team members were: Serdar Yilmaz (SDV), Yakup Beris (SDV), Balakrishna Menon Parameswaran (FEU), Rodrigo Serrano-Berthet (HDNSP), Varsha Venugopal (SDV), Andrea Ries-Padmanabhan (SDV), Andre Herzog (SDV), Myrtle Diachok (SDV). Guidance throughout report preparation was provided by Steen Lau Jorgensen (SDV), Caroline Kende-Robb (SDV), Laszlo Lovei (FEU), Robert Holzmann (HDNSP), and Robert S. Chase (SDV). The team appreciates the useful comments that were provided by: Carine Clert (SDNVP), Maria Emilia Freire (FEU), Kai Kaiser (PRMPS), Marisela Montoliu Munoz (FEU), Cecile Fruman (MNC01), Caroline Mascarell (ECSHD), Nilufar Ahmad (SAS), Jonas Manuel Frank (LCSPS), Keith W. McLean (ECSSD). Very helpful logistical assistance was provided by: Greicy C. Amjadi (SDV), Elizabeth Acul (SDV), and Herminia Adviento (SDV). We would like to thank Dina S. Towbin (EXTOP- consultant) for editing and Danielle Christophe (SDV) for formatting the report. The peer reviewers for this report are: Junaid Ahmad, Sector Manager (SASES) Luis Constantino (MNSRE), Jacomina P. de Regt (AFTSD), Ayse Guner, (Head, Center for Local Government Research and Training, Marmara University), and George Matovu, (Director, Municipal Development Programme (MDP)). iv Abbreviations and Acronyms ACAO Assistant Chief Administration Officer CAO Chief Administration Officer CSO Civil Society Organizations DEC District Executive Committee DSC District Service Commission LC Local Council LLG Lower Levels of Local Government LG Local Government LGDP Local Government Development Project LGIFA Local Government Integrated Fiduciary Assessment LGMSD Local Government Management and Service Delivery LGTB Local Government Tender Board MFPED Ministry of Finance, Planning and Economic Development NGO Non Government Organization NRM National Resistance Movement OAG Office of the Auditor General PAC Public Accounts Committee PAF Poverty Action Fund PEAP Poverty Eradication Action Plan PRSC Poverty Reduction Support Credit RDC Resident District Commissioners UDN Uganda Debt Network ULGA Uganda Local Government Association Vice President, SDN: Kathy Sierra Director, SDV: Steen Lau Jorgensen Director, FEU: Laszlo Lovei Director, Social Protection: Robert Holzmann Sector Manager, SDV: Caroline Kende-Robb Task Team Leader: Serdar Yilmaz Co-Task Manager: Balakrishna Menon Parameswaran Co-Task Manager: Rodrigo Serrano-Berthet v Executive Summary 1. OBJECTIVE AND TARGET AUDIENCE Social accountability approaches are broad range of actions and mechanisms beyond voting that citizens can use to hold accountable the state and providers of public services. These approaches could be as simple as pressuring local governments to publish budgets, or more complicated, such as participatory budgeting or social audits where citizen groups inspect, for example, public procurement policies and processes. Often times, when opportunities for governance reform at the national level are limited, there may be local level entry points (World Bank, 2007a).1 This Economic and Sector Work (ESW) examines how social accountability approaches--together with the public sector (supply-side) approaches--can help create such entry points, when institutionalized into local government operations. The main objective of this report is to present a coherent and comprehensive approach for decentralizing countries to achieve a high degree of local government discretion accompanied by robust accountability mechanisms. To this end, this ESW presents an analytical framework of local governance through the lens of accountability. This ESW intends to establish a methodological basis for analyzing linkages between decentralization and accountability. The report is based on desk reviews and interviews--it does not provide empirical research testing the report's arguments. Rather, it provides many country specific examples, and three detailed case studies (on Uganda, the Philippines, and Kerala, India) to demonstrate how the presented analytical framework could provide a useful method for studying accountability implications of decentralization reforms. It should be noted here that presenting such a framework runs the risk of generalizing issues that are mostly local context specific. The goal here is definitely not to build a one-size fits all prescription as such. This is why the local governance strategy proposed in the final chapter underlines the importance of developing systematic analytical tools and operational approaches that would rely on country specific information on local power structures, interests, and socio- economic conditions. Accordingly, the use of the analytical framework proposed here is very much dependent on local realities and it is the local policy makers in client countries that would ultimately decide what is suitable for their country and what is not. Recently, client countries, donors, and the Bank have expressed increasing interest in and demand for participatory mechanisms, which could enhance local accountability, consequently strengthen governance, and reduce corruption. This report--a direct response to this growing interest--presents various social accountability mechanisms that could complement public sector accountability approaches. It also highlights the operational challenges the Bank's projects on local governance reform must often confront. Accordingly, the findings of this ESW provide, in one comprehensive framework, guidance on both macro policy-making and implementation of decentralization reforms. As such, this report provides the Bank with a basis to develop an integrated local governance strategy to guide lending and non-lending operations. The primary audience for this report comprises public policy makers and local governance stakeholders in client countries, governmental and non-governmental practitioners and policy advisers working on local governance, local development, and public sector reforms, and the Bank staff advising countries on local government reforms. The secondary audience includes academics in the fields of economics, public administration, public policy, and social development. 1The Governance and Corruption Strategy Paper of the Bank's Development Committee acknowledges that even when opportunities for governance reform at the national level are limited, there may be entry points at the local level. vi 2. A FRAMEWORK FOR LOCAL GOVERNMENT DISCRETION AND ACCOUNTABILITY Central to the framework presented in this ESW is the following argument: Decentralization reforms should extend discretion and provide accountability on three dimensions: political, administrative, and fiscal. That is, in each of these three dimensions, local governments must undertake their newly extended powers and responsibilities under the scrutiny of accountability mechanisms, cultivated through strong supply-side institutions and energized by demand side pressures. The report builds on the World Development Report (World Bank 2004b) framework to identify local accountability actors and establishes the linkage between accountability and good local governance. In this framework, · Decentralization reforms grant local governments new powers and responsibilities in three dimensions: political, administrative, and fiscal. These dimensions provide local governments a discretionary space (chapter 2). · Accountability relationships may be driven in different directions: upward and downward. Within their discretionary space, local governments are accountable to upper levels of government (upward accountability) as well as to citizens (downward accountability). · Public accountability mechanisms, introduced for oversight and accountability of local governments, safeguard against abuse of local discretion. · New forms of social accountability mechanisms, which enable direct engagement of citizens with government, emerge to complement public accountability mechanisms (chapter 3). · Public and social accountability approaches must be bridged to ensure that citizens have the ability and opportunity to demand accountability and local governments have the means and capacity to respond to these demands (chapter 4). 2.1 Expanding Local Government Discretionary Space and Improving Accountability: Policy Recommendations In our framework, the laws and regulations governing political, administrative, and fiscal institutions and processes, which are decided primarily by the central governments, define the local government's discretionary space in decentralization. Local governments are then subject to accountability requirements for their use of discretion at each level. The following sections explain the details on how local discretion is given, and accountability is strengthened. Political Setting and Accountability Political setting is defined by arrangements that set the rules for local populations to elect and interact with their local leaders. These arrangements also define the degree of oversight functions that those local leaders enjoy over local bureaucracy. These arrangements are established through the electoral system, local party system, and the oversight powers assigned to locally elected leaders in planning, budgeting, and service delivery. An appropriate local political setting conducive to downward accountability requires free and fair elections and local political competition for mayors and councilors. This kind of political setting also supports competitive multi-party systems and provides a clear separation of power between the executive (i.e., mayor) and legislative (i.e., council) bodies of local governments. It also establishes mechanisms and provides tools to the legislative body in overseeing the operations of the executive body. Political accountability is a process whereby citizens hold elected their officials to account for their behavior and performance (Aucoin and Heintzman 2000). This could be, for example through elections vii directly. Political accountability can also be improved through elected local officials overseeing local executives, through activities that increase awareness about the policy performance of local governments, or through direct citizen involvement in policy decision making beyond elections. On the supply side, the most common measures to improve downward political accountability include ensuring representation of marginalized/vulnerable groups through reserved seats or quota systems; improving the quality of the electoral system with recall, write-in, and independent candidate options; establishing, monitoring, and enforcing transparent rules for financing local elections; reviewing and revising the role of campaign financing rules (that may favor centralized, established parties, and make it more difficult for local candidates or new parties to compete); and securing and strengthening the role of elected body (council) in overseeing local government operations. On the demand side, political accountability measures allow for citizen-initiated legislation (i.e., petitions), referendum, or recall of elected public officials; establish procedures for public petitions to adopt, amend, or repeal an act, legislation, or executive order; empower citizens to demand public hearings on policy decisions and action; and institute citizen ombudsman offices in local governments. Administrative Discretion and Accountability Administrative arrangements establish the rules, autonomy, and incentive structure for well-functioning and accountable local bureaucracies. In order to claim administrative autonomy, local governments should have discretion in at least three areas: First, local governments must control the regulatory structure. That is, they should be able to make, change, and enforce laws and regulations pertaining to local administrative affairs (i.e., on spatial and sector planning, environmental measures, and local economic development). Second, local governments should enjoy a degree of autonomy over the local procurement process. Third, local governments should have control over local civil service and employment policies. Local governments gain greater administrative autonomy when they are able to pay their staff from own-budgets, have a decisive control over staffing levels and skill mix, and can offer their staff with enough incentives, flexibility and opportunity in career advancement and performance outcomes. Administrative accountability refers to local civil servants being accountable to their top administrative officers and to outside bodies such as public audit officers, ombudsmen, and regulators or a particular administrative agency, or a board/committee about the use of their administrative discretion. On the supply side, measures to improve downward administrative accountability include independent judicial/quasi-judicial agencies that investigate cases of local government corruption and misconduct; regular administrative audits (external audits) of local government transactions through permanent or ad hoc independent entities; administrative courts that deal with compliance on local regulatory decision making and actions; a national procurement strategy for local governments; well-defined procurement rules and processes; independent procurement audits; standards for service delivery and reward performance; strong information systems to monitor service quality; clear rules regarding civil service practices (payrolls, staff levels, recruitment practices, staff performance);and systems for periodic monitoring and timely revisions (often through local government commissions) of civil service rules; and a flexible career and performance management supported by a civil service system with merits-based recruitment. On the demand side, administrative accountability is achieved with citizen participation in decision-making processes regarding administrative affairs directly or indirectly, through neighborhood and community councils. In this setting, citizens could hold their local government accountable by challenging administrative decisions; public is regularly involved in contracting and implementation of public works projects through social audit committees; and citizen charters empower the citizenry in their interactions with public service providers. Finally, public oversight mechanisms provide citizens a channel to oversee local service delivery, express their satisfaction with local services (i.e. through scorecards and service delivery surveys). All these mechanisms provide incentives for good performance to the local public service providers viii Fiscal Discretion and Accountability Fiscal decentralization rearranges roles and responsibilities among different levels of governments with the intent of transferring some of fiscal decision-making powers from central to sub-national governments. Fiscal decentralization rules regulate four areas: expenditure assignment; revenue assignment; intergovernmental transfers, and sub-national borrowing (Bird 2000). Expenditure autonomy goes beyond merely assigning service delivery responsibilities to local governments--the extent of this autonomy defines, in the eyes of citizens, the relevance of local governments in service delivery. Revenue autonomy and transfer systems define the funding level for local governments to deliver these services. Borrowing powers provide local governments scope for additional decision making in investments. In the absence of an appropriate fiscal discretionary space for local governments, intergovernmental fiscal arrangements remain incomplete. This incompleteness characterizes intergovernmental fiscal frameworks in many countries, unduly burdening local governments with new mandates and responsibilities without the means to obtain resources necessary to undertake them. Fiscal accountability seeks transparency in the management of public funds. It requires that governments manage finances prudently and ensure integrity in their financial and non-financial reporting, control, budgeting, and performance systems (Sahgal and Chakrapani 2000). It also necessitates reporting comprehensively on what local governments have achieved with their expenditures. Supply side measures to improve downward financial accountability include strong local capacity for budgeting and public financial management; standards for control on intergovernmental transfer revenues (i.e., clean audit reports, submission of financial statements); publication of transfer figures; transparent local public audit systems--with publicly available audit findings-; clear rules for responsible local borrowing (including rules regarding defaults), public access to borrowing information, and clearly defined rules regarding hard budget constraints for local governments. Demand side measures for fiscal accountability include publicly accessible local government financial information (including budgets and end-of-year financial statements); strong public involvement in budgetary process through participatory budgeting practices; gender-sensitive planning, budgeting, and resource allocation, reinforced by gender audits; independent budget analysis and participatory public expenditure tracking programs that monitor budget execution and leakage of funds. 3. STRENGTHENING THE BANK'S WORK IN LOCAL GOVERNANCE Currently, the World Bank is engaged in decentralization programs in over 80 countries.2 This broad support for decentralization is not incidental--local governance is one of the entry points for strengthening governance and anticorruption efforts in client countries, even when governance reform opportunities at the national level are limited (World Bank, 2007a). Therefore, the process of reformulating accountability relations within decentralization reforms has important operational implications for the success of the Bank's decentralization programs. Yet, the Bank's decentralization and social accountability agendas are often weakly connected. This report provides a method for integrating these agendas under the accountability framework, as part of a broader effort to improve local governance in Bank's client countries. It describes the main components of such a local governance strategy, identifies the research gaps that must be filled to fully develop this strategy, and discusses the operational implications of merging decentralization and accountability agendas. 2 According to the Independent Evaluation Group's review of operational database, between 1987 and December 2006, the Bank committed about US$32 billion to about 89 countries through 458 programs, projects, and grants in which decentralization was noted as one of the key themes or classified as an activity. ix In broad terms, a local governance strategy should provide support to all stages of public policy making in client countries. United Nation Capital Development Fund's Practitioner's Guide (UNCDF 2005) provides guidance for building such a strategy. Following similar lines, the report recommends that the Bank's assistance focus on four areas: (i) National policy debate: The Bank needs to be involved in national level policy discussions that establish macro objectives (i.e., in drafting cabinet papers and white papers to set the macro policy guidelines). This involvement should include both the executive and legislative branches of governments in the country- as they may have differing objectives and interests in shaping policy decisions. (ii) Support for legislation and the statutory framework: Once the general direction of decentralization policy is established, it should be implemented through legislation (i.e., a decentralization bill). The Bank can provide support to clients in drafting legislations, consistent with other standing legislations. (iii) Support for the regulatory framework: The Bank should engage with client countries during the design of the regulatory framework to ensure consistency and timely issuance of secondary legislations (bylaws, ministerial decrees, circulars, and so on). (iv) Support for establishing systems, procedures, guidelines, and practices: The Bank should provide support to client countries in developing financial and administrative systems, procedures, and guidelines (i.e., budget planning guidelines, service delivery norms, and so on). The key features of Bank's local governance strategy should include: Policy Pilots: Closely monitored pilots should, whenever possible, precede the full-scale introduction of new policies and delivery mechanisms. Policy Innovations: A widely acknowledged by-product of pilots and policy trials is their role in encouraging and facilitating innovation. Scaling-up and Policy Impact Strategy: It is also important for the Bank to develop a policy impact and replication strategy. Local Institutional Innovation: The Bank's local governance strategy should be flexible to accommodate local institutional innovations (i.e., engaging local traditional authorities) Establishing Historical Benchmarks for the Evolution of Local Governance Structures: The formation of the nation state in developed countries could be used as a historical benchmark to compare the local government evolution processes in developing countries. Building the Knowledge Base: It is very important to systematically collect information about different dimensions of a local governance structure in a country. Building a knowledge base facilitates better decision making. Capacity Building: Capacity building should take place at the individual or group levels (covering skills and knowledge requirements), at the institutional or organizational levels (covering operational and administrative aspects), and at the strategic or systemic level (covering legal, political, and economic frameworks). In order to make such a local governance strategy concrete, a number of country-specific analytical and diagnostic studies pertinent to decentralization and accountability warrant attention. These studies should focus on four key areas: a) Actors, powers, and accountability relationships (research on the impact of the local political environment on the subsequent outcome of decentralization reforms); b) Economic and social inequalities (to better understand the relationship between socioeconomic inequalities and elite capture as they affect the design of decentralization reforms); x c) Local electoral structure and the party system (to grasp the structure of local government electoral systems and processes and their impact on the degree of accountability of local governments to citizens); d) The drivers of locally elected leader accountability and performance (to examine factors that drive the performance of locally elected leaders, including accountability) These analytical studies should also feed into the Bank's lending operations. The Bank's lending instruments (investment loans and development policy loans) provide financing for a wide range of activities aimed at creating physical and social infrastructure. Some of these instruments aim to make operational local governance support, but a portfolio review of Bank project documents reveals that many lending projects fall short of establishing good local governance institutions. Thus, the Bank's local governance strategy should standardize the use of both supply and demand side approaches for improving downward accountability in lending operations. xi Chapter 1: Introduction 1.1. BACKGROUND In the past two decades or so, central governments around the world have been decentralizing fiscal, political, and administrative responsibilities to lower-level governments. This drive to decentralize government authority and functions has gained importance as a key component of public sector reform efforts in both the developed and developing world. As noted by Bird (2000), decentralization is particularly widespread in developing countries for a variety of reasons: in some countries, it is an outcome of the advent of multiparty political systems and the deepening of democratization; in others, it marks the transition from a command to a market economy; in yet other countries, it counters the challenge of ethnic and geographic diversity; and finally, in most cases, it is driven by the need to improve the delivery of local services to diversified populations and jurisdictions. Even as the motivation behind decentralization in each country differs (Shah and Thompson 2004), its core concern is always to bring decision making closer to people and thereby to improve state­society relations. This is possible through establishing a representative government "that works best when it is closer to people" (Stigler 1957). Such orientation towards people is posited in two arguments supporting decentralization: responsiveness and accountability.3 A government is representative when it is responsive and accountable to citizens. Yet there is some empirical evidence to suggest that decentralization does not always achieve the goal of making local governments more responsive and accountable. According to Shah and Shah, decentralization "spread like wildfire in developing and transition economies in the 1990s yet fruits of this democratic participation remain largely elusive" (Shah and Shah 2007: 72). In fact, local governments are often susceptible to rent seeking by public officials and elite capture by well-off and well-connected groups. Decentralization can become counterproductive to accountability where timely, open, and fair elections are not in place, information is not readily accessible, clarity about local government's role is lacking, performance standards and monitoring mechanisms are lax, and local governments do not have sufficient resources and capacity to deliver mandated tasks. Decentralization as such can aggravate existing social problems, further polarize local politics, marginalize vulnerable groups, and reduce efficiency and equity in public expenditures and service delivery even more. Decentralization reforms reshape accountability relationships among actors and agents within the public sector. They also affect the way local governments interact with local populations. Accordingly, institutions and processes that are shaped by decentralization reforms have a substantial impact on establishing/reformulating these accountability links, and therefore these links cannot be avoided or side- stepped when developing an effective decentralization strategy. The process of reformulating accountability relations in decentralization reforms has also important operational implications for the World Bank as it broadens its support for and deepens engagement in decentralization policies and programs in over 80 countries.4 And although decentralization and social 3We are using responsiveness and accountability as necessary components of establishing representative local governments in the context of decentralization. Representative local governments are both responsive and accountable to citizens in their localities. Our particular focus is on accountability and how it is achieved in the context of decentralization reforms, but we also attempt to link accountability to government responsiveness in local governance frameworks. 4According to the Independent Evaluation Group's review of operational database, between 1987 and December 2006, the Bank committed about US$32 billion to about 89 countries through 458 programs, projects, and grants in which decentralization was noted as one of the key themes or classified as an activity. 1 accountability are two major components of the Bank's operations, the agenda attached to each is often weakly connected. This economic and sector work (ESW) aims to integrate these agendas under an accountability framework. It presents an original analytical approach by focusing on local governance through the lenses of accountability. The ESW intends to provide a normative discussion on decentralization--it is not an empirical study. It establishes a methodological basis for country specific analysis of decentralization and accountability. The ESW is a direct response to a growing interest and demand from country clients, donors, and the Bank to institutionalize participatory mechanisms to enhance the accountability of local decision makers and service providers to strengthen governance and reduce corruption.5 In this respect, the ESW intends to inform client countries about the availability of mechanisms for downward accountability, where public sector accountability approaches are complemented by social accountability approaches. The findings of this ESW could provide guidance to upstream dialogue and downstream implementation in the decentralizing client countries under a comprehensive framework. Furthermore, such an integrated framework could be mainstreamed into the Bank Group's country analytical works (such as CASs [country assistance strategies] and PRSPs [poverty reduction strategy papers]) and inform lending operations of the regions. Within the Bank, the ESW is a collaborative effort of the Social Development Department and the Urban and Human Development Anchors with inputs from other anchors and regional Sustainable Development Network (SDN) departments; it embodies the synergies of the newly created SDN. The ESW is also a product of communication with regions through interviews, informal meetings, and/or activities such as brown bag lunch meetings in a bid to integrate the Bank's supply-side and demand-side governance work. 1.2. OBJECTIVE AND TARGET AUDIENCE Social accountability approaches are broad range of actions and mechanisms beyond voting that citizens can use to hold accountable the state and providers of public services. These approaches could be as simple as pressuring local governments to publish budgets, or more complicated, such as participatory budgeting or social audits where citizen groups inspect, for example, public procurement policies and processes. Often times, when opportunities for governance reform at the national level are limited, there may be local level entry points (World Bank, 2007a).6 This Economic and Sector Work (ESW) examines how social accountability approaches--together with the public sector (supply-side) approaches--can help create such entry points, when institutionalized into local government operations. The main objective of this report is to present a coherent and comprehensive approach for decentralizing countries to achieve a high degree of local government discretion accompanied by robust accountability mechanisms. To this end, this ESW presents an analytical framework of local governance through the lens of accountability. This ESW intends to establish a methodological basis for analyzing linkages between decentralization and accountability. The report is based on desk reviews and interviews--it does not provide empirical research testing the report's arguments. Rather, it provides many country specific examples, and three detailed case studies (on Uganda, the Philippines, and Kerala, India) to demonstrate how the presented analytical framework could provide a useful method for studying accountability 5The exploration of the merits or demerits of decentralization are beyond the purview of this ESW. 6The Governance and Corruption Strategy Paper of the Bank's Development Committee acknowledges that even when opportunities for governance reform at the national level are limited, there may be entry points at the local level. 2 implications of decentralization reforms. It should be noted here that presenting such a framework runs the risk of generalizing issues that are mostly local context specific. The goal here is definitely not to build a one-size fits all prescription as such. This is why the local governance strategy proposed in the final chapter underlines the importance of developing systematic analytical tools and operational approaches that would rely on country specific information on local power structures, interests, and socio- economic conditions. Accordingly, the use of the analytical framework proposed here is very much dependent on local realities and it is the local policy makers in client countries that would ultimately decide what is suitable for their country and what is not. Recently, client countries, donors, and the Bank have expressed increasing interest in and demand for participatory mechanisms, which could enhance local accountability, consequently strengthen governance, and reduce corruption. This report--a direct response to this growing interest--presents various social accountability mechanisms that could complement public sector accountability approaches. It also highlights the operational challenges the Bank's projects on local governance reform must often confront. Accordingly, the findings of this ESW provide, in one comprehensive framework, guidance on both macro policy-making and implementation of decentralization reforms. As such, this report provides the Bank with a basis to develop an integrated local governance strategy to guide lending and non-lending operations. The primary audience for this report comprises public policy makers and local governance stakeholders in client countries, governmental and non-governmental practitioners and policy advisers working on local governance, local development, and public sector reforms, and the Bank staff advising countries on local government reforms. The secondary audience includes academics in the fields of economics, public administration, public policy, and social development. 1.3. ANALYTICAL CONTENT The main argument of this ESW is that decentralization reforms should support both the discretion and accountability of local governments. Decentralization reforms are designed to grant local governments new powers and responsibilities that define a discretionary space on three dimensions: political, administrative, and fiscal. Against the abuse of newly extended powers along these dimensions, the decentralization framework should also establish oversight and accountability mechanisms. Local governments are accountable to upper levels of government as well as to citizens vis-à-vis their domain of discretion (Ribot 2004). In an ideal scenario, decentralization reforms are fully implemented at all dimensions (political, administrative, and fiscal) and by integrating both public and social accountability approaches, accountability is strengthened (see figure 1.1 for a visual presentation of the analytical framework for local government discretion and accountability). However, in reality, countries may be at different stages of decentralization, with varying degrees of focus on the linkages between discretion and accountability.7 This study provides a benchmark framework against which we can evaluate existing decentralization practices and identify gaps leading to accountability deficits (or excessive forms of accountability). Such an exercise would improve our understanding as to how and why decentralization reforms may actually end up undermining good local governance and what kind of measures could prevent such negative outcomes. 7 Please refer to the section on local governance categorizations (section 1.5.) in this chapter reflecting variant combinations of discretion and accountability under decentralization structures. Chapter 4 focuses in more detail on these different categorizations and introduces a discussion on trajectories that countries may follow in reaching the ideal combination of optimum discretion and accountability. 3 With this understanding, this report outlines the main elements of an analytical framework that links discretion and accountability in the context of decentralization structures. First, this framework assesses the role of different actors in accountability relationships using the World Development Report (World Bank 2004b) framework and adapts it to local context to evaluate how local governments, provided with a meaningful level of discretionary powers, enter into accountability relationships with their surrounding local actors. The report uses downward accountability to describe the ability of the government to be accountable towards its citizens, and upward accountability to describe accountability of local governments to higher levels of government. In addition, public accountability is used to refer to measures introduced by the public sector to improve both upward and downward accountability, whereas social accountability is described as an indirect form of accountability referring to civil society-led activities especially oriented towards improving the downward accountability of local governments. 4 Figure 1.1: Framework for local government discretion and accountability Intergovernmental Institutional Design LOCAL GOVERNMENT ACCOUNTABILITY SPACE Vertical cooperation Dispute resolution Public Accountability Approaches Horizontal cooperation Vision/strategy (white paper) Local council oversight Local public financial management (planning, Intergovernmental roles/ Electoral accountability budgeting, reporting, responsibilities Side measures internal control/audit, ply Administrative external audit) accountability measures Sup Political Decentralization Fiscal Decentralization Fiscal Decentralization Separation of powers Expenditure assignment Electoral system Revenue assignment Party system Local Governance Outcomes Intergovernmental transfers Responsive, effective, efficient, sustainable services Borrowing Enhanced political, financial and administrative accountability Greater local control over economic development planning and decision making Strengthened accountability through greater citizen monitoring and vigilance over power holders CDD/Social Accountability Approaches Participatory planning Participatory monitoring and Side Participatory budgeting evaluation Participatory expenditure Participatory management of mand tracking investments/projects De Citizen access to Citizen feedback for services information campaigns (Report cards, Social audits) Local Governance Administrative Decentralization Framework for Downward Source: Authors. Regulatory discretion Accountability Procurement discretion Employment/civil service discretion 5 1.3.1. Actors of Accountability The WDR 2004 framework for making services work for people, especially the poor, is founded on strengthening accountability in the relationships between four key actors involved in service delivery, namely, citizens-clients (i.e., the poor and vulnerable); policy makers (politicians, lawmakers, regulators); organizational providers (ministry, department, agency, private sector, non-profit, and so on); and frontline providers (those who comes in direct contact with clients--teachers, doctors, engineers, and so on). It also identifies four types of relationships that govern the interactions between the actors: (i) politicians to citizens: voice and politics; (ii) organizational provider to the state: compacts; (iii) frontline professionals to the organizational provider: management; and (iv) provider to the citizen-client: client power (see figure 1.2). Decentralization incorporates a new set of actor(s) and additional relationships in this framework: local governments (see figure 1.2). Schroeder (2006: 5) categorizes four different actors that are typically relevant in local accountability systems: local residents, local governments, producers of local government services, and higher level governments (including central government). Schroeder depicts local governments as the provision units rather than producers of services. He describes provision as governance decisions regarding what, how many, and the quality of services that are to be made available in the locality, how those services are to be financed, and how they are to be produced. He ascribes the production tasks of combining inputs to produce outputs to local public employees, private contractors, or even by higher-level jurisdictions under contract to the local government. In this sense, "bureaucracy (and its supporting personnel) is the set of actors that either carry out the production of local services or helps to oversee the private (or public) contractors that serve as production units" (Schroeder 2006: 9) We will use the term local appointed officials interchangeably with local bureaucracy and emphasize the important role that local elected representatives play in local accountability systems, especially in terms of oversight of local governments. Beyond service delivery, we will be interested in the role local appointed officials play in all aspects of fiscal, administrative, and political decisions, and their accountability relationship with local elected officials, with other local public employees (i.e., employees of other government agencies), and of course, with local residents. We will also explore the role of nongovernmental organizations (NGOs), civil society organizations (CSOs), community-based organizations (CBOs), resident associations, traditional leaders such as chiefs, or other powerful individuals in mediating accountability relationships between local citizens/residents and local governments. Each of these actors has particular relations of accountability with other local actors and has certain types of powers to deliver the account. These relations depend on the historical, social, and political constitution of the powers of each actor, which may be based on ideology, wealth, heredity, election, appointment or other means (Agrawal and Ribot 1999). Analyzing accountability relationships among these actors can be quite challenging. First of all, local actors and central governments may pursue different, and sometimes conflicting roles and incentives. Such variance in roles and incentives makes it even more necessary that we clearly identify actors and the nature of accountability relationships among these actors. Second, the concept of accountability itself is elusive and its use in different contexts can sometimes be confusing-- making any analysis of accountability even more challenging (Mulgan 2000). Our analysis pays particular attention to the way local governments (both appointed and elected local officials) establish and maintain their accountability relationships with their surrounding environment containing these actors. 6 Figure 1.2: Actors of accountability Center Politicians Policymakers The framework of accountability relationships in service delivery Devolution TheState Subnational government Delegation Politicians Policymakers Politicians Policymakers Deconcentration Longrouteofaccountability Voice Voice Compact Compact Citizens/clients Providers Client power Citizens/clients Shortroute Public&privateutilities Nonpoor Poor Frontline Organizations &providers Client power Coalition/inclusion Management Services Nonpoor Poor Frontline Organizations With decentralization new tier is added Services Source: Adopted from World Bank (2004). In the local space where many actors interact with each other, we define accountability as a relationship in which one of these actors (or group of actors) is able to keep in check the power of another actor (or group of actors). Accountability is broken into: 1) answerability or the obligation of one party to explain and justify actions to another; and 2) enforcement or the ability of one party to sanction the other when the explanations are found inadequate (Ackerman 2003; Brinkerhoff 2004; Goetz and Jenkins 2005; Bovens 2006). Goetz and Jenkins (2005: 8) remark that accountability then is "as much about public debate and discussion as it is the sheer imposition of will" and that answerability and enforcement are its necessary components--with neither being sufficient by itself. Answerability is composed of explanation and information.8 Enforcement is constituted through positive and negative sanctions (Manin, Przeworski, and Stokes 1999). After identifying these roles and incentives, there is also a need to further qualify the accountability actors as to whether or not they serve as actors or forums and the direction of this relationship. Accountability, defined as the obligation to explain and justify conduct (Bovens 2006), implies a relationship between an actor--the accountor--and a forum--the account-holder or accountee (Politt 2003). According to Bovens (2006), the relationship between the actor and the forum consists of three elements: (i) obligation of the actor to inform the forum about his conduct; (ii) the forum's power to interrogate the actor and to question the adequacy of the information or the legitimacy of the conduct; and (iii) the forum's power to pass judgment on the conduct of the actor--it may approve of an annual account, denounce a policy, or publicly condemn the behavior of the actor. Bovens' (2006) definition of narrow accountability also includes the possibility of sanctions--not the actual imposition of sanctions. As he notes, the possibility of sanctions "makes the difference between non-committal provision of information and being held to account" (Bovens 2006: 10). In this accountability relationship, both the actor and the forum can be either an individual or an organization. For example, in our study the actor can be a local or central government official (i.e. minister, mayor, member of parliament, or local council etc.) as well as a public sector institution 8 Information is the evidentiary basis of explanation, which limits the possible range of rationality. It is the requirement to provide the data on which a decision is justified (Goetz and Jenkins 2005). 7 belonging to central or local governments (i.e. ministries, departments, councils, and so on). The account- holder, or the accountability forum, can also be the local or central government officials or institutions. However, in democratic governance the ultimate account-holder is the general citizenry. In this study, our main concern will be to search for ways in which the local accountability links between public officials or institutions (accountor/actor) and citizenry (account-holder/forum) can be strengthened within decentralization contexts. 1.3.2. Decentralization, Local Government Discretion, and Accountability One of the critical assumptions in this report is that representative local governments are basic institutional elements of decentralization. Accordingly, granting them with meaningful discretionary powers (local government discretionary space) is both necessary and desirable for effective local governance (Mawhood 1983; Ribot 1996; Romeo 1996; Crook and Manor 1998; Agrawal and Ribot 1999; Mandondo 2000; Smoke 2000). A decentralization program can achieve this goal by transferring the legal and political authority, administrative decision-making abilities, and fiscal powers to plan public expenditures and manage public functions from the central government to sub-national governments. The goal is to improve the quality of public sector decision making by granting substantial discretion to local governments in responding to citizens' needs and preferences and being accountable to them. If local governments do not have discretion power, they "cannot gain the legitimacy they need to effectively represent local populations" (Ribot 2003: 57). Provided with a meaningful level of discretionary space, local governments then need to be checked and balanced against abuse of using their new discretions. In this respect, central governments have a set of accountability instruments to enforce implementation of their policies, rules, and regulations. However, such formal rules and norms can be vague and incomplete to deal with every situation and circumstance. They also tend to be oriented more towards upward accountability--accountability towards upper levels of government. Accordingly, a great deal of work is needed to strengthen downward accountability that emphasizes accountability mechanisms that can link local public administrations with their surrounding communities and give the poor a greater voice. Public sector accountability tools alone cannot achieve this outcome, even if public officials are willing to be responsive to citizen demands. A new and integrated approach marrying supply-side public sector accountability with demand-side social accountability is needed. The following sections describe the linkage between discretion and accountability in more details. Intergovernmental Relations and Local Government Discretionary Space The laws and regulations governing political, administrative, and fiscal institutions and processes, which are decided primarily by the central governments, define local government discretionary space in decentralization. The process is often very complicated. Many countries face problems in designing decentralization policies to structure a viable vertical distribution of functions and powers between central and local governments. Political arrangements set the rules for local populations to elect and interact with their local leaders, and they define the degree of oversight functions that those local leaders enjoy over local bureaucracy. These arrangements are established through the electoral system, the party system, and the oversight powers assigned to locally elected leaders in planning, budgeting, and service delivery. Administrative arrangements establish the rules for defining the local regulatory environment, procurement, and civil service that determine the autonomy and incentive structure for enabling accountable and well-functioning local bureaucracies. Local governments should be endowed with at least three administrative discretions. First, they should be able to make, change, and enforce laws and regulations pertaining to local administrative affairs. Second, they should be able to govern the local 8 procurement system. And third, they should be able to have control over the local civil service and employment policies. Fiscal decentralization is about sorting-out roles and responsibilities among different levels of governments with the intent of transferring some of fiscal decision-making powers from central to sub- national governments around four pillars: expenditure assignment; revenue assignment; intergovernmental transfers; and sub-national borrowing (Bird 2000). These pillars focus mainly on the existence of revenue and expenditure autonomy, a transparent and rule-based transfer system, and sound local government borrowing mechanisms. Expenditure autonomy determines the relevance of local governments in service delivery by assigning service delivery responsibilities. Revenue autonomy and transfer systems, on the other hand, define the funding level for local governments to deliver these services. Borrowing powers provides local governments space for additional decision making in investments. The absence of an appropriate fiscal discretionary space for local governments implies that intergovernmental fiscal arrangements are incomplete. In many countries, such incomplete intergovernmental fiscal frameworks load local governments with new mandates and responsibilities without the necessary resources to undertake them. From Discretion to Oversight and Accountability The discretionary space as discussed above implies a significant degree of autonomy and an enabling environment for local governments. Yet such autonomy also means that it is just an action space, which does not necessarily guarantee the kind of local government behavior that would prioritize accountability towards citizens. Despite the benefits and practicality of discretion, there are tendencies for misuse and abuse of discretionary power at the local level--as well as at the central level. In many cases, local governments use discretion for extracting rents; they are also unduly vulnerable to influence and capture by elite groups due to a lack of checks and balances on discretionary power. In the absence of a robust legal and institutional framework that institutionalizes accountability--both upward and downward-- discretion as provided by decentralization can actually significantly impair the perceived benefits of decentralization. In a decentralization framework, local governments are accountable to upper levels of government as well as to citizens as allowed by their discretionary space. As part of upward accountability structures, local government officials and agents have mandated functions that define their accountability to higher political, administrative, and fiscal hierarchy. Their job is to implement specific mandates set by central government actors and their administrators to whom local governments are accountable. A number of accountability mechanisms are at the disposal of central governments to guarantee compliance with mandates. These include all means ". . . through which state agencies check one another (division of powers between legislative, executive and judiciary, auditing agencies, ombudsmen, etc.)" (Goetz, Joshi, and Moore 2004: 5). Downward accountability relates to the ability of the government to be accountable towards its citizens. The term downward accountability is synonymous with vertical accountability, a term used in the social development literature; it is the ability and mechanism of the state to be accountable towards its citizens. Local governments account to forums (i.e., other levels of government or citizens and citizen formations) on three dimensions: political, administrative, and fiscal. They account on how they fare regarding political performance (i.e., through elections to citizens, or how are local candidates financed for their campaigns), administrative performance (i.e., about who they hire and fire and why; or what kind of tenders they grant to whom), or fiscal conduct (i.e., how they spend their money and the resources are for fiscal decisions). Political accountability is a process whereby citizens hold their governors to account for their behavior and performance directly through elections (Aucoin and Heintzman 2000) and elected local officials oversee local executives. It also involves activities for increasing civic awareness about the policy 9 performance of local governments and direct citizen involvement in policy decision making beyond elections. Administrative accountability refers to local civil servants being accountable to their top administrative officers and to outside bodies such as public audit officers, ombudsmen, and regulators or a particular administrative agency, or a board/committee about the use of their administrative discretion. Fiscal accountability seeks to ensure the transparent management and expenditure of public finances in line with policy decisions. It requires that governments manage finances prudently, that they integrate their financial and non-financial reporting, control, budgeting, and performance, and that they report comprehensively on what they have achieved with their expenditure of funds (Sahgal and Chakrapani 2000). Improving accountability in all these dimensions requires approaches to promote downward accountability that combine both supply- and demand-side efforts. The following section explains how these approaches can complement each other for this purpose. Approaches to Promote Local Government Downward Accountability: Public vs. Social Accountability The downward accountability of local governments are strengthened by integrated approaches that marry supply-side public sector accountability with demand-side social accountability instruments. The supply side of accountability, which is also known as public accountability, is the hallmark of and a sine qua non for good governance (Bovens 2005). It is the obligation of public authorities (governments, elected representatives, corporate, and other governing bodies) to explain publicly, fully and fairly, how they are conducting responsibilities that affect the public in important ways. Public accountability refers to the institutionalized practice of account giving--it focuses on public sector managers who spend public money, exercise public authority, and manage a corporate body under public law. Public account giving provides political representatives9 and voters with the necessary inputs for judging the fairness, effectiveness, and efficiency of their governance system (Manin, Przeworski, and Stokes 1999). Until recently, efforts to foster good governance focused on strengthening the supply side of accountability within state institutions. Accordingly, donor initiatives supported institution building in developing countries to increase the supply of governance processes by reforming public institutions. These efforts included improving public financial management systems, establishing audit institutions and anti-corruption commissions, rehabilitating procurement systems, and reforming civil service structures. Recently, the development community has concluded that domestic demand for accountability, originating within civil society entities and the public at large, is at least as important for development as supply-side mechanisms. Also referred as social accountability, demand side refers to an approach toward building accountability that relies on civic engagement--in which ordinary citizens and/or civil society organizations demand accountability. Recognizing the limitations of both electoral and public accountability mechanisms, demand-side/social accountability approaches require concerted civic education efforts during the decentralization reform process. Such new understanding encouraged an 9 The formal doctrine of elected representatives' responsibility rests upon the notion of representative democracy and parliamentary sovereignty. The elected representatives of people--parliament--bring the government--the executive--to account. Parliament enacts administrative laws to enforce accountability. On the other hand, poor governance and lack of public sector accountability may result from a number of factors including civil service incompetence, lack of efficient institutions, and pervasive corruption (Schaeffer 2005). 10 expansion in the repertoire of instruments through which citizens can hold the state to account, beyond voting. These instruments include traditional practices such as public demonstrations, protests, and investigative journalism, as well as more innovative ones such as participatory budgeting, social audits, or citizen report cards. The main expectation of this kind of new approach is that it will "make citizens aware of their continued role of principals beyond election day. To the extent that a particular decentralization reform allows for such a role, a critical element to the successful implementation of the reform then becomes citizens' awareness of their workload as principals beyond what is required on election day . . . It is thus essential to enhance the skills, capacities, and willingness of citizens to carry out the additional tasks assigned to them as a result of the decentralization reform process in order for that process to overcome the limits of elections as the sole means of agent selection and oversight" (Hiskey 2006: 7). As social accountability emerged in response to limitations in supply-side approaches, the real challenge is now how to bring those approaches together in practice so that local governments could be downwardly accountable to their citizens. This report attempts to address bring them together under a coherent and comprehensive decentralization framework. The following section provides a snapshot of the framework that this report is built on. Subsequent chapters focus in more detail on each component of this framework. 1.4 A FRAMEWORK FOR LOCAL GOVERNMENT ACCOUNTABILITY As described above, this report presents a theoretical framework for accountability that has two critical components: discretion and accountability. This accountability framework draws from both Bovens' characterization of accountability as the obligation to explain and justify conduct between an actor and a forum and the WDR-2004 framework for making services work for people. We identify actors at local accountability space based on WDR framework and use Boven's accountability analysis to determine directions and types of accountability relationships among these actors. We then build on this accountability framework to establish the linkage between accountability and good local governance (see figure 1.3). The following brings together all the elements that define accountability relationships to explicate such a framework for analyzing and improving accountability in the context of decentralization. In this framework, · Local government is at the center of a web of accountability relationships, given the primary focus on decentralization. In some relations, local government is the actor (accounter) while in others it is the forum. · Decentralization reforms grant local governments new powers and responsibilities in three dimensions: political, administrative, and fiscal. These dimensions provide local governments a discretionary space through which they are held to account (chapter 2). · Accountability relationships may be driven in multiple directions: upward, downward, or diagonal. Through their discretionary space, local governments are accountable to upper levels of government (upward accountability) as well as to citizens (downward accountability). Upward and downward accountability of local governments are distinct but interlinked, deeply affecting each other. · Recognizing the need to have safeguards against abuse of local discretion, public accountability mechanisms are introduced for oversight and accountability of local governments. 11 · In response to limitations in public accountability mechanisms, new forms of social accountability have emerged. Social accountability mechanisms enable a more direct engagement of citizens with government. · Yet, there is also a need to bridge these two approaches of accountability (chapter 3). This will ensure more integrated systems where citizens have the ability and opportunity to demand accountability and local governments have the means and capacity to respond to these demands (chapter 4). Figure 1.3: Good local governance and accountability Central government sloot ng rot sec Cooperation desi.tv cilbuP ytilibatn stcartn & compacts Co rtenifoseluR go cou ac Other Regional planning Local Compacts Service local/provincial providers lai government Service delivery soc sloot governments Competition contracts Regional investments dna or NGOs/ Voice Private ct ytilibatn se CSOs ti noitarg Sector cilbuP accou eci Vo snoitceEl niotapiictraP Ex Mi Local government charges discretionary space power Citizens UserClient Formal rules of accountability Informal norms of accountability 12 1.5 LOCAL GOVERNANCE CATEGORIZATIONS As mentioned before, local governments with a high degree of discretion and with a high degree of accountability represent an ideal scenario in decentralization structures (quadrant III in figure 1.4, below). This scenario assumes that decentralization in all dimensions (political, administrative, and fiscal) is fully implemented, and it successfully integrates both public and social accountability approaches. In reality, countries may be at different stages of decentralization, with varying degrees of focus on the linkages between discretion and accountability. Countries that are at high end of discretion but lacking focus on accountability falls under quadrant I in figure 1.4. This kind of "high discretion­low accountability" combination may actually create perverse incentives for local governments, making them vulnerable to elite capture or prone to reckless decision making. Quadrant II represents countries that provide local governments with a limited discretionary space. Local governments in these countries are also subject to weak accountability systems. With limited discretion and accountability, local governments in these countries are mere extensions of national governments and are not accountable to citizens. Quadrant IV, on the other hand, signifies countries that are more focused on accountability structures governing local governance systems. Yet, in the absence of high discretion for local decision making, local governments in these countries are more likely to be overloaded by accountability requirements. Absence of local discretion also implies that local governments have skewed incentives to be heavily accountable to upper levels of government, with little incentives for being accountable towards citizens--as most critical decisions are made by the central governments. Chapter 4 focuses in more detail on these different categorizations and introduces a discussion on trajectories that countries may follow in reaching the ideal combination of optimum discretion and accountability (quadrant III). Figure 1.4: Discretion and accountability graph Discretion High I III II IV Low Accountability Source: Adopted from Levy (2006). 1.6 SUMMARY AND OUTLINE OF THE STUDY Decentralization reforms are not uniform across countries. We observe varying degrees of decentralization, often ignoring the link between discretion and accountability. Accompanied by weak capacities at local levels to undertake new powers and responsibilities, it is not uncommon to see 13 reactions to decentralization by central governments in the form of stricter control over local governments, ending up with stripping local governments of their powers and discretions. Yet, reducing powers and functions is not a form of accountability; it just makes local governments irrelevant. It is not the increased discretionary space but lack of effective accountability systems that undermine local governance. In other words, discretion and accountability do not work against each other, provided that both supply- and demand-side accountability approaches work well, supporting one another. In the best case scenario, local governments enjoy significant powers and discretion, and they are downwardly accountable through effective public and social accountability measures. In a nutshell, this kind of decentralization reform foresees empowered local governments and requires central governments to move away from micromanaging local affairs to oversight and to providing an enabling environment as well as setting/observing minimum standards. The analysis in this report distinguishes and links discretion and accountability recognizing different, albeit sometimes conflicting, roles and incentives that local actors and central governments might pursue. Such variance in roles and incentives makes it necessary that we clearly identify actors and the nature of accountability relationships among these actors. This is why we started to discuss the analytical content of this report by introducing the WDR-2004 framework and adapting it to a local context. Adding to the WDR framework, there is also a need to further qualify the accountability actors whether or not they serve as actors or forums and the direction of this relationship. Local governments are either actors or forums in these accountability relationships and they are restricted by their discretionary space as defined by intergovernmental relations in pursuing these relationships. Based on this premise, in the following chapters, we will study political, administrative, and fiscal domains of local government discretion and accountability. Chapter 2 discusses that formal structures and informal practices in decentralization process that provide local governments the necessary space in which they would primarily be accountable to citizens. We argue that, in a given environment conditioned by many external political, economic, and social factors in a country, different policy choices are still available to decision makers. Chapter 2 focuses on such design choices along the lines of political, administrative, and fiscal decentralization that are available to policy makers in defining local government discretionary space. Chapter 3 examines approaches to improve transparency and accountability within local government structures. It also highlights deficiencies in supply-side public sector accountability approaches, and discusses how these approaches are complemented by demand-side social accountability initiatives. The chapter's main message is that, in the absence of a more comprehensive approach to improve accountability of local governments that would bring together public sector and social accountability mechanisms, there is a risk that decentralization may actually have perverse effects on local accountability. Chapter 4 will further elaborate these discussions and explore options for institutionalizing such an integrated approach to decentralization and accountability. It also intends to discuss the need to mainstream this kind of integrated approach into operational projects of donor community. In doing so, the chapter highlights challenges that client countries as well as donor community face in operationalizing some of the recommendations provided in this report. 14 Chapter 2: Decentralization and Local Discretion 2.1 INTRODUCTION This chapter describes the role of decentralization reforms in providing local governments with a discretionary space so that local governments can be responsive to citizens by using their discretion and be held accountable for their performances (for both the process and the outcome). Decentralization reforms grant local governments new powers and responsibilities that define such a discretionary space in three dimensions: political, administrative, and fiscal. In our analytical framework introduced in chapter 1, we defined accountability as a function of this discretionary space. With the implementation of a decentralization program, the legal and political authority to plan projects, make decisions, and manage public functions and funds is transferred from the central government and its agencies to sub-national governments. This chapter focuses on the intergovernmental division of powers, roles, and responsibilities in decentralization that defines the boundaries of the local discretionary space through political, administrative, and fiscal arrangements. These arrangements are external to local governments as central governments primarily decide the laws and regulations governing political, administrative, and fiscal institutions and processes. Political arrangements set the rules for local populations to elect and interact with their local leaders and they define degree of oversight that those local leaders have over the local bureaucracy. Administrative arrangements establish the civil service rules that determine the autonomy and incentive structure for enabling accountable and well-functioning local bureaucracies. Fiscal arrangements determine the relevance of local governments in service delivery by assigning service delivery responsibilities. Furthermore, they define the funding level for local governments to deliver these services. This chapter is not intended to provide an overview of intergovernmental arrangements around the world, nor is it a stocktaking exercise of local governance practices. The chapter's scope is limited to the analysis of political, administrative, and fiscal dimensions of decentralization to the extent that they affect local government discretionary powers within public sector structures. The following sections provide a detailed analysis on these three dimensions. 2.2 POLITICAL DIMENSION The political dimension of decentralization sets the rules for local populations to elect and interact with their local leaders. These rules also define the degree of control that those local leaders have over the local bureaucracy. The political setting under which local governments operate could have significant impact on and be affected by local power relationships (especially among local officials, local businesses, elites, and families). Accordingly, this setting shapes accountability relationships in decentralization contexts. Box 2.1, below, provides a series of questions to determine as to whether political arrangements promote downward accountability. The ideal political setting in decentralization contexts requires a suitable environment for local elected leaders to act independently and in line with the demands of local populations, even if these demands are in conflict with their own parties or with the central government. Such a suitable local environment implies institutional arrangements for the separation of powers among the executive, legislative, and judicial bodies; election laws and electoral system; and existence and implementation/functioning of political party laws and party system. The following sections expand on how these arrangements play important roles in defining the political setting for local government actions. 15 Box 2.1: Relationships between the political dimension of decentralization and local political settings · How do election systems affect the local political setting? · What is the impact of the structure of political parties and of the party system, and the general political culture on accountability relationships at the local level? · Is there a clear separation of powers among the executive, legislative (council), and judicial bodies at the local level? Source: Authors. Existence and Functioning of the Institutional Separation of Powers at the Local Level An effective decentralization strategy requires the existence of the separation of powers among key institutions that provide an appropriate political setting for different local actors to perform their roles in local rule making. This is possible by introducing countervailing institutions to the executive bodies at the local level, such as the legislature (local councils), judiciary (with local specialization), and civil society groups. In such an environment where roles and functions are clearly identified, local councilors are then expected to undertake independent oversight functions over local executive bodies, local courts take the role of impartially resolving conflicts,10 and civil society performs monitoring roles over these other institutions. A key prerequisite for the existence of separation of powers as such is the presence of a specialized local court system that is able to resolve local conflicts. In the absence of an effective court system that can respond to local conflicts, the local political setting would not be conducive to downward accountability. An impartial and efficient judiciary system that is accessible and affordable can play a significant role in expanding the space for the exercise of powers extended through decentralization. In all of these, there is a need for special attention to ensure the access of poor and vulnerable groups, including those living in remote rural areas, to the justice system in case of conflicts with local authorities over issues affecting their livelihoods. Similarly, it is equally important that political decentralization provides local councilors with the powers and functions necessary for overseeing the work of executive agencies or appointed local officials-- whether it has investigative functions and capacities for obtaining information on the work of local government departments and agencies, or whether the local council actually exercises the authority to investigate or censure the operations of administrative departments and agencies (IDEA 2004: 10). Finally, the availability of effective citizen involvement in local government affairs provides an important institutional check and balance against the local executive bodies. The impact of organized interest groups, community-based organizations, civil society, and the media is elaborated in chapter 3. We will suffice here to say that active citizen involvement implies much more than just voting during election times. Some examples of how citizens can play an active role in demanding accountability from local governments will be discussed under "direct citizen involvement" in chapter 3. 10 Chapter 3 discusses how these roles can be strengthened to improve local political accountability. Here we introduce the basic political setting where oversight and judiciary roles can effectively be undertaken. 16 Existence and Quality of Electoral System for Mayors and Council Members Despite many weaknesses, elections remain the principal method whereby all eligible local residents can potentially have a voice in the outcome and can hold decision makers accountable (Schroeder 2006: 9). It is necessary to closely look into electoral system as it applies at the local level. Whether the system is based on majority (plurality), proportional representation, or a combination of the two principles could have significant influence on the local political processes, how candidates are elected, and how citizens are involved in these processes. Party systems also respond to the realities of local politics and electoral systems, and produce corresponding structures at the local levels. Electoral systems affect the accountability incentives of elected local leaders--during and between elections. Plurality systems aim to produce a majority winner. In the process, elections can also end up disproportionately representing the winning party. In proportionate systems, allocation of seats is roughly proportionate to votes, and in that sense, it reflects public's choice better. However, proportionate systems are also vulnerable to elite capture and to undesirable candidates making backdoor entry to the legislature. Many variations that combine (mixed systems) plurality systems with proportionate systems are also possible. Depending on its features, an electoral system may favor big parties, undermine alternative voices and dissent, or encourage strict hierarchies within parties (due to procedures for candidacy, party members may feel a need for closer ties with "the center" rather than their local constituency). Thus, when elections are based on proportional representation, which is significantly influenced by the national party, such elections may not fully provide the desired accountability to hold local policy makers accountable for their actions.11 More importantly, electoral systems can be structured in such a way to systematically exclude certain groups. Or they may encourage political parties to simply "win" the votes of particular groups over others. In short, the ways in which the local electorate's preferences translate into local council seating allocation may affect local campaigning and the performance of elected leaders after the elections. The qualities of local electoral systems have a significant impact on how issues of importance in the lives of different groups--such as the poor and vulnerable, women, youth, disabled, or marginalized groups--affect the local governance agenda. A system that favors multiparty systems with a differentiated party agenda would lead to a very different outcome in the local council agenda before and after elections, as opposed to a system favoring only one party and a central agenda. Nature of the Party System and Party Structures The extent to which citizens' voices are effective through the electoral system also depends on factors beyond the citizens' control. Representative democracy provides citizens with an opportunity to vote, but without certain mitigating features, citizens are often forced to accept the agendas set by political parties. Party system features in a particular country mainly depend on elements such as the number of parties competing in local elections, rules governing the financing of parties, rules governing the participation of disadvantaged groups, such as women or certain minority groups, and the availability of parties based on ethnicity or religion. 11Conventional political theory, however, advocates some form of proportional representation for new democracies as a means to ensure a voice for smaller groups thereby giving them a stake and presence in the new democracy rather than shutting them out (World Bank 2000: 114). While such an approach may certainly be necessary to allow disparate regional and ethnic interests to have a distinct voice in a new democracy, such proportional elections in a country with a strong national party, which greatly influences who gets on the local party list, may undermine the accountability of sub-national elected officials to local citizens. 17 Box 2.2: Uganda's "no-party" system In Uganda, the government until recently followed a "no-party" system because the National Resistance Movement (NRM) leadership held that a multi-party system would revive ethnic and religious cleavages. Blair (2000) argues that the existence of competitive political parties is a necessary requirement for encouraging a culture of local accountability. Azfar et al. (2001) argue that the no-party system affects the local level more than the national level. The state and the local councils have greater authority in the provinces than in the capital and other large cities, and observers cite evidence that both the local councils and the resident district commissioners do much of the movement's work in the interior, providing platforms for candidates, spreading the movement ideology, administering political training, and other functions. Thus, the perpetuation of the no-party state in Uganda may have compromised its development. Sources: Blair 2000; Azfar et al. 2001; Francis and James 2003. An analysis to determine the political setting defining the boundaries of local government actions should start with the very basic question about the party system--whether or not more than one party can genuinely compete for local elections (see box 2.2 for an example where a no party system dominates). In some countries, even though competitive multi-party elections may occur at the national levels, local electoral systems do not have a similar competitive process. Restriction of local party competition significantly reduces the ability (and willingness) of local government leaders, especially the elected ones, to make and implement decisions independently and in line with preferences of local populations. Moreover, opposition parties (or simply nonruling party candidates) are important sources of information during and between elections for local citizens and other interested parties about the incompetence and failures of local policies, or corrupt activities and the misconduct of local officials. Other elements for defining the relationship between the party system and the discretion of elected local government officials are related to the rules that govern the participation of disadvantaged groups, such as women or certain minority groups, and the availability of parties based on ethnicity or religion. In the absence of clear rules that stipulate the inclusion of certain disadvantaged or minority segments of society, a party system may easily engender a system of dominance by the majority groups and more powerful social groups. Exclusion of the disadvantaged and minority groups--be it based on race, gender, ethnicity, or religion--from the party system reflects a party agenda (and inevitably, a local council and government agenda) that might be favoring the majority and the more powerful, undermining the interests of the excluded. In the absence of sanctions against such exclusion, downward accountability of locally elected leaders is damaged, leading to a skewed policymaking and greater rent seeking. Similarly, party loyalty structure can also have direct or indirect effects on the ability of local political leaders to exercise their new powers. Membership structure of the parties, the role of national party leaders in selecting candidates and preparing party lists for local elections, and the hierarchy structure within parties can shape the behaviors of political parties in such a way that national priorities and national leaders of the party may dominate local priorities and agenda (see box 2.3 for a discussion on party systems and within party structures in West Africa). For example, the way that candidates for local elections are chosen is an important factor that affects party loyalties. If national party officials formally or informally control an individual's candidacy, then the accountability incentives of local political leaders are expected to be upwards, at least during the candidacy process. South Africa represents an example of such a situation. In South African municipalities, half of the elected local councilors come from a party list that is, in part, influenced by the national party. The other half is elected from wards, but even there the national party influences who stands for election. In the final analysis, both party list and ward candidates are accountable, largely, to the national party that put them on the party list for their election, not to the local citizens who voted for the party. Such proportional representation gives tremendous power to the people who control the creation of the party list and determines who stands for ward elections. In particular, it allows the party boss to reprimand and ultimately "fire" council members who do not follow the 18 national party line (Wittenberg 2003). On the contrary, decentralized party loyalty structures and hierarchies may actually encourage local political leaders to focus on and respond to the demands of local constituencies. Party system and party structures interact with electoral systems. Both party systems and party structures can affect the discretion of elected local leaders and how they use their discretion in conducting local level politics. If the interaction between the electoral process and party structures is not set properly, local elected bodies may actually follow an agenda that fits their party goals nationally, even if it conflicts with the interests and needs of the local populations--a situation that is goes against the spirit of local discretion and downward accountability. Box 2.3: Strengthening political parties in Africa Are African political parties sleeping ghosts that wake up to haunt the goodwill and political resources of the electorates only during election years? This was one of the questions posed in the "Inter-regional Consultative Forum on the Role of Political Parties as Agents of Democratization," held on October 19­20, 2006, in Windhoek, Namibia. International IDEA, in partnership with the Southern African Development Community Parliamentary Forum, organized the forum. The forum highlighted the following trends for political parties in Africa, which differ widely among the sub- regions: In West Africa, political parties are now flourishing--with 130 parties in Côte d'Ivoire, 77 in Senegal, and 22 in Liberia--although this is no guarantee in itself of competitive party politics. Poverty and illiteracy in the sub-region make it difficult for political parties to raise the required resources for effective political work and mobilization. In reality, many parties exist in name only and not in practice. Political parties are weak in organizational capacity, internal democratic culture, financial resource mobilization, social inclusiveness (especially of women, youth, and the physically challenged), and in their ability to articulate how to meet development challenges. Party systems in Southern Africa vary from the no-party system in Swaziland to multi-party systems in Mauritius. In between these two systems, most countries operate with one dominant party and fragmented, enfeebled, and disjointed opposition parties (such as in Botswana, Namibia, and South Africa). In other countries, party systems are in crisis (such as in Malawi, Zambia, and Zimbabwe), due to minority regimes in the first two, and dictatorial rule in the latter. Lesotho stands out as a case where the party system remains non-institutionalized due to the serious state of fragmentation that often afflicts both ruling and opposition parties alike. In East Africa, Kenya and Tanzania seem to have increased multi-party competition and individual freedom, which is also the case to a lesser extent in Uganda. Opposition parties in Uganda and Ethiopia operate under severe constraints, and in Sudan, they can only resume activity at the consent of the government. Still the party systems in these countries bear some similarities: they lack managerial capacity and show structural weaknesses when not associated with the incumbent government; there is a lack of adherence to formal rules, regulations, procedures, and programs; leadership is centered on a dominant personality, family, or clique, and major decisions are often made at this level; there is a tendency for parties to repeatedly breakup and fuse with others; they survive with a weak and unreliable financial and human resource base; and many parties are moribund in between elections. The forum participants concluded that political parties and party systems in Africa face challenges of institutionalization, representation, and democratic practices. They suggested that more efforts be directed towards a more level playing field, where reforms on the funding of political parties, women's representation, internal democracy of political parties, and the role of incumbent parties are needed for promoting party development and democratization. Source: http://www.idea.int/africa/strengthening_pp.cfm. 2.3 ADMINISTRATIVE DIMENSION Empowered by the new political setting formulated through decentralization, local governments need to be endowed with administrative autonomy concerning affairs under their jurisdiction. A local government that can make and enforce regulatory decisions, govern a procurement system (based on national 19 standards), as well as control and manage its workforce can be considered administratively autonomous. Box 2.4 provides a series of questions to determine whether local governments really enjoy an appropriate administrative autonomy. Only this kind of autonomy would expand the discretionary space of local governments so that they could be held accountable for their administrative actions. In this arrangement, local governments would be primarily responsible for their decisions and actions, and they would be empowered to control and manage their workforce to improve service delivery. Box 2.4: The relationship between administrative decentralization and local discretion · Are there clear sets of rules defining the administrative autonomy of local governments (as described in the Constitution, laws, decrees and other regulations), or are the rules ad hoc and granted as privileges? · Can the local governments make, change, and challenge regulatory laws affecting their jurisdiction (i.e., city planning, environmental regulations, and business regulations)? Do the local governments possess the discretion to enforce local regulatory decisions (fines, enforcement of fines, and so on)? · What are the limitations with respect to decision-making power for employment/civil service policy? Source: Authors. To extract downward accountability, local governments should be endowed with at least three administrative discretions, as follows: · First, they should be able to make, change, and enforce laws and regulations pertaining to local administrative affairs. · Second, they should be able to enjoy a degree of autonomy over the local procurement process. · And third, they should have control over local civil service and employment policies. In enabling these, local governments may be subjected to certain national level benchmarks and standards. The following subsections present detailed discussions on each of these three discretions. Our thesis is that local administrative autonomy based on these three discretions increases the ability of local governments to respond to issues affecting populations in their jurisdiction. Granting administrative autonomy to local governments can get very complicated. First, what is on paper may not match what is on the ground. Second, there may be different, conflicting practices that endanger the administrative autonomy of local governments. For example, central governments may introduce complex structures for local rule making that dilute local administrative autonomy on key local economic and social development issues. Central governments may even set counter-effective policy environments that undermine elected local governments by granting deconcentrated units of central government the upper hand in rule making and authorization procedures (complicating lateral accountability arrangements). This situation may create a sense of frustration among the local population about the local government's (in)ability to respond to their needs and resolve local conflicts. In addition, local governments themselves may create adverse practices, generating administrative rule making that is ineffective or detrimental to local socio-economic development. This may very well be the case due to a lack of experience, of capacity, or simply of foresight of local governments. In this regard, we will also argue at the end of this section for the need to clearly define administrative autonomy of local governments by rules and regulations (i.e., through a Constitution, laws, decrees, and other regulations). The absence of clear sets of rules can result in contestations across levels of government, often leading to "what might be called imbalanced decentralization (Campos and Hellman 2005: 242). 20 The Local Government's Ability to Make, Change, and Enforce Regulatory Laws/Decisions To claim administrative autonomy, local governments should have a minimum set of powers and capacities to initiate regulatory legislation on critical issues affecting their jurisdiction. For example, they need the authority to approve and issue generally binding ordinances regarding public matters in their jurisdiction, subject to national and state level laws. In places where they are empowered with significant powers, local government administrative powers usually extend to local economic development, land use planning and management, zoning regulations, and public safety. In certain cases, these powers could even stretch to some aspects of public health, social protection, education, and environmental protection. Local governments also need certain tools such as the power to sanction and punish for non-compliance to ensure a meaningful level of administrative autonomy on local affairs. Yet these powers do not automatically mean criminal prosecution. An often-used tool for enforcement by local governments is administrative penalty, done through a locally issued ordinance and administered by a local governing body, such as a local council or a special committee. The penalties usually entail monetary fines or revocation of certain licenses or rights related to, for instance, an economic activity or land use. Such an ordinance-based enforcement mechanism could prove more cost effective and practical than prosecution through court activities. On the other hand, this also implies a need for qualified workforce endowed with powers to monitor compliance routinely. Yet administrative penalties are not a replacement for criminal prosecution. In fact, courts have a particular importance in the context of local administrative autonomy. Enforcing local administrative rulings would be effective and legitimate only when there is recourse to challenge these in case a conflict arises between citizens and local governments. Chapter 3 provides more details on the role of administrative courts in ensuring such administrative accountability of local governments. Moreover, as discussed earlier in the section on political decentralization, existence of a specialized administrative court on local affairs or a similar body would be very important to have a check and balance on local executive and legislative bodies. Local Government Discretion over Procurement Process and Administration of Services Increased mandates and responsibilities for new services require that local governments be endowed with discretion over a procurement process that covers procuring goods and services through contracts. New responsibilities assigned to local governments require increased flexibility in the procurement laws and regulations as well as high quality of employees who are well trained in public procurement, ethics, and contract management (Brennan and Miller n.d.). Such flexibility and discretion implies that, under national standards and guidance, local governments can develop procurement strategies, identify associated processes, and accordingly, issue contracts for goods and services. The discretions over the procurement process require clearly stipulated rules set by national laws and regulations mandating that all local governments adopt a procurement policy on an annual basis. This procurement policy defines how local governments manage all purchases and contracts. Such a procurement policy requires that local governments develop procurement strategies in line with the overall planning process, also considering workforce implications. Procurement processes are also closely associated with the quality of service administration and how contracts are awarded to partnerships in service delivery. These partnerships include procurement relationships with firms of various sizes, social enterprises, minority groups businesses, and voluntary and community organization suppliers. The procurement process also includes managing the procurement relationship after the contract is issued (i.e., contract management). 21 Local Government Discretion Over Civil Service and Employment Policies Decentralization reforms have significant human resource implications. The reforms reallocate powers and jobs, geographically as well as institutionally. Issues related to statute, prestige, and labor mobility often impede this relocation or dislocation across tiers of government. Bureaucracies tend to resist decentralization reforms on political (career perspectives, institutional allegiances, and political networks) and efficiency terms (economies of scale and scope, coordination, and principal­agent problems, and gender inequalities) (http://www1.worldbank.org/publicsector/decentralization/fiscal.htm). Such resistance may result in a situation where field officers maintain strong links with their original line ministries, thereby enjoying some insulation against local control. This is often the case in countries where central government officials simply transferred to local governments after decentralization reforms. The result is that staff burdens are transferred to local governments, without the discretionary powers over civil service. The local bureaucracy's allegiance in such situations is at best blurry between local governments and line ministries of central government, promoting upward rather than downward accountability. Due to similar bureaucratic resistance, in India, even though most state acts have clarified the functions to be devolved to local governments, the states have mostly not been able to transfer administrative and technical controls over locally administered programs. Some states such as Kerala have taken measures to transfer staff but have faced strong resistance from unions (World Bank 2004a). Similarly, in Indonesia, local governments have considerable control over expenditures for basic public services, but little control over the civil servants that provide those services. In such arrangements, these governments cannot downsize the workforce, alter remuneration packages, or introduce new recruitment and promotions to coincide with changing expenditure priorities (Campos and Hellman 2005: 242). Box 2.5: Mexico's health sector human resource management and decentralization reforms Mexico decentralized considerable powers in the health sector from the federal to the state governments. The aim was to increase health care accessibility and coverage. Some 116,000 health workers were transferred from federal to state employment as part of this. One of the most damaging results of decentralization was the fragmentation in labor policy. Most states now hire personnel through (at least) two different mechanisms, i.e., 'federal' and 'state' contracts. These contracts result in quite different labor benefits and working conditions for personnel, with the consequence that two health workers who hold the same type of post and perform similar tasks may have very different earnings. The fragmented labor policy is a very divisive factor for labor relations at both the federal and state levels. Finding a solution to this remains a key human resource challenge for Mexico. Source: Kolehmainen-Aitken 2004: 10, based on information provided by Dr. Armando Arredondo and Mr. Emanuel Orozco, Health Systems Research Center, National Institute of Public Health, Cuernavaca, Mexico, including the referenced documents they cite. Local government competency and discretion over civil service and employment policies, a critical component of human resource management in the context of administrative decentralization, should ideally cover the following items (Evans 2004): pay policy autonomy (setting local hardship and remote allowances, and setting overall wage rates); budget transparency (paying staff from one's own budget); budget and establishment control (controlling staff numbers and the authority to remove surplus staff); recruitment autonomy/authority (recognition as formal employer, authority to hire, and individual merit- based recruitment mechanisms); career management control (vertical and horizontal mobility, including transfers to deconcentrated units within the local government system,); and performance management (directing and supervising activities and tasks, conducting evaluations, and the ability to discipline and fire). Local governments gain greater administrative autonomy when they are able to pay their staff from their own budgets, have decisive control over staffing levels and skill mix, and can offer their staff with enough incentives, flexibility, and opportunity in career advancement and performance outcomes. Each of 22 these factors increases a local government's level of administrative autonomy and their accountability relationship with local bureaucracies. With adequate checks and balances, a decentralized framework for civil service management can help break dependencies upon higher levels of governments and promote downward accountability. Box 2.6: Administrative discretion over local staff and the tendering process in Uganda Who has the administrative decision-making power over the local staff and tender procedures? What kind of division of labor exists between elected and appointed staff at local levels? In Uganda, the local government civil service in each district headed by a chief administrative officer (CAO) is in juxtaposition with an elected political authority, the district council, led by its chairman (LC5), also elected by popular vote at the district level. In a number of districts this has produced instances of conflict over decision- making responsibilities and expenditure priorities between technical and political personnel (Livingstone and Charlton 2001). In Uganda, District Service Commission (DSC), which appoints all district staff, is a crucial decision-making arena and carries an enormous scope for patronage. The District Council (consisting of elected politicians) on the advice of its district executive appoints the members. Local community members and administrators allege that the patronage of the District Executive Committee over membership of the District Tender Board, with its privileges including generous sitting fees, enables them to influence the decisions of the board unduly. Tenders (e.g., for tax and construction contracts) are supposed to be allocated on the basis of a points system that takes account of a range of criteria including price, experience, and record of tax payments. In practice though, a letter of recommendation from a politician is believed to be a crucial prerequisite, and it is widely believed that successful tenderers are friends, relatives, or protégés of the political class, or proxy companies operating on their behalf. The DSC appoints and disciplines and has the power to remove all district staff. Civil servants feel that this makes them vulnerable to undue pressure or even victimization should they go against the wishes of local politicians. The DSC of one district, under pressure from the District Executive, dismissed a Deputy CAO after he questioned the use of the unconditional grant for politicians' allowances (Francis and James 2003). Source: Francis and James 2003. There is no universally accepted practice for the division of roles and responsibilities between central and local levels in determining staff pay policy in decentralization contexts. Staff pay policy usually falls under the control of partly local and partly central government. Salary budgets may be: (a) determined at the central level for all staff and transferred to decentralized units as a separate line item; (b) incorporated in a grant from the national government to the decentralized units; or (c) determined centrally and transferred to decentralized units as a salary budget for seconded civil servants, and determined and funded by decentralized units for local hires (Kolehmainen-Aitken 2004: 3). In many cases, local governments' autonomy in changing salary levels is restricted by centrally determined benchmarks, nationally unified salary scales and packages, and overall budget ceilings (ibid: 4). Local governments need to exercise budget and establishment control for real administrative autonomy in civil service management. For genuine budget and establishment control, they should be able to decide on staff numbers and mix as well as have the authority to discharge surplus labor, based on their needs and budget constraints. A centrally controlled civil service structure may not allow local managers to create new posts, or their budget may be insufficient to increase staffing levels, even if they have this power (ibid: 3). Of course, managing and controlling local staff numbers and mix would not be possible without recruitment autonomy. Such autonomy would imply recognition of local governments as a formal employer and granting them the authority to hire. It is equally important to set a merit-based system to recruit individuals (which could be monitored by the central government or an independent authority). In India, recruitment autonomy often remains at the national or state level; for example, in some states, 23 powers to suspend and dismiss panchayat (local government) staff are vested with the state bureaucracy. As a result, local government institutions are in a disadvantageous position vis-à-vis even middle-rung state government officials (Matthew 2003). Having hiring and firing power though is still not a guarantee of full control over local civil service. In Uganda, for example, local governments have been given the authority to hire and fire devolved personnel, and there has been a de-linkage of the local government and national civil service. However, in both cases the political influence of public sector health workers brought about central imposition of salary level, benefits, and employment conditions. This represented a major constraint on local decision space, not only in human resource management per se, but also in an indirect effect on the control of financial resources since human resources represent a high percentage of recurring costs and budget allocations (Bossert and Beauvais 2002). Local governments also need the flexibility to pursue appropriate career and performance management tools, as needed. Implicit in this is that local government human resource managers should have enough capacity and competency to administer such performance management schemes. At the minimum, commensurate with their capacities, local governments should at least be able to respond to identified needs and performance gaps with appropriate policies and programs regarding working hours and conditions, training, and career structures, as well as by transferring staff between geographic areas and facilities within their jurisdiction. 2.4 FISCAL DIMENSION The fiscal dimension of decentralization is mostly associated with allocative efficiency arguments, postulating that local governments can provide differentiated public goods to heterogeneous populations at the local level.12 This section attempts to establish the relationship between fiscal decentralization and accountability. As argued in chapter 1, this link is established primarily through the available fiscal discretionary space. Box 2.6 identifies the main questions to ask in delineating such space. These questions relate to four pillars of an intergovernmental system design: expenditure assignment, revenue assignment, intergovernmental transfers, and sub-national borrowing (Bird 2000). They focus mainly on the existence of revenue and expenditure autonomy, a transparent and rule based transfer system, and sound local government borrowing mechanisms. Box 2.7: Relationship between fiscal decentralization and local discretion · What's the degree of local government control over local revenues, expenditures, and investment decisions? · What's the degree of local government dependency on transfers? How does dependency on transfers affect accountability relationships? · Are there other informal transfers (i.e., cost reimbursements that are not recorded in the intergovernmental fiscal system) that create hidden dependencies on the central government and lead to moral hazard (and soft- budget constraints) on the part of local governments? · Are local governments allowed to borrow? If so, what are the rules and controls on local borrowing? Source: Authors. 12See Musgrave 1959; Oates 1972. 24 Fiscal decentralization is about sorting out roles and responsibilities among different levels of governments with the intent of transferring some fiscal decision-making powers around these four pillars from central to sub-national governments. In principle, this provides local governments with the necessary fiscal space to make its own decisions, finance them through locally mobilized revenue resources, and account for their decisions and actions to local populations. This change reshapes accountability incentives of actors in the intergovernmental system in favor of downward accountability. Absence of an appropriate fiscal space implies that intergovernmental fiscal arrangements are incomplete. In many countries, such incomplete intergovernmental fiscal frameworks load local governments with new mandates and responsibilities without the necessary resources to undertake them. For example, in Uganda, the district governments' taxing powers are not commensurate with the responsibilities being devolved. Specifically, while the local revenue sources and instruments available to local governments have remain unchanged from those operational in the pre-National Resistance Movement era, local expenditure responsibilities have expanded exponentially (Livingstone and Charlton 2001). While some local governments in the country could mobilize local resources to finance new responsibilities, others could not achieve the same results and became even more dependent on central government funding (Jeppsson 2001). Even where resources may be available, local governments are often not empowered to decide on how to use them. This perpetuates a fiscal dependency on the central government that reinforces excessive upward accountability of local governments to the central government, leaving little incentive for public officials to promote citizen participation. As a corollary, when local governments are assigned limited resources, local populations have little interest and incentive to be engaged with them as they know that the real decisions are made at the central government level. The following sections present a detailed discussion about the relationships between the four pillars of fiscal decentralization and fiscal discretionary space. Expenditure Assignment: Defining the Role of Local Governments in Service Delivery Fiscal decentralization requires that public services that have high local public good characteristics be assigned to local governments. Devolving expenditure responsibilities of local impact to local governments is an important step in increasing the participation of citizens in local decision-making processes. Thus, a genuine spirit of decentralization requires assigning a meaningful level of expenditure responsibilities to local governments to underscore their service autonomy to respond to local needs, a crucial aspect for establishing downward accountability linkages. Assignment of public services to sub-national governments--local or state--are based on considerations of economies of scale, economies of scope (appropriate bundling of public services to improve efficiency through information and coordination economies, and enhanced accountability through voter participation and cost recovery), cost-benefit spillovers, proximity to beneficiaries, consumer preferences, and flexibility in budgetary choices on composition of public spending. Box 2.7, below, identifies the criteria for assigning responsibilities between central and local governments to get the incentives right for efficient and equitable delivery of public services. An important consideration to be borne in mind is that accountability is often best promoted by establishing a clear and proximate linkage between the costs and benefits of services. If people who benefit from a service bear its financial costs, they are more likely to demand accountability (chapter 3 will discuss issues of accountability with respect to fiscal decentralization in more detail). A clear assignment of service responsibilities requires an unambiguous and well-defined institutional framework that defines the roles and responsibilities of different levels of government and provinces sufficient budgetary autonomy to carry out assigned responsibilities at each level of government. The clear assignment of expenditure responsibilities becomes even more important in certain sectors where line ministries and other government agencies may also deliver services at the local level--often times, in the same geographic area. This kind of situation often exists for education, health, and social services for many developing countries. Here, it is 25 important to distinguish between deconcentrated/delegated and devolved responsibilities. Responsibilities could be transferred to central units at the local level (deconcentrated) or to local governments only for the actual delivery of services (delegated), while all decisions on planning, budgeting, and financing remain at the central level. In South Africa, for example, health and education services are the concurrent responsibility of central and provincial governments. However, the concept of concurrence--who is responsible for what--is not defined clearly, which gives incentives for budget gaming. Since the South African Constitution specifies central government responsibilities as policy making, setting service delivery standards, and financing, provincial governments are responsible for the implementation of those decisions taken by the center. Therefore, provincial governments spend grants allocated to these services and let the central government worry about the funding gap (Ahmad, Devarajan, Khemani, and Shah 2005). In countries where decentralization does not end up transferring such responsibilities to local governments, central government departments and public sector companies continue to deliver most services that have high local characteristics. Examples include services such as primary education, health, and public security that are still delivered by central government (through deconcentrated units) in many countries. Such a discussion about deconcentrated/delegated versus devolved forms of service delivery responsibilities is important because in many countries, division of roles and responsibilities, therefore of discretion over services, between deconcentrated and decentralized units of government, are a source of tension and confusion in pursuing decentralization reforms. If the delivery of a service involves both deconcentrated and decentralized units, discretion is split between the two parties.13 If the relationship is hierarchical, wherein decentralized units are dependent on or responding to deconcentrated units (which are extensions of central government), there will be little coordination. As a result of such confusion over discretion over service responsibilities, (lateral) accountability relationships may get even blurrier and difficult to sort out as actors and their incentives would naturally diverge. Similar considerations may come into play in assigning responsibilities in seemingly "easier" sectors as well. Collection, transport, and treatment services in solid waste management provide a good example. Here too expenditure responsibilities might be assigned to multiple actors based on criteria such as economies of scale, commercial viability, and externalities. Provision of these services may again require a certain level of division of responsibilities among local governments, higher-level government units, and probably private sector and communities. Clear assignment of roles and responsibilities will be decisive in shaping accountability relationships among various actors at the local level. Another aspect of expenditure assignment is decision-making power over expenditure allocation. Beyond clarifying their expenditure responsibilities, local governments should also be given the discretion to make their own expenditure allocation decisions (with necessary reporting, monitoring, and sanctioning mechanisms). When it comes to giving power and authority to local governments in expenditure allocation, decentralization generally becomes a contested terrain between the central and sub-national levels. If the contest over the extent and contours of fiscal arrangements between levels of government is not resolved, it can easily constrain local autonomy and reduce the local government's credibility and responsiveness in service delivery. 13We described this form of accountability relationship as "lateral" accountability in chapter 1 under sections on actors of accountability as well as directions of accountability. 26 Box 2.8 Criteria for assigning responsibilities between central and local governments 1. Economic Efficiency: Functions should be assigned to jurisdictions: a. that are large enough to realize economies of scale and small enough not to incur diseconomies of scale [economies of scale]; b. that are willing to provide alternative service offerings to their citizens and specific services within a price range and level of effectiveness acceptable to local citizenry [public sector competition]; c. that adopt pricing policies (for partial or full cost recovery) for their functions whenever possible [public sector pricing]. 2. Fiscal Equity: Appropriate functions should be assigned to jurisdictions: a. that are large enough to encompass the cost and benefits of a function or that are willing to compensate other jurisdictions for the service costs imposed or for benefits received by them [economic externalities]; b. that have adequate fiscal capacity to finance their public service responsibilities and that are willing to implement measures that insure inter-personal inter-jurisdictional fiscal equity in the performance of a function [fiscal equalization]. 3. Political Accountability: Functions should be assigned to jurisdictions a. that are controllable by, accessible to, and accountable to their residents in the performance of their public service responsibilities [access and control]; b. that maximize the conditions and opportunities for active and productive citizen participation in the performance of a function [citizen participation]. 4. Administrative Effectiveness: Functions should be assigned to jurisdictions: a. that are responsible for a wide variety of functions and that can balance competing functional interests [general-purpose character]; b. that encompass a geographic area adequate for effective performance of a function [geographic adequacy]; c. that explicitly determine the goals of and means of discharging public service responsibilities and that periodically reassess program goals in light of performance standards [management capability]; d. that are willing to pursue intergovernmental policies for promoting inter-local functional cooperation and reducing inter-local functional conflict [intergovernmental flexibility]; and e. that have adequate legal authority to perform a function and rely on it in administering the function [legal adequacy]. Source: ACIR (Advisory Commission on Intergovernmental Relations). 1974. Governmental Functions and Processes: Local and Areawide. ACIR, A-45, Washington DC. Revenue Assignment: Financing Local Service Delivery Oates' decentralization theorem (1972) states that local governments should provide services to identifiable recipients up to the point at which the value placed on the last (marginal) amount of services for which recipients are willing to pay is equal to the benefits they receive. The implication of the theorem is that local governments must be given the authority to exercise "own-source" taxation to self- finance local services at the margin. The next section is based on the premise that local governments are more accountable to citizens when relying on their own tax bases. In contrast, local governments would be less accountable when the pleasure of expenditure benefits is separated from the pain of taxation (Bahl and Schroeder 1983). 27 A significant degree of revenue autonomy and some tax effort incentives are critical to encourage downward accountability and increase the efficiency of local government operations. While there is no set of prescribed rules in revenue assignment, according to Bahl, sub-national governments should have (at least) tax rate-setting authority over locally assigned revenues (Bahl 1999). Local governments must have discretion on three aspects of revenue generation: (i) setting the tax rate; (ii) defining the tax base; and (iii) administering the revenue collection (figure 2.1 below illustrates varying degrees of local autonomy that can be provided to local governments). A completely local tax is one that is assessed and collected by local governments--at rates decided and with proceeds accruing to local governments (Bird 2000; Bird and Vaillancourt 1998). Central governments may develop ways to intervene with local government revenue autonomy, putting direct or indirect restrictions on local government discretionary space. For example, in Uganda, the local government may adopt additional taxes, but only with the approval of the Ministry of Local Government. This essentially limits local governments to minor variations from the list, since the law provides no standard by which the Minister approves or disapproves proposed new revenue sources (Azfar et al. 2001). Similar restrictions on revenue generation exist in Philippines as well. While boosting local taxing authority, the Local Government Code in Philippines constrains local revenue collection through rules on rates, assessments, appeals, and administrative responsibilities (USAID [United States Agency for International Development] 1999). The code also regulates earmarked funds such as the Special Education Fund, specifying the property levies through which these funds are raised and the rules for distributing the proceeds. These requirements significantly compromise local revenue autonomy. In fact, Figure 2.1: Revenue Autonomy Scale NONE LOW MEDIUM HIGH HIGHER HIGHEST Central Revenue sharing Tax sharing SNG sets tax SNG sets SNG sets tax government sets with share whereby rate, but only tax rate rate and base rate and base of determined central/local within only "Sub-national unilaterally by revenue split can centrally government central authority be only changed permissible (SNG) revenue" with SNG consent ranges of SNG Source: World Bank (2006a) as adapted from Ebel and Taliercio (2005). Note: SGN = sub-national governments whenever fiscal (deficit) problems have become serious, the central government has conjured up schemes to reduce mandated, formula-based transfers, further limiting the flexibility and responsiveness of local governments (Azfar et al. 2000; Alonzo 2003). The accountability implication of constraining a local government's revenue raising ability is that it might create incentives for them to make inefficient investment decisions, especially if they can shift costs to central budgets (Campos and Hellman 2005: 242). In designing revenue system for local governments, it is important to pay attention to the revenue buoyancy of assigned sources. Low buoyancy means that a revenue system that is adequate today will be inadequate in the future, since revenue growth will not keep pace with expenditure needs. The buoyancy of revenue systems comes from a combination of growth in tax base, through improvements in tax 28 administration, and a rate increase. Low buoyancy of revenues will make local governments more dependent on centrally transferred revenues as well as more upwardly accountable. Financing the Fiscal Gap: Intergovernmental Transfer System Design Intergovernmental transfers are an essential component of local government finance in all countries. The intergovernmental transfer system design has important accountability implications, as it affects the fiscal dependency on central government for the services local governments are required to provide. As explained above, transfers may also affect local revenue raising ability. The rules governing the following four items of the transfer system determine local government's fiscal dependency on central governments, thereby affecting local discretion in expenditure and revenue decisions: Box 2.9 A simple model for analyzing intergovernmental transfers Alternative Forms of Intergovernmental Grants Ultimately, intergovernmental transfer systems rely on one of the several alternative forms as described in the table below, originally produced by Bahl and Linn (1992). The grant's form in this table is a product of a combination of properties based on the methods of allocating the divisible pool among eligible units and on the method of determining the divisible pool itself. The transfer system may use more than one particular form, even though the goal of one particular type of grant may actually conflict with the goal of the other. Grant type A in this table represents a grant form where the transfer pool is defined as a fixed share of the national tax revenue and it is distributed based on where the tax revenue was originally collected. This form tends to favor richer and more urban areas, as these areas are usually more likely to collect much higher tax revenues compared to poorer regions. In other words, the distribution is based on revenue collection performance rather than the needs of the communities. Grant type B is based on the same method in terms of determining the pool as grant type A, yet its distribution is based on a formula. A formula-based approach tends to be more predictable for the local governments as they could plan ahead relying on the estimated revenue from transfers. Yet, depending on the formula, the grant could end up favoring the local government units with certain qualities, and not others. For example, a formula that emphasizes the size of the city, and not the socioeconomic characteristics, might result in a transfer system favoring urban areas with larger populations--these usually tend to be economically more active areas--not necessarily poorer regions. Or in the opposite case, a formula that strongly favors poorer regions, but ignores the revenue collection efforts, may actually end up introducing disincentives for larger and richer municipalities to collect taxes. This is to say that a formula based system does not necessarily guarantee a smooth functioning transfer scheme. Grant type F is based on the same method in determining the grant distribution, yet the definition of the transfer pool depends on the ad hoc decisions of the central authorities. For obvious reasons, such a grant system may end up producing an unpredictable environment for local governments. When the grant distribution is ad hoc, whether the transfer pool is determined by a specified share of the national tax revenue (grant type D), or by the ad hoc decisions of the central government (grant type H), the end result is likely to be one where local governments enter into patronage relationships with the central government to gain favors in the transfer system. Grant types C, G, and K are all based on a distribution method using reimbursement principles, ideally aiming to provide grants based on the needs (or the services provided) of the local governments. The differences among these three forms stem from the difference in the method used in determining the transfer pool (fixed share of national income, ad hoc, or reimbursement based). A good transfer system is bound to use a combination of these different forms of grants as each of the grant types has its own merits and disadvantages. Considering the highly differential needs of local governments depending on their size, demographic profile, socioeconomic qualities, and so on, the central government would find it difficult to satisfy those needs by relying on only one type of grant in the transfer system. The ultimate blend of the transfer system would largely depend on the country context and the preferences of the central and local governments as well as their constituencies. 29 Alternative Forms of Intergovernmental Grant Programs Method of Determining the Total Divisible Pool Method of Allocating the Divisible Pool among Specified Share of Ad Hoc Decision Reimbursement of Eligible Units National or State Approved Expenditures Government Tax Origin of Collection of Tax Grant Type A NA NA Formula Grant Type B Grant Type F NA Total or Partial Reimbursement of Costs Grant Type C Grant Type G Grant Type K Ad Hoc Grant Type D Grant Type H NA Source: Text from World Bank (2006a), table from Bahl and Linn (1992). NA = not applicable. 1. The rules that determine the total amount of transfer--distributable pool. There are three ways to determine the total amount to be transferred: (i) as a fixed share of national government revenues; (ii) as a part of annual budget decision; (iii) as a proportion of approved specific local expenditures to be reimbursed. A rule-based transfer system will bring greater stability and predictability, and promote good planning and efficient service delivery effort. If the distributable pool is determined by the central government in an ad hoc and opaque manner, there would be uncertainty at the local level in receiving the transfer revenues, not to mention poor budgeting practices and the lack of an appropriate hard budget constraint. This would, in turn, weaken the accountability linkage between local governments and citizens. 2. The way the distributable pool is allocated among local governments. Once the amount in the distributable pool is determined, the next issue is how to distribute the pool among local governments. There are four general categories of distributing: (i) using the derivation principle (allocating revenues based on jurisdictions where they were collected), (ii) ad hoc, (iii) applying formula(s), and (iv) reimbursing costs. In general, distribution of funds by formula or derivation principle is a more effective system.14 Certainty over revenues improves local governments' capacity for better planning, which, in turn, facilitates greater engagement with citizens. Ad hoc allocation, on the other hand, may become a political decision on the part of central government, making fiscal planning impossible for local governments. Distribution rules may be structured to perpetuate existing power dynamics. For example, in Philippines, provinces and municipalities combined receive 57 percent of revenue transfer and shoulder 92.5 percent of cost of devolved function, highly urbanized cities and villages bear only 7.5 percent of the costs but receive 47 percent of the transfers (Emboltura 1994). This skewed distribution reflects the reality that governors are seen as bigger threat than mayors. There are fewer big city mayors than there are governors, and most legislators represent districts without highly urbanized cities in them. Thus by making provinces responsible for more services than they could pay for with automatic revenue transfers, Congress ensured that governors 14Even when transfer system is formula-driven, political concerns can play an important role in the selection of variables for the formula (Khemani 2007). 30 would remain dependant on legislators, who could subsequently offers services as brokers of additional revenue (Eaton 2001). 3. The way the purpose of a transfer system--unconditional general purpose grant versus conditional specific transfer--is identified. To the extent that transfers are conditional, local governments are dependent on the transferring authority. Depending on their purpose, conditional transfers can have specific spending objectives and give local governments' limited autonomy to meet local preferences. Further, if local governments do not have any own-source revenues and all of the central government transfers are conditional, one can expect them to be under significant restrictions to their discretion over expenditures and revenues. Since this situation makes local governments more responsive to central government demands, use of these funds may then reflect the priorities of central government rather than local citizens. For example, in Uganda, conditional grant terms contain not only affirmative duties but also list of things that grants cannot be used for. The conditional grant reporting requirements are said to be onerous (Azfar et al. 2001). The center's tight control over the use of intergovernmental transfers may have impeded, in some ways, the development of local accountability as officials and elected representatives have focused their attention on meeting grant conditions (Devas and Grant 2003). While a high proportion of central transfers is not as such incompatible with local autonomy, the conditions frequently attached to these transfers can undermine genuine local decision-making (Francis and James 2003). The question now being debated in Uganda is whether this was an inappropriate recentralization or whether it was a legitimate fine tuning, putting in place mechanisms which should have been there at the outset (Jeppsson 2001). 4. Local governments' involvement in the design and management of the intergovernmental transfer system. Intergovernmental system design is a dynamic process that should be responsive to changes in citizens' needs and preferences with constant adjustments and systematic reviews for "sorting out" rules and responsibilities among different types and levels of governments. By making local governments a part of that process, central governments can empower local governments by increasing their role in a significant portion of their revenue resources. A high degree of dependence on transfer revenues shifts the focus of local government accountability away from citizens to the central government (Rodden 2002). In other words, this shift creates an imbalance between downward and upward accountability incentives at the local level, negatively affecting local government discretion and their incentive to respond to citizen demands. In Nigeria, for example, local governments' overdependence on intergovernmental transfer revenues coupled with uncertainty about the amount and timing "facilitates local evasion of responsibility under the guise of fiscal powerlessness" (Khemani 2004: 22). In another case, in Uganda, transfers, as earmarked conditional grants, finance approximately 88 percent of the local government expenditures. As local governments still finance their expenditures mainly from transfers from the central government budget and as own source revenues are significantly constrained, a substantial degree of line ministry control over local expenditure decisions is maintained, with the disadvantage of curtailing district autonomy against the original intentions regarding the decentralization of decision making. The financing that comes through a transfer system leaves little room for local decision making and prioritization, and they are often not sufficient to meet all expenditure needs. As such, with lack of sufficient local discretion over expenditures and with the shortage of funds to cover recurrent expenditure, the end result is a failure to deliver services. This kind of situation is revealed in the uniformly run-down, even deteriorating condition of urban infrastructure and, in rural areas, in poorly maintained access roads. This, and the feeling that people's taxes are being used largely to cover the salaries of the local administration, have in turn generated a widespread unwillingness to pay taxes, exacerbating the problem of poor service delivery and 31 generating a vicious circle of noncompliance. This feeling has also been a reason for the major outbreaks of rioting, which have occurred when graduated tax,15 in particular, has been suddenly raised (Livingstone and Charlton 2001). Financing Infrastructure: Local Government Borrowing The number of developing countries with efficient markets for local borrowing is limited.16 Since local government borrowing has major fiscal implications, many central governments limit, control, or even totally prohibit the issuance of debt by local governments. In countries where the transfer system falls short of responding to local investment needs and local governments have limited ability to raise additional revenue through their own sources, strong restrictions on borrowing may actually limit local discretion in addressing investment needs. In countries where the central government does not prohibit local borrowing entirely, there are measures to restrict local borrowing. The limitations imposed on local government borrowing may have an impact on local discretion over key development projects. There are four different approaches to limit local government borrowing: 1. Market discipline. In countries where there are vibrant financial markets, market forces regulate local government borrowing. Local governments are monitored by market actors and forced to follow a disciplined approach to borrowing. 2. Cooperative arrangements between local and central governments. In several European countries and Australia, a negotiation process between central/federal and local governments designs local borrowing controls. Local governments are actively involved in establishing these control rules and voluntarily comply with them. 3. Rule-based controls. Central governments can also unilaterally exert controls on local borrowing by imposing fiscal rule. Rules may take the form of restrictions on overall budget deficits (Austria, Spain), operating budget deficits (Norway), indicators of debt-servicing capacity (Spain, Japan, Brazil, Republic of Korea), level of accumulated local debt (Hungary), or level of spending (Belgium, Germany). If the rules for local borrowing are not transparent and subject to negotiation, local governments will loose their domain of discretion about financing infrastructure, making it more difficult to trace investment decisions made at the local level and how they are financed. 4. Administrative constraints. In several countries, central governments are empowered with direct control over local borrowing. This control may take various forms, including the setting of annual limits on the overall debt of individual local governments (Lithuania); special treatment or prohibition of direct external borrowing (Indonesia, Mexico); review and authorization of individual borrowing operations, including approval of the terms and conditions (India and Bolivia); or the centralization of all government borrowing with on-lending to local governments (Latvia and Indonesia). In these 15Recently, Uganda abolished its graduated tax, which was approximately 80 percent of the total local tax revenue in Uganda (Steffensen et al. 2004). The government of Uganda did not substitute this tax with any other source of revenue (Francis and James 2003). Although the tax was a controversial one and caused a lot of discontent among local populations (in the form of nonpayment and riots), abolishment of this tax without a replacement made local governments increasingly more dependant on central transfers for their revenues. This is still an issue of contention in Uganda. 16In inefficient markets, moral hazard--the presumption by capital markets that central government will bail out local governments in case of bankruptcy--is a major concern for central governments. 32 countries, borrowing rules make local governments more submissive to the central government, with limited autonomy over key decisions on investments. Box 2.10: Municipal borrowing in Brazil In contrast to other Latin American countries, municipal governments in Brazil may borrow from commercial banks on a short-term basis. However, as a response to widespread sub-national debt, related to high inflation rates and a lack of fiscal discipline, especially among states and large municipalities in the 1990s, measures were taken to ensure fiscal control. The Fiscal Responsibility Law introduced new requirements for state and municipal authorities to publish revenue and spending accounts, restricting deficits at the sub-national level and making public officials legally liable for any infractions. It also prohibits the bailing-out of sub-national debt by the federal government. To be able to borrow either from the internal or the external market, local and state governments now need: (i) the approval of the legislature, (ii) to submit a request to the Central Bank for analysis, which then issues technical report that is sent to the Senate, recommending approval or the rejection of the request, and, (iii) the Senate's approval. Source: Môra and Varsano 2001. 2.5 CONCLUSION This chapter explained the basic elements of decentralization that provide local governments with an appropriate discretionary space for local decision-making and service delivery. It focused on the ideal case where local governments are endowed with significant autonomy about the political, administrative, and fiscal dimensions of decentralization. Yet when it comes to the implementation of reforms, decentralization often leads to contentions across different levels of government as well as among various local actors that are affected by the outcomes. To what extent local governments should be given the discretion and autonomy over local decision making? When different country contexts and conflicting interests come into play, the answer is not as simple as the theoretical framework suggests. As the reforms proceed, the questions about who will have what kind of power over what kind of functions remain a controversial issue. All such tensions are actually not surprising as decentralization in its essence changes power dynamics within a political process, where different actors may have opposing interests in the reforms. More specifically, political decentralization has implications for the electoral and party systems. As a result of new rules of the political game, new local leaders emerge, introducing new actors in the political competition. Accordingly, the nature of interaction between local leaders and constituencies are redefined. Decentralization reforms also change the way local institutions work and bureaucracies behave. It requires different set of skills and flexibilities in job descriptions. It empowers some actors within bureaucracies and undermines others. It changes the locations of jobs and their associated benefits. As a result of new regulatory powers and roles over procurement and service delivery processes, decisions and actions of local governments have greater impact than before on local economies. As such, decentralization has fiscal implications and challenges some of the vested interests in the older version of the decision-making process. It leads to new kinds of interactions and contractual relationships between local governments, small or big private sector, providers and producers of services, communities, and NGOs. 33 With reshaped power relationships and increased stakes at the local level, decentralization reforms often face strong resistance during their implementation. Coupled with the inexperience and lack of capacity on the part of local governments, decentralization is often blamed for not fulfilling its promises and in fact having negative impact on the development processes. As a result, reforms are frequently revisited and sometimes reversed. Many central governments around the world respond to such frictions in the form of stricter control and excessive accountability measures over local governments, sometimes revoking some crucial aspects of discretionary powers and resources extended to local governments. Yet, as mentioned in the introductory chapter, reducing powers and functions is not a form of accountability; it just makes local governments irrelevant. The negative impacts of decentralization do occur, but the reason behind failures is the absence of effective accountability systems--both upward and downward- not that local governments have excessive powers. If designed carefully, discretion and accountability support each other. The important issue here is to emphasize the fact that upward accountability mechanisms introduced by central governments are necessary but not sufficient to ensure appropriate use of local discretion. In the absence of sound mechanisms for downward accountability --accountability of local governments towards citizens--the sole emphasis on upward accountability measures ends up limiting the autonomy of local governments in local decision-making and service delivery--negating the intended goal of empowering local governments. The next chapter focuses on ways in which downward accountability of local governments is ensured through both supply- and demand-side accountability approaches. 34 Chapter 3: Decentralization and Local Accountability Devolving discretion or autonomy to local governments without putting in place strong, performance- oriented accountability mechanisms can be a recipe for disaster. As we argued in chapter 2, a decentralization strategy's underlying premise is to give local governments a meaningful degree of discretionary powers. Such discretionary space should be accompanied by the safeguards against its abuse. In the absence of such safeguards, discretionary space alone may actually leave the door open for misuse and abuse of new powers. Public officials could be unduly influenced and captured by elite groups due to a lack of checks and balances to discretionary power. In addition to safeguarding from abuse, accountability mechanisms should create incentives for performance. The traditional approach to accountability of local governments has relied on supply-side or public sector accountability instruments. Increasingly, both practitioners and academics are recognizing the critical role of the demand side or civil society in contributing to accountability. So far, however, these two approaches to accountability have developed in parallel tracks. There have been few attempts to explore the complementarities, synergies, and tensions. This chapter brings these two approaches together around the three critical dimensions of decentralization: political, administrative, and fiscal/financial. The chapter begins with an overview on the actors, objects, and approaches to accountability and then looks at political, administrative, and fiscal/financial accountability from the combined perspective of supply- and demand-side approaches. 3.1. THE MULTIPLE DIMENSIONS OF ACCOUNTABILITY 3.1.1. Actors and Directions of Accountability As mentioned in chapter 1, accountability implies a relationship between a principal (or forum or account holder) and an agent (or accountee). Accountability in local context involves such relationships among central governments (both as ministries managing decentralization strategies and deconcentrated units of central governments delivering services locally), executive and legislative arms of local governments (bureaucrats and councilors), local citizens, and firms. In this report, we are concerned with four sets of relationships as they apply to devolutionary policies: · Vertical accountability is the accountability of individuals and organizations within the state apparatus to citizen­voters through electoral and explicitly political channels; · Horizontal accountability refers to the control that state institutions exert on the performance of other public agencies and government branches within the same level of government (e.g., legislature over the executive); · Upward accountability refers to the accountability of a lower level of government entity towards its overseeing government agencies; and, · Downward accountability refers to the local government's ability to be accountable towards its citizens. 3.1.2 Objects of Accountability The relations of accountability differ not only according to the actors that inhabit them, but also according to the object of accountability--i.e., what is the agent accountable for? The term accountability is 35 originally derived from financial accounting, but its meaning has expanded to include administrative and political accountability:17 · Political accountability refers to accounting to the electorate about the way government business is conducted. This form of accountability usually refers to political leaders being held accountable through local elections, but it also refers other political aspects such as campaign finances, legislative performance, and so on. · Administrative accountability covers administrative decisions (i.e., decisions about recruitment, procurement, land use and planning, and so on) of local officials. Administrative accountability can be an upward, hierarchical relationship between the local government and higher levels of government, such as laws giving power to a central government ministry (e.g., the Ministry of Local Government) to audit local governments' administrative transactions. It can also involve horizontal relations such as when a local government's internal control inspectors audit local government units. · Fiscal/financial accountability refers to the local financial planning and decision-making process and its outcomes. This includes accountability for sound and transparent public expenditures and financial management systems, overall fiscal discipline, allocation of resources to priority needs, and efficient and effective allocation of public services according to an effective, efficient, transparent, and rule-based public financial management (PFM) system. The object of accountability can also be categorized in terms of whether accountability is rendered about the process or outcome of a decision or action: · Accountability for process refers to "whether or not governments are acting in accord with rules that prescribe how they should or should not behave. Accountability for outcomes, on the other hand, focuses whether the decisions of governments are in the best interest of their citizens" (Schroeder 2006: 2). · Accountability for outcomes emerged as part of "the new public management" paradigm in response to the concerns about government performance in terms of providing high-quality services that citizens value. It is an attempt to "shift accountability from complying with rules and regulations to achieving results" (Behn 2001: 27). In this line, public officials are expected to respond to individual citizens through innovative and flexible approaches. The discussion about process versus outcomes is still relevant in each dimension. For example, in the context of administrative accountability, citizens may be highly satisfied about the quality of a particular service they receive (outcome), but whether or not the contract to the service provider was awarded in a fair and competitive process is still an important process question that needs to be addressed through accountability mechanisms. A proper accountability system should seek to reconcile the tension between accountability for performance and accountability for due process. Ideally, accountability is expected to be about process and performance. Accountability systems focusing too much on "process" will more likely be inflexible so much so that everything would be prescribed as to how things should be done. On the other hand, focusing too much on outcome and results and ignoring how things are done will more likely leave room 17Historically, the concept of public accountability is derived from accounting and bookkeeping (Bovens 2005). 36 for corruption or unfairness in the process of getting the results.18 The challenge here is whether we can encourage public managers to be "entrepreneurial," while we make sure that we can get enough information about the conduct. 3.1.3. Approaches to Accountability There are two basic approaches to pursue accountability, regardless of whether it is about political, administrative, or financial/fiscal decisions, or whether it is about process or outcomes: · Public or supply-side accountability is the obligation of public authorities (governments, elected representatives, corporate, and other governing bodies) to explain publicly, fully and fairly, how they are carrying out responsibilities that affect the citizens in important ways. Public accountability refers to institutionalized practice of account giving--it focuses on public sector managers who spend public money, exercise public authority, and manage a corporate body under public law. Public account giving provides political representatives and voters with the necessary inputs for judging the fairness, effectiveness, and efficiency of their governance system (Manin, Przeworski and Stokes, 1999). · Social or demand-side accountability is an approach toward building accountability that relies on civic engagement--in which ordinary citizens and/or civil society organizations demand accountability. Social accountability puts a strong emphasis on the proactive role of civil society actors. The diffusion of social accountability initiatives can be seen as part of the emergence of a broader "new accountability agenda" (Goetz and Jenkins 2005). This new agenda responds to the systemic failures of conventional institutions of democratic accountability to overcome entrenched processes of elite capture and elite bias, and delivers accountable governance to the poor and marginalized groups in society. Citizens' accountability demands can contribute to: (i) improved governance (by monitoring that authority is exercised in a transparent, responsive, and fair way); (ii) improved service delivery (by providing feedback to service providers on areas that need improvement and helping to discipline them); and (iii) deepening democracy (by allowing citizens to exercise their right to expect and ensure that government acts in the best interests of people). The role of citizen oversight is not to replace, but to complement and enhance public accountability mechanisms.19 The next section explains how these approaches can complement each other for this purpose. 18Behn (2001:29), in an American context, argues that "in an environment with excessive oversight and regulatory requirements," most public managers choose to satisfy the test for accountability for finances and fairness (in search for being audit-proof for process). According to him, most public managers make sure that no one can hold them accountable (that is, punish them) for not handling the finances properly; then, if they still have time, resources, organizational capacity, or flexibility left over, they will try to improve performance. Other risk-taking and result- seeking executives, on the other hand, may seek "innovative ways to circumvent the formal rules." 19The concepts of "police patrol" and "fire alarm" oversight mechanisms illustrate well this complementary role. Police patrol oversight is the traditional modality in which supervisory agencies operate, trying to keep a constant eye on those they are supposed to monitor. In contrast, fire alarm oversight happens when an agency relies on external actors to detect when there are problems (to "sound the alarm") and then focuses its attention particularly carefully on those areas (McCubbins and Schwartz 1984). A similar emphasis on the complementary role of citizen oversight appears throughout the literature. See Paul (1992) on the relationship between "exit"/ "voice" and control. 37 3.2. POLITICAL ACCOUNTABILITY IN DECENTRALIZATION In chapter 2 we discussed the need to provide an appropriate political setting for local governments. This section describes approaches to make the local government accountable in the political dimension. Strengthening the political dimension of accountability at the local level requires strengthening public accountability by introducing some safeguards in the local electoral system and local council oversight, as well as social accountability through mechanisms that promote citizen oversight. 3.2.1. Public Accountability Approaches Introducing Safeguards in the Electoral System An important issue in strengthening the political dimension of accountability is the local electoral system. Chapter 2 discussed the broader features of electoral system that set the rules of game in elections--one of the most crucial aspects of political accountability on the supply side. The structure of local electoral system has an impact on "the success or failure of decentralization reforms" (Hiskey 2006: 12). If the local electoral system does not secure a real competition among local politicians, decentralization reforms might end up strengthening the hands of local political strongmen. Political competition among local politicians increases the chances for vulnerable groups to be included into the decision-making processes. More importantly, . . . [g]reater and more meaningful political competition for local political offices, then, should have a positive impact on reforms that expand the decision-making powers of local agents. Where citizens have a means of control over elected officials through the presence of viable electoral alternatives, the increased level of uncertainty among local political elites should enhance the responsiveness of these elites to the concerns of the citizenry (Hiskey 2006: 12). To avoid some of the negative implications of use of the electoral system, some countries introduced supply-side measures such as allowing independent candidates to run in local elections, reserving seats for women and/or other vulnerable groups, allowing for recall of elected officials from public office, and limiting the length of term for these officials. The list of these measures is not intended to be exhaustive, but to illustrate how some supply-side measures that improve the electoral system work in local context. Allowing Independent Candidates and Local Political Organizations to Run in Local Elections Electoral systems must also be analyzed for allowing participation of independent candidates or the participation of local political organizations. For example, in countries such as India and Uganda where independent candidates are admitted in local elections, authorities are more likely to be accountable to people (Ribot 2004: 27). Reserving Seats for Women and/or other Vulnerable Groups Elections fail to become effective accountability mechanisms when local elites dominate elected bodies, and the poor and minorities are not represented (Schroeder 2006: 11). To overcome this potential disadvantage in local elections, some countries have set aside some proportion of elected seats to minorities (Schroeder 2006: 11). Many countries reserve seats for women and/or other vulnerable minorities. These can be a fixed number or percentage of open seats, as in India, where one-third of the seats are set apart in many states for women. The impact of reserved seats is debatable (See box 3.1, below, for examples on India and Uganda). On the other hand, in countries where such a quota system does not exist, the outcome may be even worse. Unlike India and Uganda, for example, the Philippines Local Government Code 1991 does not have quotas for women and minorities. As a result, women hold only about 10 percent of elected local offices (Blair 2000). 38 Box 3.1. Impact of reserved seats in India and Uganda Despite affirmative action in India in favor of women, however, men still tend to dominate decision-making processes. In Karnataka state, for example, women hold one-third of the seats in local councils. Yet they play a marginal part in council affairs, tending to either remain silent or participate only as their husbands direct. Similarly, the scheduled castes in Karnataka's local councils either do not participate or do so at the direction of their elite patrons (Blair 2000). In Uganda, the reservation of seats for women, youth, and disabled does not seem to have had much of an impact on outcomes, partly because of the limited skills and effectiveness of those occupying such seats (Rakodi 2002). Female councilors acknowledge that, while men have become more aware of women and children issues as a result of statutory requirements for women to be represented on all local councils, this does not mean that their decisions are more gender sensitive. In addition, female councilors are found to be more hesitant and less vocal higher up the local political system; whereas discussions with neighbors at the village or ward level (LC1) are more effective, at the district level (LC5) the issues are more complex, and female councilors are intimidated about speaking out (Saito 2000). Allowing Recall of Elected Officials. Many local government systems contain formal (i.e., a vote of the town council; a minister's discretion) or informal means for pressuring or legally removing local officials from office, absent of any public input, for purely political reasons. There needs to be a legitimate process for removing corrupt and incompetent officials (public referenda, formal financial reviews, and so on) for good governance. At the same time, because local and other pertinent institutions are weak and the level of politicization is high, such mechanisms for ensuring good governance can be used as fronts for the political manipulation of local administrations by party leaders or nonlocal interests (Bland 2006). Limiting the Length of the Term of Office. The length of the term of office can be too short or too long to benefit democratic growth. A primary consideration is the time elected local officials need to develop a coherent team and an action plan and implement it, without continually having to worry about electoral politics. If a new mayor will need a year or so to get the administration going reasonably well, and at least the final 6 months will be largely driven, if not consumed, by the upcoming election, then a 2- or 3- year term is too short. On the other hand, a 5- or more year term, especially if unlimited re-election is permitted, may be an invitation for local bossism. Too much time between elections could reduce the local officials' sense of accountability to the electorate or become a temptation to abuse the power of incumbency to remain in office indefinitely (Bland 2006). India follows the municipal corporation system in most states, where the councilors usually chose the mayor in the municipal corporation from among themselves through an indirect election for a one-year renewable term. The mayor generally lacks executive authority because of his short one-year tenure compared to councilors elected for five-year terms (www.citymayors.com). Additional Criteria. Sufficiently separate elections of national and local political leaders allow local concerns to predominate during the local electoral cycle; they raise the profile of local government, allow local leadership to function more independently of higher authorities, and further strengthen the tacit contract between the local elected official and the voter. Factors such as campaign financing rules may favor large, centralized parties and make it more difficult for local candidates or new parties to compete. Media access can give candidates an unfair advantage. Improving Local Council Oversight A second public accountability component to improve local government downward accountability pertains to local council oversight over local appointed officials. Local council oversight in decentralizing countries is an important component of institutional checks and balances and is a critical entry point to achieve good local governance and effective public services. 39 As argued before, representative local governments are at the center of the accountability framework that this ESW is presenting. Local councils are the core units of such representative governments. Functioning local council oversight relies on the assumption that local elected representatives have more incentive to respond to the needs and preferences of local populations and that they are more downwardly accountable as compared to local bureaucrats. In the ideal setting, local councils make policies and laws--as representative bodies for collective decision making, and they work with the local government's executive branch to deliberate on policies and implementation through public financial management as well as service delivery. Local councils are assumed to be representing citizens--by giving voice to individual citizens, civil society organizations, and business groups, and by representing the needs of local constituents in policy making. This assumption leads to public accountability measures that aim to strengthen local council oversight so that elected local leaders oversee local government operations on behalf of local citizens. The local council oversees executive policy implementation and service delivery and holds the local bureaucracy accountable for their performances. The accountability and oversight relationship on service delivery between the local executive branch and local councils requires a constant information flow between the two on how local service producers are performing. Local service producers, in other words, the local bureaucracy, carry out the production of local services according to policies that the local elected representatives decide on.20 Local service producers are answerable primarily to the local government's executive branch (see the section on administrative accountability for greater details on this accountability relationship). A well functioning local council oversight assumes that they are also accountable to local elected representatives. The relationship between elected local councilors and executives also pertains to budget planning, execution, and monitoring and evaluation. Local councilors are supposed to oversee local government's executive branch during the entire public financial management process and provide local executives with constant feedback. This requires establishing the linkage between planning and budgeting (whether budgets reflect planning) as well as between planned and executed budgets (budget execution performance), and producing policy-oriented budgets (outcome-oriented budgeting in terms of responsiveness to demands and preferences of local citizens). Such arguments in favor of local council oversight rely on the assumption that citizens are in control of local decision-making through their local legislative council representatives. In the real world, local council oversight is one of the weakest links in accountability and oversight mechanisms. In Uganda, for example, it is unlikely that local councils exercise effective oversight of budgets and expenditures by their governments and subordinate local governments, since few council members appear to understand the budgetary and expenditure figures they are provided (Azfar et al. 2001). Therefore, building capacity of local councilors to better exercise their oversight role remains a challenge. Moreover, central governments may also resort to institutional arrangements that diminish the power and authority of local councils. For example, in Ghana, only 70 percent of the local council is elected; the central government chooses the remaining 30 percent and appoints the chief executive (mayor) (Schroeder 2006: 10). Deficiencies in local council oversight often stem from two major weaknesses in local government systems in decentralizing countries: 1. A lack of strong oversight mechanisms (de facto and de jure) in formal and informal practices of oversight functions by elected local leaders: This is mainly due to weak definitions of oversight mechanisms in laws and regulations or ineffective utilization of formally defined oversight 20In the case of private production of local public services, they oversee the private contractors. 40 mechanisms. In an ideal setting, availability of an appropriately defined and used oversight system would affect the behavior of local governments in favor of transparency, accountability, and responsiveness to citizen demands and preferences. 2. Weak capacity: Lack of capacity is often cited as one of main reasons why local councilors cannot function as effective oversight agents at the local level. Bearing in mind such observed weaknesses in local council oversight in many decentralizing countries; the following section highlights the role of demand-side initiatives to improve local governments' political accountability. These initiatives center on the more direct role that local citizens can play in improving downward accountability of local governments. 3.2.2. Social Accountability Approaches As mentioned earlier, public accountability instruments in electoral system and local council oversight are insufficient to ensure downward accountability of local governments. Deficiencies as such reveal accountability failures both between the elected official and the electorate, and between the elected bodies and the bureaucracy. They manifest in patronage, clientelism, elite capture, or elite bias. Social accountability mechanisms can act as a source of countervailing power, giving poor people a direct voice in the policies that local governments formulate and implement. These mechanisms are often part of broader efforts to deepen democracy and to ensure the existence of a robust public sphere that is needed for citizens to give feedback and control government action.21 Two types of interventions through which citizens have expanded powers to hold political leaders accountable are: (i) generic legislation that empowers citizens to demand explanations and justifications from local governments; and, (ii) the institutionalization of formal bodies or processes that act partly or exclusively as citizen oversight mechanisms. Generic Legislation that Empowers Citizens to Demand Local Accountability. Some initiatives have focused on introducing legal mechanisms that empower citizens for grievance redress or with the right to request explanations regarding municipal legislation. For example: · Public hearings and consultations are probably the most common instrument of dialogue between citizens and the local government. In most cases the hearings are consultative and nonbinding. They make government answerable to their constituency, but they lack the enforcement dimension of an accountability relationship. · The right to demand a public hearing. As part of the process of adopting normative acts, the municipality must hold a public hearing on the proposed act if it is requested by at least a minimum number of persons or an association having a minimum membership. · Public petitions. Any person or organization may petition the municipality to adopt, amend, or repeal a normative act, and the petition must be reviewed and responded to in writing. · Administrative complaints. The municipality must go beyond the minimal provisions of the country's administrative appeals law by giving complainants an opportunity to be heard and 21These efforts represent a new approach to democratic governance that some scholars have called "Empowered Participatory Governance." It looks at how alternative political and administrative designs can "surpass conventional democratic institutional forms on the quite practical aims of enhancing the responsiveness and effectiveness of the state while at the same time making it more fair, participatory, deliberative, and accountable" (Fung and Wright 2001: 8; Fung 2004). 41 shifting the burden of justification to the government to prove that they followed rules and processes, as opposed to the complainant having to show that the government failed to do so. · The right to initiate a recall or referendum. The Local Government Code in the Philippines establishes the mechanism of recall as an immediate accountability mechanism for elected local officials. Their tenure may be terminated by popular vote under a special recall election that can be initiated by a petition. The code also guarantees citizens the right to pass key legislation directly or oppose proposed legislation with the instrument of referendum. Specific Bodies and Processes for Citizen Oversight. Citizen oversight bodies are institutional structures that citizens form to provide a direct channel for citizen oversight over local government's work. All the citizens in the municipality (Gram Sabha in India), several citizen representatives (Vigilance Committee in Bolivia), or an elected member (citizen ombudsman in Japan) can form these oversight bodies (see box 3.2. for details on these different forms of citizen oversight bodies). Box 3.2. Citizen Oversight Bodies in Bolivia and Japan In Bolivia, the Law of Popular Participation created local vigilance committees to monitor the activities of elected local government bodies and participate in local planning and budget creation. Vigilance committee members are selected from various traditional governance systems including peasant associations and indigenous communal institutions. The participation of non-elites is encouraged. Vigilance committees that suspect wrongdoing by local councils can invoke a legal complaint procedure in which a special Senate committee reviews the case and has the power to suspend funds to the local council if it is found to have acted inappropriately. In Japan, in response to the widespread perception of corruption in local governments, a civic movement began establishing citizen ombudsman in several municipalities. This successful initiative has spread throughout the country and led to the formation of a National Citizen Ombudsmen Liaison Council and to the recognition of this mechanism in government statutes. In addition, the National Council developed a survey to rank the level of transparency of local governments, which was used as an additional source of pressure over local government to improve its performance. Source: Serrano 2006. The main political decision that elected leaders make concerns the allocation of public resources to respond to local population preferences. Consequently, municipal planning and budgeting is the municipal function that registers the highest level of civic engagement and oversight. In many countries, this has led to the creation of specific bodies through which citizens can demand accountability regarding municipal planning and budgeting. In the Philippines, for instance, the Local Government Code mandates that all provincial, municipal, and barangay (village/district) governments establish a local development council to set the direction of economic and social development and review local governments' budgets. One-quarter of the council members should come from NGOs and CBOs (Estrella and Iszatt 2006). Similarly, the expansion of participatory budgeting practices across the world has provided an important opportunity for citizen oversight of the allocation of municipal resources.22 Participatory budgeting is one way to address the problem of elite capture of the local parliament. Although the ultimate decision power over the budgets rests with the parliament, the process can give marginalized 22After its start in Porto Alegre, Brazil in 1988, participatory budgeting has been taken up on voluntary basis by over 140 municipalities in the country. In other countries such as Bolivia, Peru, or the state of Kerala in India, participatory budgeting is mandatory and regulated by a national legislation. 42 groups a direct voice in the decisive preparatory phase of the budget preparation. The municipality of Porto Alegre in Brazil, for instance, employs community organizers on a regular basis to encourage unorganized citizens to participate and to prevent the process to be taken over by clientelistic networks. Participatory budgeting forces municipalities to make budget information available to ordinary citizen and to report on a regular basis on the status of execution of the previous year's commitments. As citizens gain ownership of the process, they are motivated to oversee the implementation of their approved projects. The deeper involvement of citizens in the planning and budgeting process creates favorable conditions for them to demand accountability. Creating a Political Culture for Citizen Oversight through CDD operations. Programs that follow a CDD approach23 have been an important vehicle for empowering citizen oversight. By introducing mechanisms for poor people to participate in decision making and mechanisms for transparency and accountability at the local level, CDD programs have helped to develop a culture of citizen oversight. A government program that empowers citizens to be informed and have a say over local public expenditure creates the expectation that all programs should follow the same standards (see box 3.3). 23A CDD approach tries to improve the well-being of poor people by increasing their control over the way investment resources are planned, executed, and managed. Since communities do not act in isolation but in a local space where they interact with local governments, sectoral service delivery units, the private sector, and other civil society organizations, the CDD approach has lately been expanded into a broader local governance approach that captures the quality of these local interactions (Helling, Serrano, and Warren 2005; McLean et al. 2006). 43 Box 3.3. Building a culture of citizen oversight: a vignette from Indonesia's Kecamatan Development Project It was a brilliantly clear morning in central Sulawesi when the villagers first spied the large pile of lumber. One of the delivery truck drivers stood lazily by the wood, blowing cigarette smoke over his steaming coffee. He had come from Palu, the provincial capital. The gold lettering embroidered on his hat told the villagers that he and the silent man in the neatly pressed green safari suit, also sipping his coffee, worked for the Palu Public Works Department. The villagers were curious. Just last year they had gotten funds from the Kecamatan Development Project (KDP) to build a stone road from their rice fields to the market route, and now here were the materials to repair a bridge. Had the government finally noticed their plight? "Friend, what is this wood for?" "It's to build a bridge." "How much wood is there? What did it cost?" "That's none of your business. Just be thankful that the government will be building you a bridge." "But we want to know. This is our new rule here. You have to come to the balai desa and tell us about the project. Then you have to post a signboard so that all of us know how much this bridge costs. If KDP does it, we want you to do it too." "You are mistaken. KDP is KDP, and it has KDP rules. This is a government project, and we follow our rules. Just be thankful that you are getting a bridge." The villagers were troubled. That night the village elders met. Some people said they should just accept the wood because the village needed the bridge. But many more villagers were angry. This was now the era of reformasi, and people had a right to know about projects in their village. Early the next morning, even before the first rays of sunlight pierced the dark clouds, the villagers had hauled the wood back onto a large truck owned by the son of the village council head. Two truckloads of villagers and scores of motorcycles joined the procession to the district parliament. When the first parliamentarians arrived for work that morning, they were met by a quiet delegation of the villagers standing atop a large pile of wood wrapped in an enormous white cloth. "What is this?" they asked. "This is the cloth we use to wrap our dead," the village head replied, "and dead is what this project is. We would rather have no bridge and no wood than go back to the corrupt ways of the New Order. From now on we only want projects that involve us in decisions. If KDP can do it, other projects can do it too." And with those words, the villagers got back on their trucks and went home. Source: Guggenheim 2006. 3.3. ADMINISTRATIVE ACCOUNTABILITY IN DECENTRALIZATION In a decentralized public sector, administrative accountability is about local civil servants being accountable to their top administrative officers and to outside bodies such as public audit officers, ombudsmen, and regulators or a particular administrative agency about the use of their administrative discretion. We argued in chapter 2 that to have administrative autonomy, local governments need to be endowed with at least three administrative discretions that reinforce downward accountability. First, they should be able to make, change, and enforce laws and regulations pertaining to local administrative affairs (regulatory autonomy). Second, they should be able to govern the procurement and service delivery processes in their jurisdiction. And finally, they should be able to have control over the local civil service and employment policies. 44 In the decentralization process, in line with the increased administrative decision-making power, strong administrative accountability mechanisms are needed to ensure that local governments use their newly granted powers properly. These administrative accountability mechanisms should balance upward and downward accountability. In the case of upward accountability, a higher-level authority should oversee local governments' administrative actions for compliance with national laws and regulations. In the case of downward accountability, administrative accountability mechanisms should provide citizens with tools to be the ultimate forum. The following sections underlie both public and social accountability approaches to improve downward accountability of local governments in administrative affairs. 3.3.1. Public Accountability Approaches Administrative decision making is a complex procedure that takes into account the objectives set by political leadership, statutory provisions, and other governmental regulations. Consequently, administrative accountability is achieved through the use of a wide variety of mechanisms designed to prevent corruption and ensure that public officials are answerable to their hierarchies, relevant independent public bodies, courts, and the people they serve. Three major mechanisms that public sector approaches rely on to improve administrative accountability are structures within bureaucratic hierarchies, specially designed independent bodies, and administrative courts. 1. Accountability structures in the bureaucratic hierarchy imply that civil servants are responsible to higher administrative officers, public audit offices, supervisors, and legislative bodies. Higher- level authorities may instigate investigations or audits on the use of administrative discretion by lower-level bureaucrats. These methods include internal control and audit, and are important as a first step to uncover information about maladministration by local governments. 2. Independent bodies can conduct administrative audits on local governments. Accountability through such specially designed independent bodies emerged as a mechanism in response to the increased complexities of bureaucracy and new government challenges that require specific expertise. As a result, creation of independent agencies has become a widespread accountability measure. Examples of independent bodies include independent auditors (external auditors) who scrutinize the use of public funds for signs of misuse, ombudsmen that hear citizens' complaints about local governments' regulatory decisions and actions, or theme-specific bodies (such as the counter corruption commissions, environmental review boards, or commissions for sustainable development) that address compliance issues with sector-based standards set by the national government.24 Theme-specific bodies analyze whether or not the local administrative decisions are in line with the sectoral strategies that the national government has set. These bodies may also focus on the process through which decisions are made or on compliance with national standards in the same area of the administrative decision or action. The presence of independent bodies is limited at the local level, but the demand for their creation is growing. 3. Administrative courts with local expertise may play a crucial role in addressing local conflicts arising from local governments' regulatory decisions. These institutions are also helpful in ensuring the compliance with national laws, regulations, and standards. Existence of such a court system is especially important as it has binding decisions on legal compliance by local 24The legislature has set up these agencies to make inquiries, to obtain information, and to make regulations or judgments. Although their judgments may lack sufficient legal sanctions, they can cause departmental embarrassment and, to a limited extent, governmental changes in policy and decisions (Zarei 2000). 45 governments, and it may resolve cases that the local governments and associated independent bodies failed to address. Along the discretions mentioned in chapter 2, these mechanisms address concerns for administrative accountability of local governments in relation to three areas: a) regulatory decisions and actions; b) civil service policies and practices (employment and pay policy); and, c) service delivery and procurement administration. Improving administrative accountability for regulatory decisions and actions. Chapter 2 discussed administrative autonomy for regulatory decisions. With administrative autonomy, local governments would have the authority to approve and issue generally binding ordinances regarding public matters in their jurisdiction, subject to national and state laws. Regulatory powers of local governments may extend to local economic development, housing, land use planning and management, zoning regulations, and public safety. In certain cases, these powers could even stretch to some aspects of public health, social protection, education, and environmental protection. Such extended powers by local governments increase the effect and importance of local government regulatory decisions in local public life. Therefore, introducing measures to improve accountability of local governments in this area has become increasingly more important with decentralization reforms. Exercise of regulatory powers in these areas can take many forms, including passing ordinances and giving/restricting/revoking permits and licenses. Empowering local governments with regulatory powers required constant attention to the exercise of these powers. This attention needs to be concentrated on three levels: 1. Whether or not the regulatory decisions are in conflict with national laws, regulations, and standards; 2. Whether or not local regulatory decisions are in harmony with sectoral policies at local levels (i.e., as set by central ministries and applied at the local level); and, 3. Whether or not these decisions are applied in a just, equal, and impartial manner to all citizens living in that jurisdiction (whether maladministration causes injustice). Improving administrative accountability for civil service practices. As discussed in chapter 2, decentralization grants local governments greater administrative autonomy by enabling them to pay their staff from own budgets, to have decisive control over staffing levels and skill mix, and to offer their staff enough incentives, flexibility, and opportunities for career advancement and performance outcomes. A decentralized framework for civil service management can help break dependencies on higher levels of governments and promote downward accountability. Yet adequate checks and balances should also complement exercise of these new discretions. The most important aspects of oversight over the exercise of civil service autonomy include measures to improve budget transparency on staff payments, policies and practices of new hires (existence of individual merit-based recruitment mechanisms), and on budget and establishment control practices (controlling staff numbers and the authority to remove surplus staff). A civil service system with merit-based recruitment and flexible career and performance management could foster downward accountability, provided it is supported by appropriate budget transparency measures. Public administrative accountability approaches foresee that the system (including payrolls, staff levels, recruitment practices, and staff performance) should have transparent mechanisms that can be internally and externally monitored. These should include channels for administrative audits that can be initiated by both internal (i.e., elected leaders) and external actors (i.e., civil society groups) for accountability to be more downward and citizen-oriented. The role of citizen-oriented activities will be discussed in the next section under social accountability approaches. 46 Improving administrative accountability for service delivery and procurement. Public approaches to promote downward accountability on the use of administrative discretion over service delivery and procurement entails monitoring steps on decision making, service delivery, and so on. The process often gets very complicated as procurement relationships and production of services involve third party actors such as the private sector, communities, or civil society organizations. These relationships, in the absence of a robust system for accountability, may be prone to capture, corruption, favoritism, and patronage. The risks are not limited to who is granted certain tenders or who is awarded procurement for production of certain services. The need for strengthened accountability mechanisms is also important to improve the quality of services after the contractor or service provider is selected. In this sense, the relationship between the actors who produce the services and the local executive branch becomes a critical area that needs constant vigilance. Local-level service delivery relationships usually involve accountability relationships between local executives and local bureaucrats, and between local executives and deconcentrated units. According to Schroeder (2004), "the underlying idea is that local governments are to be accountable to the local residents through the services provided locally (but produced through the efforts of the bureaucracy). They are also accountable to the central (or regional) government that has set the rules governing local government behavior and has provided at least some of the local government's resources." The relationship between local executives and the local bureaucracy embodies a complex set of practical problems. Local service producers, in other words, the local bureaucracy, are supposed to carry out the production of local services according to policies decided by the local elected representatives.25 Local service producers, in turn, are usually answerable primarily to the local government's executive branch.26 The relationship between the central government or its deconcentrated units and local governments presents no less complicated problems in establishing sound administrative accountability systems. As discussed in chapter 2, central governments or their deconcentrated units set the rules under which local governments operate, and provides a portion of local government's financial resources for service delivery. Local governments, mostly local executives who decide about the service delivery and procurement, are accountable to higher-level hierarchies about their conduct. Public approaches are necessary to provide the appropriate channels for uncovering basic information on regulatory decisions, civil service practices, and services as described above. Yet public approaches alone fall short of ensuring proper use of administrative discretion. The deficits in public accountability approaches in the administrative dimension are manifested in a number of ways, all of which tend to hurt the poor disproportionately: i) information on resource flows is withheld, making it easy to divert funds in public works programs or capture supplies such as medicine, textbooks or seeds/fertilizers; ii) administrative decisions are taken and implemented, without properly consulting the concerned stakeholder groups or without expert advice; iii) from the citizens' perspective, the most critical accountability failure is the lack of oversight of local government services provision, which often leads to a poor standard of services. Bureaucratic reporting systems tend to have explicit anti-poor biases. The following section highlights the role of social accountability approaches in addressing such deficits in public accountability approaches. 25In the case of private production of local public services, the local bureaucracy oversees private contractors. 26 Under the political accountability section in this chapter, we also discussed the role of elected officials in overseeing local service delivery through mechanisms available to them for local council oversight. The focus here is on the relationship between local executive branch (service provider) and bureaucrats (who produce the services). 47 3.3.2. Social Accountability Approaches As a result of shortcomings in public accountability mechanisms, social accountability approaches developed mechanisms to achieve downward accountability through the use of administrative discretion at the local level. These citizen-oriented efforts mostly concentrate on producing the most relevant information about local governments' administrative decisions and actions that affect local citizens. In doing so, such efforts tend to focus more on monitoring the quality of services and on the way contracts and tenders are given at the local level. Recently, innovative approaches have emphasized the need to have a role for citizens in initiating administrative audits to reveal more information on employment and pay policies as well as relationships between local governments and service providers. The following explains each of these aspects of the civic role in promoting downward accountability of local governments in administrative matters. Information provision as a basis for citizen monitoring. The lack of information on financial allocations to local authorities often leads to the abuse of funds. To make information available to the public is therefore often the starting point of many social accountability initiatives. In India, a small Indian NGO in Rajasthan initially introduced public hearings in 1994 to stop fraud. Detailed accounts of official expenditure records were presented in a public meeting, and citizens were invited to testify discrepancies between official records and their personal experiences. Laborers, suppliers, and contractors were invited to verify payments on the documents. This initiative led to a major campaign ("Every Citizen has the Right to Information Campaign") and in 2005, India's Right to Information Act came into effect. The central government act provides for every citizen's right to public information. On payment of a small fee--which is waived for persons living below the poverty line-- every citizen can demand and receive details of expenditures on the work done over the last five years in his/her village. In many countries the right to information does not exist for many administrative activities, and citizen groups have to establish their own networks to make information public. The Uganda Debt Network (UDN) established local monitoring committees in order to track local public expenditures (de Rezio, Aseem, and Ramkumar 2006).27 In 2002, UDN set up community-based monitoring and evaluation system in eight districts. Village- and district-level committees monitored local expenditures and promoted dialogue around problems that could be solved locally. For example, if community monitors found that certain teachers were not doing their jobs properly or were absent from work, the problem was discussed with the village authorities, and the individual could be disciplined or transferred. On the other hand, if monitors noted that building material for school construction did not conform to the relevant specifications, it was brought to the attention of the district authorities. Often a non-confrontational approach is used to demand corrections, building on the committee's capacity to negotiate, the openness of government officials, and the respective public accountability processes. At the same time, UDN backs up its claims through the use of media, like radio programs or regular anti-corruption campaigns. UDN's work on at the district and sub-district levels on budget implementation, service delivery, and anticorruption allows it to generate inputs and policy recommendations for the national level with a local perspective. For example, UDN was monitoring the implementation of a national school facility grant and discovered a widespread inefficiency in fund utilization. Its subsequent national campaign led to a change in the guidelines and a blockage of the loopholes. Monitoring Procurement and Implementation of Local Government Contracts. The contracting and implementation of public works and services are functions that suffer from a high risk of corruption and 27UDN was formed in 1996 and developed an expertise and reputation for its national budget advocacy. 48 of mismanagement. A number of citizen-based initiatives have emerged to complement the internal government accountability mechanisms that manage these risks. They can be grouped under three categories: (i) supervision during the contracting and bidding process; (ii) monitoring of the construction process; and, (iii) supervision through ex-post controls and audits of budget execution. A typical source of local government corruption and collusion involves drafting the tender documents in ways that unfairly benefit one contractor over others. Citizens have organized public consultations in which different parties get a chance to comment on the draft tender document before the start of the bidding process. In addition, independent outsiders can be involved to conduct an in-depth analysis of the tender document (see box 3.4). In the Philippines, the Local Government Code defines special bodies in which accredited NGO and CBOs have a seat in the pre-qualification, bids, and awards committee for local contracts. To increase transparency, citizens have also been involved in overseeing the opening and analysis of the bidding offers. In Nicaragua, for instance, the Social Investment Fund encouraged community organizations receiving investment projects from the municipality to be present during the opening and analysis of the bidding offers (Grun 2000). Citizens can also oversee the construction process while it is taking place. Citizens are trained to oversee so that funds allocated to an investment are spent as budgeted and that the physical construction follows the standards agreed to in the contract, e.g., the correct amount of cement, thickness of the walls, or depth of the well. Box 3.4. Overseeing the local government tendering process In Argentina, the Municipality of Morón, assisted by the local chapter of Transparency International, introduced two mechanisms to monitor the contracting of the waste collection service, which had been widely criticized for alleged corruption during the previous administration. At a public hearing (extraordinary session of the City Council) that 500 people attended, participants discussed the draft tender document with the bidders. Next, through an integrity pact, the hearing helped establish mutual commitments between the local government and the bidders on issues such as sanctions for bribery and public disclosure of the award decision. As a result of the hearing process, the contract value for waste collection services was reduced from about $45 million to $32 million. Source: Transparency International 2001. In El Salvador, the municipality of San Antonio del Monte entered into a partnership with the beneficiaries of a six-kilometer local road. The beneficiaries formed a social audit committee that monitored the physical construction process, from the receipt and quality of the materials to their proper use. The committee interviewed individuals and entities such as the mayor, the head of the institutional procurement and contracts unit, project technical staff, and the general public to ascertain the project's budgetary characteristics, the quality and quantity of resources, timetable, and needs. Guided by the committee, the community conducted ongoing evaluations of the physical progress of the public works project, a task that required precise technical expertise. The committee reported to the local development committee and the local government through out the process (World Bank, 2003; Vidaurre, 2003). Monitoring Local Service Provision. The actual service provision quality and level are probably what citizens care the most about. There are a number of strategies that citizens have used to exercise oversight in this area. Some rely on participatory assessments and feedback surveys, and are often accompanied by agreements on the expected standards of services. Others rely more on public representation in service- specific institutions that channel citizens' complaints and allow them regular oversight. One of the main innovations that drew attention to the potential of the social accountability approach was the experience of citizen report cards in Bangalore, India (Paul 2002). Citizen report cards are participatory surveys that solicit user feedback on the performance of public services. They are used in situations where there is no demand-side data, such as user perception on quality and satisfaction with 49 public services. The citizens' report cards are instrumental especially in gathering demand-side data about state-owned monopolies many of which lack incentives to be responsive to their clients. The report card process relies on extensive media coverage and civil society advocacy to achieve greater accountability. The initial experiment in the municipality of Bangalore proved that a powerful way to improve local government services was by collecting citizens' feedback about the performance of local services in a structured way and using that assessment as a yardstick against which to measure future improvements. This basic concept has led to a proliferation of initiatives that use different types of participatory performance assessments to evaluate local government services. In Uganda, for instance, Kampala conducted its first citizen report card in early 2005. It provided the city council and other basic public service providers with feedback on water and sanitation, health, education, roads and public transport, solid waste management, public toilets, the management of the city environment, maintenance of law and order, and management of city infrastructure (Kampala City Council and Uganda Management Institute 2005). A complementary strategy has been to develop citizen charters. These are pacts between the community and the service providers that spell out expectations and roles, enabling the citizens to interact more effectively with the municipality. They specify the expected standards of the services, identify who is responsible, and outline the procedures for the redress of complaints. For example, the Citizens' Charter in the Municipality of Mumbai, India, covers detailed public services for each municipal department (box 3.5). Box 3.5. The citizens' charter in the municipality of Mumbai, India This charter covers specific public services for each municipal department. For example, for the Solid Waste Management Department, it specifies such items as the disposal of dead animals, the authorized collection spots from where garbage is to be picked up, the maintenance of public toilets, the collection and transportation of garbage, the sweeping of roads, the cleaning of dustbins, and other sanitary measures. Citizens can lodge a complaint if the garbage disposal truck is not covered properly. They can demand a response from the municipal corporation within 24 hours from the time the complaint is filed. Other services included in the charter are: solid waste management, waterworks, sewerage, storm water drains, road maintenance, traffic, public health, licenses, environment sanitation, education, electricity, transport, building proposals, shops and establishments, gardens, and the Mumbai fire brigade. Source: World Bank, 2005. Other strategies have relied on the creation of new institutions to promote citizen oversight over a specific service that the local government provides. Usually these are multi-stakeholder councils formed by different combinations of users, civil society organizations, government, and private sector representatives. Examples include the local school councils in the city of Chicago, in the United States (Fung 2004), and citizen community boards and school management committees in Pakistan (ADB [Asian Development Bank]/DfID [Department for International Development]/World Bank 2004). 3.4. FINANCIAL ACCOUNTABILITY IN DECENTRALIZATION In chapter 2, we discussed how fiscal decentralization expands the revenues and expenditures that are under the discretion of local governments. Fiscal decentralization and increased local discretion over public finances and expenditures also imply dramatic changes in terms of accountability relationships, as the budget process spans different local government branches and links national and local government. On the one hand, fiscal decentralization has the potential to improve public financial accountability, as it 50 brings agents closer to the principal and, as such, reduces the information asymmetry with regard to the oversight of public resources.28 On the other hand, if not designed adequately, fiscal decentralization may also lead to accountability failures as local governments are granted greater fiscal autonomy. Decentralization reforms, therefore, need to address these risks and ensure financial accountability, taking into account significant changes as well as an increased complexity in accountability relationships. In this respect, decentralizing countries increasingly focus on public financial reforms to strengthen the intergovernmental architecture and public financial management systems to ensure that sub-national governments have both the resources and the incentives to perform the required tasks prudently. Yet public financial accountability approaches alone usually fall short of achieving this goal. More recently, in response to deficiencies in public financial accountability measures, demand-side initiatives have grown in numbers and depth to support public financial accountability about the fiscal conduct and financial performance of local governments. The following sections focus on each of these approaches and explain how supply-side public financial accountability approaches and demand-side social accountability approaches could work well together to improve downward accountability of local governments on financial issues. 3.4.1. Public Accountability Approaches Public financial accountability seeks to ensure the transparent management of public finances in line with policy decisions. It "requires that governments manage finances prudently, that they integrate their financial and nonfinancial reporting, control, budgeting, and performance; that they report comprehensively on what they have achieved with their expenditure of funds; and that stakeholders behave ethically" (Sahgal and Chakrapani 2000: 3). Ensuring financial accountability is especially important when greater autonomy is being granted to local governments in the fiscal decentralization context. Public financial accountability measures in this respect focus on establishing effective, efficient, transparent, and rules-based public financial management. The term "public financial management" (PFM)29 is used below to refer to all components of the budgeting process and relates to both "upstream" (preparation and programming) and "downstream" (execution, accounting, control, reporting, and monitoring and evaluation) phases of budgeting and budget execution processes (see figure 3.1). 28The principal­agent (PA) theory provides for a conceptual framework for understanding of public accountability processes. Its basic insight is that the principal, P, hires the agent, A, to undertake a particular task, but P suffers from an information asymmetry problem in motivating and controlling A. In both economics literature (e.g., Arrow 1985; Rees 1985; Laffont 1989) and its public management extensions (e.g., Jensen and Meckling 1976; Andrews and Shah 2005) the solution to a PA problem involves the development of an optimal incentive package that furthers the principal's objectives and constrains the agent's actions. 29 In practice, the terms "public financial management" and "public expenditure management" are used interchangeably to refer to all components of the budgeting process. 51 Figure 3.1: The public financial management cycle Planning Budget execution Accounting, External Legislative (Managerial Monitoring Audit Oversight (Budget + Activities+ and Reporting Strategic) Procurement) Internal Control Internal Audit Source: Baltaci and Yilmaz 2006. The Public Expenditure and Financial Accountability (PEFA) Initiative (see box 3.6) in this sense summarizes the core dimensions of good public financial management that promotes transparency and accountability in each step of the cycle. Box 3.6. The Public Expenditure and Financial Accountability Initiative defines core dimensions of an effective PFM system: 1. Budget credibility: The budget is realistic and is implemented as intended. 2. Comprehensiveness and transparency: The budget and the fiscal risk oversight are comprehensive, and fiscal and budget information is accessible to public. 3. Policy-based budgeting: The budget is prepared with due regard to government policy. 4. Predictability and control in budget execution: The budget is implemented in an orderly and predictable manner, and there are arrangements for the exercise of control and stewardship in the use of public funds. 5. Accounting, recording, and reporting: Adequate records and information are produced, maintained, and disseminated to meet decision-making control, management, and reporting purposes. 6. External scrutiny and audit: Arrangements for scrutiny of public finances and follow up by executive are operating. Source: PEFA 2005. Without having an effective public financial management system at both the central and local levels, unintended consequences of a fiscal decentralization program can be fiscal imbalance, weak accountability, political capture, and deterioration in public services. The risks are even higher at the local level. There is increasing evidence that weak or absent local public financial management systems "are likely to negate any advantages that might be inherent in bringing public services `closer' to local communities" (Ahmad, Albino-War, and Singh 2006: 405). Ensuring financial accountability, therefore, needs to be an integral element of the decentralization process. The outcome of fiscal decentralization critically depends on the ability of local governments to manage revenues and expenditures effectively. In the absence of sound local public financial management systems, financial accountability cannot be ensured, as local governments are vulnerable to waste, corruption, and inefficiencies (Baltaci and Yilmaz 2006). Where public financial management mechanisms are in place, they often show significant weaknesses, ranging from the lack of control systems for safeguarding against abuse, misuse, fraud, and irregularities (e.g., in Bosnia), to inefficient cash management, collusive practices in procurement, and weaknesses in internal control and audit (e.g., in India) (Baltaci and Yilmaz 2006). Evidence so far indicates that there 52 are significant shortcomings with regard to public financial accountability in preventing the misuse of public resources, resulting in fraud and corruption, among others. In many countries, public financial management systems do not necessarily produce information about public sector expenditures in a timely manner for people to keep track of service delivery expenditures. There are also long delays in conducting the audits and making the results public in many developing countries (Olowu 2001). Delays in auditing emasculate the effectiveness of the external audit as an accountability mechanism. Producing reliable information on expenditures, service delivery, and budgets through external audits is important in addressing information asymmetry problems in two respects; it facilitates local councilor oversight over local executive bodies, and it uncovers crucial information about critical services and associated expenditures needed for effective citizen monitoring. Other public financial accountability shortcomings are related to weak links between policy making, budget planning, execution, and outcomes achieved. Policy decisions are compacts between political leaders and citizens. Policy making and budget planning should be interlinked as budgets are policy choices, resulting from a compact with citizens. However, in many countries, these links are not in place. More importantly, the lack of hard budget restraints on political decision makers during the budgeting process leads to inadequate funding for important services such as operations and maintenance. Furthermore, it leads to poor expenditure control and the unpredictable flow of funds to service delivery in crucial sectors such as health and education. In the absence of such linkages, it is quite difficult to ascertain outcomes based on policy or budget planning. This makes it all the more difficult to identify who is responsible for which outcomes. Table 3.1 summarizes shortcomings with regard to financial accountability at the local government level in selected countries. As a response to these shortcomings, additional, complementary social accountability tools have emerged where citizens play a key role. To prevent the misuse of public finances in connection with fiscal decentralization processes, these social accountability tools should be promoted as complementary to public financial accountability mechanisms. 53 Table 3.1: Local governments and internal controls in selected countries Country Internal Control System at the Local Level Issues Arising Due to Lack of Internal Controls Argentina Lack of legal instruments and no political High level of indebtedness in local governments willingness to improve internal controls and failure to provide urban services and audit systems Bosnia Lack of contemporary internal controls Impaired safeguarding measures abuse, misuse, and audit systems fraud, and irregularities; widespread corruption; misconduct and misuse of public funds; and public disaffection against government institutions China Ex-ante expenditure control and Common problems in compliance with laws and compliance audits regulations; unlawful tax practices Colombia Unclear legal framework defining the Negligence, corruption, and misuse of public funds functions and responsibilities of the fiscal control agencies; ineffective internal control mechanisms India- Inefficient controls and audit practices, Frequent case of abuse, misuse, and fraud; Karnataka old fashioned rule books, lack of timely irregularities and malpractices in procurement; State and reliable information; focus on lack of coherence to the stated rules and compliance audits, and inadequate procedures follow-up and audit findings Indonesia Weak internal control and audit systems Unethical and uneconomic operations due to pervasive corruption, inefficient cash management; and collusive practices in procurement Philippines Weak internal control environment, Lack of compliance with laws, rules and nonexistent internal audits, and lack of regulations; fraud and irregularities; and overpaid timely financial information public purchase and procurement Source: Baltaci and Yilmaz 2006. 3.4.2. Social Accountability Approaches As a response to weaknesses of public financial accountability systems, social accountability mechanisms have emerged in many countries, aiming to address these shortcomings. While social accountability initiatives were originally introduced by civil society organizations, governments have picked up some of the tools and instruments and incorporated them into federal and state legislations. These have become known as hybrid accountability mechanisms, where the distinction between public and social accountability have become blurry. Citizens' participation in budgeting and expenditure processes. Citizens' participation in budgeting and related processes, such as expenditure tracking, has increased substantially over the past few years. Until recently, the budget process was seen as a purely technical matter and, as such, was dominated by the executive branch of national and local governments. The legislative and the public often lacked the capacity to scrutinize budget processes, particularly at the local level. However, public interest in budgets 54 has risen dramatically over the past decade, mainly because it has become evident that budgeting processes are crucial for ensuring that policy implementation is actually in line with development strategies and planning. Today, legislators at the national and local levels are getting increasingly involved in budget debates, and the media is more active in reporting on the misuse of public expenditures. This greater openness of the public budget process has been largely the success of civil society organizations that have acquired skills and confidence to intervene in budget processes. The International Budget Project (IBP)30 estimates that close to 100 organizations in 70 countries were engaged in this type of activities in 2005, compared to 10 organizations a decade earlier. The potential for social accountability with regard to budgeting and financial management processes is vast. It includes: · Making budget information available to the public and bringing transparency into the budget process, · Demanding consistency between budget priorities and declared policy and planning objectives, · Monitoring budget execution and exposing inconsistency between budget formulation and budget execution, the leakage of funds and corruption, and, · Finally, bringing about a dynamic balance between the limited fiscal space of local governments and local needs, by increasing budget discipline and the local revenue base. Civil society organizations have developed a range of mechanisms and tools to intervene and track budgeting, financial management, and execution processes, such as independent budget analysis. The independent budget analysis is a way to make budget information available to the public, influence budget allocation and revenue policies, and initiate debate on sector-specific implications of budget allocation, (e.g., gender implications). Generally, participatory processes have opened a public space to demand for and address issues as budget discipline, tax reforms, and local revenues. Major social financial accountability tools and instruments include the following: Public hearings on budget information. This was explained above. Participatory public expenditure tracking (PET). Expenditure tracking is a method to monitor flows of public resources for the provision of public services and goods and uncover leakages and corruption in the system. Participatory public expenditure tracking (PET) for earmarked grants was first applied in the health and education sectors in 1996 in Uganda.31 Findings revealed that, among others, a large percentage of funding (80%) was captured and never reached the schools as intended. These findings promoted a mass public information campaign by the Ministry of Finance, media, and civil society--a signal to local governments to improve transparency and efficiency, as well as lower the cost of information to parents. Follow-up surveys in the education sector showed a major improvement, reducing grant capture to less than 20 percent (Reinikka and Svensson 2003). 30In 2006 the IBP has launched an Open Budget Index for 59 countries in collaboration with researchers and NGOs. See www.openbudgetindex.org. 31The data collected from different levels of administration, including 250 schools and 100 health facilities, showed that: More than 80 percent of intended capitation grant was captured and never reached schools. Large schools with wealthier parents and qualified teachers were able to obtain more of their budget allocation; enrollment trends differed from published data. 55 Performance-oriented, citizen-based, ex-post auditing. In other initiatives, citizens performed physical and financial audits of the local government accounts. By comparing the written records with the actual outputs and discussing them in public venues, citizens forced local governments to answer the following kinds of questions: Has the local government spent its money on the goods and services that its accounting books say it has? Has it paid the market price to its providers and contractors? Have the purchased goods and services been delivered to their final destination in the quantity and quality that has been paid for? Civic monitoring of inter-governmental fiscal transfers. Local governments across the world depend on central government transfers for the provision of local services. In many developing countries, central governments allocate these transfers with low levels of transparency and accountability, undermining downwardly accountability of local governments towards their citizens. For this reason, some social accountability initiatives have tried to address this failure of public accountability mechanism by monitoring the flow of grants to local governments and denouncing abuses in their allocation. In Ghana, for example, a group of civil society organizations led by the Integrated Social Development Centre conducted a study to track the disbursement of the District Assemblies Common Fund (King et al. 2003), a constitutional provision that reserves 5 percent of national tax revenue to fund development activities by districts and the largest single source of development funding for many districts. The study examined the method of allocation to local authorities, actual amounts, and uses of the allocations, guidelines, and compliance for utilization, and identified the weaknesses of the fund's administration.32 While district authorities followed established procedures for the use of the funds, the study revealed that the administration of the fund itself was problematic. This included delays in the releases of funding; discrepancies in the amounts allocated, disbursed, and received; and misuse and discrimination in the selection of projects, contract awards, and cost and quality of projects. The study made recommendations to improve awareness and public education on the processes and identified ways to minimize political interference and institutionalize tracking of expenditures through participatory methods. Impact of local governance and accountability on tax compliance. Evidence has shown that improved local governance and participatory budgeting and expenditure processes improved citizens' ownership for local policies and their willingness to pay taxes. In the Philippines and Brazil, for example, the level of tax compliance is a core accountability signal that citizens can send to local governments regarding their performance. Citizens tend to show lower tax compliance towards corrupt and poorly performing local governments than towards more transparent and effective municipalities. Increased legitimacy through participation and the improved tax collection go hand-in-hand and can be a mutually reinforcing process. Increasing the share of the local budget financed through own-source revenues is one of the key measures to motivate citizens to demand accountability from local governments. A central way to increase tax collection is by improving the legitimacy of local governments in the eyes of citizens. The city of San Fernando in the Philippines was able to increase its local revenue by around 60 percent between 1996 and 2001, after a reformist mayor institutionalized citizen participation, reduced bureaucratic expenses, and implemented a range of highly visible projects. The city government gradually gained enough legitimacy to address tax issues successfully. It launched an information campaign on taxes, created incentives for citizens to pay taxes, overhauled the tax office, and created a one-stop shop for various government services. In addition, the city government could address less-popular measures such as the creation a tax enforcement unit and a watchdog body that reported on tax avoidance of businesses (Guiza 2002). 32The study was conducted in four district assemblies in collaboration and with financial support from DfID and the World Bank. 56 In Porto Alegre, the city could was only able to allocate one percent of its budget to participatory budgeting when participatory budgeting started. Over the following 16 years, it was able to increase the citizens' tax burdens, particularly for the middle class, and to multiply tenfold the percentage of the budget dedicated to participatory budgeting. Schneider and Baquero (2006) describe how fiscal and institutional changes brought together a system of "embedded public finance" in which state actors targeted social groups with the benefits that each group wanted and, in exchange, the government secured the assets that each group could offer. Poorer segments of the population profited from participation and from pro-poor investments through participatory budgeting in exchange for political support. The middle class received better services and governance in exchange for tax contributions (Schneider and Baquero 2006). 3.5 CONCLUSION This chapter argued that local discretion provided by decentralization reforms should be accompanied by an accountability framework that combines supply- and demand-side approaches. It is important to note here that supply- and demand-side approaches should and can work in a complementary way; neither of these approaches alone can fully achieve the intended outcome of downward accountability in decentralization structures. Treating them as two necessary components of the same objective could eliminate the existing tensions between these two approaches. This is possible by developing more comprehensive local governance strategies (both by the donors and the countries implementing decentralization reforms) that bridge the two approaches together. Chapter 4 provides guidance to client countries on dealing with these issues and highlights operational challenges in putting such an accountability framework in practice. 57 Chapter 4: Reconciling Accountability and Discretion in Decentralization The previous chapters of this report outlined the main elements of an analytical framework that links discretion and accountability in the context of decentralization structures. In this framework, representative local governments are basic institutional elements of decentralization where they are accountable to upper levels of government and to citizens as allowed by their discretionary space. The framework introduced in this report evaluates how these representative local governments, provided with a meaningful level of discretionary powers, enter into accountability relationships with their surrounding local actors. Discretion is given to representative local governments so that they are able to use their best judgment to meet citizens' needs and preferences. However, there are tendencies for misuse and abuse of new discretionary powers. Hence a good local governance system is the one with strong accountability links that have safeguards against its abuse. In this regard, the main focus of this study is on strengthening downward accountability of local governments through an approach that combines supply- and demand- side accountability measures. The issue of discretion and the potential for abuse of power are critical to strong state­society relations and to cooperative development under a decentralization framework. In a given country context, the way local government discretion is identified and the nature of accompanying accountability systems have a colossal impact on local governance outcomes. To avoid governance practices that result in waste, misallocation, and corruption, local discretion granted through decentralization should be coupled with efficacious mechanisms for strengthening downward accountability. In many decentralizing developing countries, the capacity of enforcement institutions (for example, judiciary and comptroller) is very weak. In these countries, monitoring and controlling discretion against potential for abuse constitutes a major challenge to their decentralization reform efforts. Until recently, efforts to foster good governance focused on strengthening the supply-side of accountability in state institutions. Accordingly, donor initiatives supported institution building in developing countries to increase the supply of governance processes through reforming public institutions. More recently, the development community has come to realize that domestic demand for accountability, originating in civil society entities and the public at large, is at least as important for development as supply-side mechanisms. Also referred as social accountability, demand side refers to an approach toward building accountability that relies on civic engagement--in which ordinary citizens and/or civil society organizations demand accountability. Accordingly, this paper argues that downward accountability of local governments follows the discretions provided to local governments and it should be strengthened by integrated approaches marrying supply-side public sector accountability with demand- side social accountability instruments. With the emergence of social accountability, in response to limitations in supply-side approaches, the real challenge is how to reconcile the approaches in practice so that local governments can be downwardly accountable to citizens. A primary precondition for achieving this is that decentralization reforms should simultaneously empower local governments and citizens. These twin goals require prioritization and sequencing of decentralization reforms, which strengthen both supply-and demand-side accountability institutions.33 33The theoretical literature on the issue of sequencing and implementation of decentralization programs provides a normative discussion on broad steps to be taken in sequencing decentralization reforms. However, it is silent about 58 This final chapter brings together the findings and conclusions of the previous chapters with the objective of charting a way forward for the audiences34 of the report to prioritize and sequence decentralization and accountability reforms in the context of effective and responsive local governance. It also highlights challenges in operationalizing the framework presented in this study through the World Bank operations on local governance. 4.1 POLICY RECOMMENDATIONS The ultimate goal of decentralization reforms is to have a local government system with a high degree of decision-making autonomy (discretion) and accountability (quadrant III in figure 4.1). Achieving this goal requires providing support to enhance the supply of local governance processes by reforming public sector institutions and by creating demand pressure for better governance. To this end, we present a set of policy recommendations that are intended to identify issues of importance for the report's primary audiences in moving their country's local governance structure to quadrant III. Figure 4.1, below, shows three possible trajectories for local governance turnarounds and related scenarios for sequencing of local governance reforms (although, admittedly, neither one of these trajectories is superior to others). The starting point for all trajectories is quadrant II where both local government discretion and accountability are very low--highly centralized countries with no local government accountability. The end point is quadrant III where local governments have a high degree of discretionary power accompanied with a high degree of accountability towards citizens. The first trajectory signifies a country context where the focus is on discretion in the short-run (e.g., big- bang decentralization in Indonesia). In this group of countries, decentralization reforms placed an initial emphasis on increasing political, administrative, and fiscal discretionary power of local governments. Several well-known examples of decentralization reforms--most notably, in recent times, Indonesia-- exemplify this trajectory. In Indonesia, the big bang decentralization devolved functions and resources through two national laws (Law 22/1999 and Law 25/1999). Supporting laws and regulations, as well as other institutional measures to strengthen control and improve accountability of the local governments, were introduced much later. In fact, there has been an ongoing process to bring greater accountability in local governance in Indonesia's decentralization. The next step in their reform efforts is to strengthen the accountability of local governments towards citizens. The second trajectory denotes a scenario where the national decision makers put a higher priority on promoting political pluralism, administrative accountability, and financial safeguards in the short run, as was the case with ex-Communist transition countries in the Eastern Europe. The third trajectory represents a balanced situation where both discretion and accountability are strengthened simultaneously. Arguably, in the real world, there are few instances of a perfect diagonal trajectory. the linkage between decentralization and broader accountability reforms. There are very few papers analyzing this issue. See for example, Spahn (2005) and Bahl and Martinez-Vazquez (2006). 34As discussed in the first chapter, the primary audiences for this report include: (1) influential public policy makers in client countries; (2) local governance stakeholders in client countries; (3) governmental and non-governmental practitioners and policy advisers working on local governance, local development and public sector reforms; and (4) World Bank staff advising countries on local government reforms. The secondary audience includes academics in the fields of economics, public administration, public policy, and social development. 59 Figure 4.1: Local governance turnarounds--trajectories Trajectory 1 Trajectory II Trajectory III Discretion Discretion Discretion High I III II IV Low Accountability Accountability Accountability Source: Adopted from Levy (2006). Initial turnaround Desirable follow-through Whatever the eventual trajectory may be, reaching to quadrant III requires taking certain actions on the part of policy makers. To guide this process, table 4.1 provides a set of recommendations in the areas of political, administrative, and fiscal decentralization that links discretion with accountability in the local government context. These recommendations are divided into two groups of action--supply- and demand-sides--denoting a twin focus: improving local level public sector capacity while concurrently developing demand-side institutions of local governance. They aim to inform various parties involved (including policy makers in client countries and donors) about the issues in designing a decentralization strategy. It should be stressed that the table does not pretend to be a prescriptive "cookbook" that is a recipe for all countries and contexts-- rather, it aims to provide a concise list of issues that may be relevant to a larger set of countries and need to be prioritized and sequenced. The decision to decentralize is inherently a political one. Moreover, accountability and autonomy (discretion) are end attributes of particular political systems. To this end, the first part in table 4.1 discusses the actions needed to strengthen political aspect of discretion and accountability. The tools listed under strengthening political discretion of local governments include electoral laws and direct democracy tools. Their main objective is to empower representative systems of local governance and strengthen citizen participation. On the accountability side, supply-side actions aim to improve electoral system quality with ballot measures and initiatives. Demand-side actions are intended to support citizen- based initiatives to increase citizen oversight over local government policies. They mostly focus on performance of elected political leaders and how representative the local level electoral and party political systems are. The second part in table 4.1 discusses the actions needed to strengthen administrative aspect of discretion and accountability. The tools for under strengthening administrative discretion of local governments cover a wide spectrum--from legal and regulatory tools to planning and enforcement tools. On the accountability side, actions aim to generate momentum for a culture of continuous improvement and greater compliance across local governments; provide an early intervention option for central governments to local governments that experience financial distress; promote good governance and ethical conduct principles; identify and share innovation and good practice in local governments; and enable the central government to use review information to provide feedback into its work in identifying necessary legislative and policy work for the local government sector. Demand-side actions are intended to complement supply-side mechanisms by increasing citizen awareness. Active citizen participation and civic engagement are part of the system of checks and 60 balances that prevent those in power from abusing discretion. Citizen awareness and expectations are a crucial element of an effective local governance reform strategy. Supply-side actions should increase citizens' level of awareness and understanding on laws, rights, budgets, and policies. The third part of table 4.1 concentrates on how to strengthen the financial/fiscal aspect of discretion and accountability. Intergovernmental tools are used to strengthen financial/fiscal discretion. They focus on strengthening the intergovernmental fiscal framework and design aspects of the local government finance system--the rational assignment of expenditures and revenues, transparency and predictability of transfers, and so forth. On the accountability side, supply-side actions aim to establish robust public financial systems. An effective, efficient, transparent, and rules-based public financial management (PFM) system is an essential tool for a government in fiscal decentralization program implementation. PFM reforms support fiscal decentralization process by promoting transparency and accountability in the use of public resources, ensuring allocation of public resources in accordance with citizen priorities, and supporting aggregate fiscal discipline. There are a number of PFM challenges that local governments in the developing world face today: poor planning; lack of linkage between policy making, planning, and budgeting; weak accounting systems; inefficient procurement systems; waste and errors in service management; poor cash management; poor reporting of financial performance; lack of audit and control; and limited capacity. Demand-side actions address these shortcomings. Public sector institutions produce a lot of information, such as budget allocations and service delivery targets, which are easily understood by ordinary people. Local government reform operations should support nonstate institutions, i.e., social accountability institutions, in monitoring local government spending and service delivery performance and producing information for public consumption. 61 Table 4.1: Policy Recommendations to Strengthen Discretion and Accountability POLITICAL SETTING AND ACCOUNTABILITY How to Strengthen Discretion How to Strengthen Accountability Strengthen electoral system at the local level Supply side Ensure free and fair elections at the local level Review and revise, as needed, the role of campaign financing rules that may favor large, Clearly define rules regarding electoral system and broader centralized parties and make it more difficult for local candidates or new parties to compete guidelines for local political competition for mayors and councilors Establish, monitor, and enforce transparent rules for financing local elections Allow for competitive multi-party systems; remove restrictions before Improve the quality of the electoral system recall, write-in, and independent candidate local party competition options Provide legal safeguards against dismissal of elected local councils by Ensure representation of marginalized/vulnerable groups (reserved seats, quota systems) higher levels of government on frivolous grounds Institute consultation mechanisms such as public hearings and integrate these into formal decision-making processes Ensure a legally mandated term for elected public officials (i.e., not too short or too long to serve to local populations) Secure and strengthen the role of elected body/ies (council) in overseeing local government operations Establish a clear separation of power between the executive (i.e., mayor) Demand side and legislative (i.e., council) bodies of local governments Allow for citizen-initiated legislation (i.e., petitions) and referendum Assign necessary functions and mechanisms, and strengthen resources to elected local councilors to exercising their oversight functions Establish procedures for public petitions to adopt, amend, or repeal an act, legislation, or Allow, within prescribed rules, for voting measures beyond elections; executive order i.e., authority for local governments to initiate referendum Allow for citizens to initiate the recall of elected public officials. Allow for citizens to demand public hearings on policy decisions and actions Establish citizen ombudsman office in local governments 62 Table 4.1: Policy Recommendations to Strengthen Discretion and Accountability (continued) ADMINISTRATIVE DISCRETION AND ACCOUNTABILITY How to Strengthen Discretion How to Strengthen Accountability Give local governments administrative autonomy through clearly set rules Supply side Regularly conduct administrative audits (external audit) of local government transactions through Allow local governments to make and change and laws, regulations, executive permanent or ad hoc independent entities orders, ordinances affecting their locality (i.e., on spatial and sectoral planning, Establish independent judicial/quasi-judicial agencies to investigate cases of local government environmental and local economic development); remove redundant procedures corruption and misconduct of central government approvals Establish administrative courts to deal with compliance on local regulatory decision making and actions Empower local governments in enforcing local regulatory decisions; provide Develop a national procurement strategy for local governments; strengthen procurement rules and them certain tools such as powers to sanction and punish for noncompliance to processes; institutionalize independent procurement audits ensure meaningful level of administrative autonomy on local affairs Develop ethics rules, codes of conduct, and professional practice guidelines for local government employees Assign civil service and employee functions to local governments (i.e., hiring, Introduce a flexible career and performance management supported by a civil service system with firing, promotion, relocation, and so on) merits-based recruitment Establish clear rules regarding civil service practices (payrolls, staff levels, recruitment practices, Grant autonomy in local procurement and administration of service delivery (i.e., staff performance) and develop systems for periodic monitoring and timely revisions (often through in procuring goods and services or contracting out aspects of service delivery) local government commissions) Institute a grievance/complaint system as well as a whistleblower protection system Introduce a flexible career and performance management supported by a civil Set standards for service delivery and reward performance; strengthen information systems to service system with merits-based recruitment monitor service quality Demand side Secure citizen participation in decision-making processes regarding administrative affairs through neighborhood councils, community councils Allow citizens to challenge administrative decisions made by local governments in appropriate fora Support public involvement in contracting and implementation of public works projects through social audit committees Develop citizen charters for empowering the citizenry in their interactions with local service providers Support public oversight mechanisms to provide citizens a channel to oversee local service delivery and introduce measures for assessing citizen satisfaction with service delivery (scorecards, service delivery surveys) Integrate service delivery performance into formal intergovernmental processes and provide incentives for good performance 63 Table 4.1: Policy Recommendations to Strengthen Discretion and Accountability (continued) FINANCIAL/FISCAL DISCRETION AND ACCOUNTABILITY How to Strengthen Discretion How to Strengthen Accountability Clearly assign service responsibilities and revenue sources Supply side Grant meaningful autonomy over local expenditures decisions (typically, Establish minimum control conditions for intergovernmental transfer revenues (i.e., clean capital versus recurrent expenditures; kinds and recipients of services) audit reports, submission of financial statements); publicize transfer figures Strengthen local capacity for budgeting and public financial management Grant meaningful autonomy in local revenue decisions (setting tax bases Create a modern chart of accounts system for reporting revenues and expenses and rates and fee rates) Establish criteria for local revenue mobilization, expenditure efficiency, and financial management discipline, and link performance to rewards in intergovernmental systems Remove redundant procedures of central government approvals Publicize approved budgets; provide open and timely access to budget execution information; publicize end-of-fiscal-year budget Develop a performance monitoring and evaluation system to track budget and public Establish a transfer system that is flexible and responsive to the local expenditure outcomes, and make information publicly available needs and changing conditions Strengthen local level public audit systems and make audit findings publicly available Define clear rules for responsible local borrowing, including rules regarding defaults, and Bring transparency and predictability in transfers in the: (i) method for make borrowing information publicly available determining total divisible pool; (ii) method for allocating divisible pool among eligible units; and (iii) timing of transfers Define clear rules regarding hard budget constraints for local governments Demand side Establish a multilevel government coordination body to develop legal and Make local government financial information (including budgets and end-of-year financial regulatory standards for transfer schemes statements) easily accessible to public: Strengthen public involvement in budgetary process through participatory budgeting Determine the right size of the transfer pool to address vertical practices imbalances; update and publicize this information regularly Introduce gender-sensitive planning, budgeting, and resource allocation--reinforced by gender audits Support responsible local borrowing Make budget information available to the public and influence budget allocation through independent budget analysis Monitor budget execution and leakage of funds through participatory public expenditure tracking Source: Authors 64 4.2 STRENGTHENING THE BANK'S WORK IN LOCAL GOVERNANCE This section highlights the main components of a local governance strategy that the World Bank needs to develop and identifies research gaps and operational implications as next steps. 4.2.1. A Local Governance Strategy for the World Bank Currently, the World Bank is engaged in decentralization programs in over 80 countries.35 This broad support for decentralization is not incidental--as mentioned in the beginning of this report, local governance is one of the entry points for strengthening governance and anticorruption efforts in client countries, even when governance reform opportunities at the national level are limited (World Bank, 2007a). Therefore, the process of reformulating accountability relations within decentralization reforms has important operational implications for the success of the Bank's decentralization programs. Yet, the Bank's decentralization and social accountability agendas are often weakly connected. This report provides a method for integrating these agendas under the accountability framework, as part of a broader effort to improve local governance in Bank's client countries. It describes the main components of such a local governance strategy, identifies the research gaps that must be filled to fully develop this strategy, and discusses the operational implications of merging decentralization and accountability agendas. In broad terms, a local governance strategy should provide support to all stages of public policy making in client countries. United Nation Capital Development Fund's Practitioner's Guide (UNCDF 2005) provides guidance for building such a strategy. Following similar lines, the report recommends that the Bank's assistance focus on four areas: (i) National policy debate: The Bank needs to be involved in national level policy discussions that establish macro policy objectives. The Bank can help client countries in drafting cabinet papers and white papers to set the macro policy guidelines. More importantly, the Bank can help national policy makers think about difficult policy issues during this process and make sure that there no inconsistencies across different policy documents. (ii) Support for legislation and the statutory framework: Once the general direction of decentralization policy is established, it should be implemented through legislation (i.e., a decentralization bill). The Bank should provide support to clients in drafting legislation that is carefully drafted and consistent with other standing legislations. (iii) Support for the regulatory framework: The implementation of a decentralization strategy requires the enactment of secondary legislation (bylaws, ministerial decrees, circulars, and so on). The Bank should engage with client countries during this process to ensure consistency and timely issuance of this legislation. (iv) Support for establishing systems, procedures, guidelines, and practices: In the implementation of a decentralization reform strategy, it is also important to establish financial and administrative systems, procedures, and guidelines (i.e., budget planning guidelines, service delivery norms, and so on). The Bank should provide support to client countries in developing these systems, procedures, and guidelines. 35According to the Independent Evaluation Group's review of operational database, between 1987 and December 2006, the Bank committed about US$32 billion to about 89 countries through 458 programs, projects, and grants in which decentralization was noted as one of the key themes or classified as an activity. 65 The key features of Bank's local governance strategy should be: Policy Pilots Closely monitored pilots should, whenever possible, precede the full-scale introduction of new policies and delivery mechanisms. Phased introductions help not only to inform implementation, but also identify and prevent unintended consequences. A pilot is an important first stage of regular, longer-term policy monitoring and evaluation. Therefore it is important to pilot activities that can guide the policy design process. Advantage should be taken of the small-scale and explicitly experimental nature of pilots to encourage policy innovations that might otherwise be too risky or costly to embark on. If successful, the pilots can be scaled up. The impact of such pilots helps to improve certain policies, to make adjustments of policies or processes and to sometimes to totally redesign the program. Policy Innovations A widely acknowledged by-product of pilots and policy trials is their role in encouraging and facilitating innovation. It is much easier for a client government to contemplate an untested new policy or delivery method if it is packaged as an experiment. This applies especially to small changes in policy or process where experimentation--perhaps with one or several alternative approaches--is clearly the most rational option. The fact that pilot programs help to reduce the risk of expensive failures frees governments to be more courageous in considering options they might otherwise eschew. Scaling-up and Policy Impact Strategy It is also important for the Bank to develop a policy impact strategy and replication in local governance. The strategy should involve using various Bank instruments (such as development policy lending, learning and innovations loans, and technical assistance loans) to inform the client countries how to implement different components of the local governance reforms. Even in countries where there is no commitment to decentralization, the strategy can demonstrate that there are gains from implementing the suggestions from table 4.1. Local Institutional Innovation The Bank's local governance strategy should be flexible enough so that there is a room for local institutional innovations. Specific form of innovation might vary across countries, but these innovations should cover rules, procedures, and organizational structures. For example, local institutional innovation should seek ways to integrate traditional authorities into local governance. Establishing Historical Benchmarks for the Evolution of Local Governance Structures The nation state in the West developed through a bottom-up process. The political process was not easy and was often turbulent, but it was a process that was significantly influenced by local systems of governance established by local communities. Many developing countries, especially the post-colonial states in Africa and South Asia, inherited the trappings of a nation built on a centralized colonial apparatus--a top-down process. Therefore, a highly centralized system of governance evolved in those countries; political systems favor central politicians and central bureaucracy over systems of external accountability that would ensure that policy makers and service providers were responding directly to citizens. Even in countries where there were no post- colonial structures, central authorities often capture local governments. The formation of the nation state in developed countries could be used as a historical benchmark to compare the local government evolution processes in developing countries. Historical analysis of local governments will provide benchmarks for the evolution of local governance structures in the developed world. It will provide a 66 conceptual framework for thinking about local democratic governance36 in developing countries. The framework will help to capture various important characteristics of strengthening and institutionalizing an effective local government system. Building the Knowledge Base It is very important to systematically collect information about different dimensions of a local governance structure in a country. Building a knowledge base facilitates better decision making. Capacity Building Building the capacity of various stakeholders' is an important component of a local governance strategy. Capacity building should take place at the individual or group levels (covering skills and knowledge requirements), at the institutional or organizational levels (covering operational and administrative aspects), and at the strategic or systemic level (covering legal, political, and economic frameworks). The following sections present the analytical and operational areas that World Bank projects could focus on in line with the abovementioned components of a local governance strategy. 4.2.2. Investing in Analytical Products and Tools The World Bank should invest in analytical products to inform country-specific conditions, not only for operational projects, but also country assistance strategies, PRSPs, and sectoral strategies. These investments will provide country management units, regions, and, more importantly, client countries with the knowledge to enable them to make informed policy choices. Furthermore, they will help the Bank to design better country assistance strategies and public policy advocacy. A number of country-specific analytical and diagnostic studies pertinent to decentralization and accountability warrant attention to improve the analysis underpinning the Bank's work in decentralization and local governance. These studies should focus on four key areas: 1. Actors, powers, and accountability relationships: The pre-existing local political environment can have a significant impact on the outcome of decentralization projects (Hiskey 2006). Research on the impact of the local political environment on the subsequent outcome of decentralization reforms shows that designers of decentralization reforms should invest time and resources to understand a country's political dynamics to determine how to sequence demand-side reforms to empower citizens (Hiskey 2006). In their study of demand-based National Solidarity Program (PRONASOL) of Oaxaca State in Mexico, Fox and Aranda conclude "outcome was highly contingent on whether local governments are already democratic and responsive to their citizens. Where these prior conditions do not hold, however, decentralization could actually reinforce authoritarian rule at the local level" (as cited in Hiskey 2006: 11). According to Hiskey, to avoid such a situation, "designers of decentralization reforms must closely examine the political environment of the target population and determine how best to empower citizen/principals of that community rather than the elite/agents" (ibid: 11). 2. Assessing economic and social inequalities: There is a growing literature on the relationship between socioeconomic inequalities and elite capture that stresses the importance of designing decentralization reforms that take these inequalities into account. Bardhan argues that "[t]he extent of capture of local governments by elites depends on levels of social and economic inequality within communities, traditions of 36 Local democratic governance is a general rubric for the array of institutions, organizations, rules, laws, and regulations that interact to realize decentralization's benefits. 67 political participation and voter awareness, fairness and regularity of elections, transparency in local decision- making processes and government accounts, media attention and other factors" (Bardhan 2002: 194). In a given country there are different communities with varying degrees of socioeconomic power. Theoretically, decentralization empowers all principals in overseeing agents. However, the extent of "economic and social inequalities within a community may dramatically affect the ability of all principals to gain equal access to the information gathering and agent monitoring mechanisms created by decentralization" (Hiskey 2006: 14). Therefore, it is important to assess a country's socioeconomic characteristics prior to designing a decentralization strategy. If the country "is relatively equal in its human capital and economic resources [in different regions], the danger of creating a privileged set of principals is lessened' (ibid: 14). Otherwise, it would be more appropriate before decentralization to sequence the reforms that would first address these inequalities and allow the underprivileged more mechanisms of oversight and involvement in decision-making processes. 3. Local electoral structure and the party system: The structure of local government electoral systems (rules of candidacy, suffrage, rounds, proportionality, term lengths) and processes (competitiveness of elections, party structure, financing, and so on) shape the degree of accountability of local governments (councils, mayors, special districts, and so on) to citizens. According to Hiskey, "[g]reater and more meaningful political competition for local political offices, then, should have a positive impact on reforms that expand the decision-making powers of local agents. Where citizens have a means of control over elected officials through the presence of viable electoral alternatives, the increased level of uncertainty among local political elites should enhance the responsiveness of these elites to the concerns of the citizenry" (ibid: 12). Increased political competition allows for more effective preference matching and increases chances for widespread acceptance of election results, which, in return, decreases possibility for post-electoral instability (Hiskey 2006: 13). Thus, it is particularly important to gather empirical data on the relation between the structure of local electoral systems and the accountability of elected local authorities. There are important question to be probed: How does the universal suffrage shape local government responsiveness in general and to the poorest segments, in particular? Are there comparative studies on universal suffrage versus systems that exclude migrants? Does the exclusion of migrants deepen identity-based conflict? How do political processes around the choice of candidates and elections unfold in practice? How does the choice of candidates shape the accountability of local elected authorities? Is the institution that is charged with election oversight nonpartisan? 4. The drivers of locally elected leader accountability and performance: There are many factors that drive the performance of locally elected leaders, including accountability. It is important to understand accountability in the context of the range of factors that shape local government performance. For example, contributing factors may include the enabling environment of national rules and regulations; the organizational environment; donor projects; the ability of councilors to mobilize government; the ability to mobilize local revenues and resources; the powers and skills of councilors; the ideologies of councilors, the fear of sanctions, and cultural norms and values. One factor identified in the literature, for example, is capabilities. Teskey (2005: 5) defines organizational development as the ability to work as a unit. He defines institutional development as having favorable rules of the game. Individuals have skills and competencies. Organizations have capacity. They have capabilities to do specific things (manage people, conduct research, manage funds).37 Teskey (ibid) argues that these are 37See Teskey (2005:11) where he summarizes definitions of human capacity (skills), organizational capacity (collective action), and institutional capacity (rules of the game). 68 the building blocks of organization's capacity to perform. Organizations operate in a wider institutional environment that can support or undermine their ability to achieve their collective objectives. He argues that organizational reform must be distinguished and approached differently from institutional reform. Institutional reform is much more difficult, but without it, organizational reform is not likely to work or be sustainable. 4.2.3. Challenges for Lending Operations and Technical Assistance The Bank's lending instruments (investment loans and development policy loans) provide financing for a wide range of activities aimed at creating physical and social infrastructure. Some of these instruments aim to operationalize local governance support. However, portfolio review of Bank project documents suggests that many lending projects fall short of establishing good local governance institutions, including both the supply and demand side. For example, in an ongoing study, the Social Development Department's Community Driven Development and Local Governance Team conducted a portfolio review of urban projects to analyze whether or not these projects had a clear focus on both supply- and demand-side institutions of local governance (Herzog, 2007).38 The preliminary findings of the 110 urban operations from FY00-07 suggest that urban governance is an increasingly important project design feature. Interestingly, this increase is largely associated with a sharp growth of the demand-side of urban governance aspects in lending, from 40 percent in FY00­01 to more than 60 percent in FY06­07. On the supply side, urban planning and service delivery are the most prevalent project components; attention to financial and information management has been growing more recently, while concerns about local government auditing and monitoring and evaluation are still limited and inconsistent. There are also clear trends on the demand side. From the beginning, access to information and voice in decision making were important project design components, while attention to public oversight over budget execution and participatory monitoring and evaluation of service delivery has increased more recently. Yet, despite increasing level of attention to demand-side governance institutions, a key result of this review is that policy reforms and institutional development in Bank urban projects have mainly focused on the supply side of governance. The attention to the demand side of urban governance is limited to piloting, with limited impact on institution building and policy setting. Here are the lessons learned from these projects: In the design stage, many physical infrastructure projects that finance local roads, water supply systems, local schools and clinics, and so on, do not develop local governance institutions--i.e., investment choices may not match local preferences, there is a lack of local ownership. Many projects did not provide enough support to local governments as agents of change. Some established parallel structures to channel funds. Many projects focused heavily on supply-side institutions and did not strengthen demand-side institutions. Many projects did not design strong monitoring and evaluation systems to measure developmental impact. Despite such common deficiencies, a growing number of lending projects try to develop a more comprehensive local governance approach. The following boxes (boxes 4.1. and 4.2) demonstrate two examples as such that aim to bridge supply- and demand-side approaches with a monitoring and evaluation component. 38The review covered 110 operations, including 80 mapped to Urban Development and an additional 30 operations in urban areas mapped to CDD, which were used as control. 69 Box 4.1. Examples of operational entry points: local governance support project in Bangladesh and demand for good governance project in Cambodia Local governance support project in Bangladesh The Government of Bangladesh recently took a number of steps to strengthen local governance, building on the long history of elected Union Parishads (UPs)--lowest tier of local government--and more recently conducted some innovative pilot projects. The government introduced the direct transfer of untied block grant to UPs and revised local revenue generation procedures. Although small, the block grant established the important principles of local discretion and accountability, with community oversight. In this context, the Local Governance Support Project assists the government to scale up and institutionalize the system of fiscal support and mechanisms of accountability for UPs. The LGSP development objective is to develop accountable local governments (LGs) that provide services that meet community priorities, supported by predictable and transparent fiscal transfer systems. The project aims to institutionalize accountability into existing systems based on incentives associated with the expanded block grant and complemented by mandatory disclosure by UPs to both communities and to higher tiers of government. Regular open meetings, public disclosure, and regular reporting will be at the core of the local participatory process and are the conditions for block grant release. Under the project, UPs submit quarterly reports to Local Government Division (LGD) and communities, and the government carries out annual financial, procurement, environmental, and social assurance audits for each UP to become or remain eligible for the expanded block grant. The Audit Quality Assurance Cell of the Comptroller & Auditor General verifies these audits on a sample basis. The project also provides technical and financial support to upper- tier monitoring agencies and capacity building at all levels to ensure adequate tracking of project inputs and outputs. The project is supported by a proposed competitive grant facility that will assist civil society in undertaking activities that support the inclusion of marginalized communities in UP decision making. Specific requirements for public disclosure are indicated in the audit strategy and monitoring procedures. In particular, three major elements of participation and disclosure are required in the project: open meetings; UP-level notice boards; and signboards displaying information of each infrastructure scheme. A block grant coordination committee at the upazila level ensures coordination, peer learning, and Upazila Nirbhai Officers (UNO) oversight, and provides an avenue for grievance redress. Citizens have three avenues to seek redress, in case all safeguards and checks built in to ensure accountability fail: (i) open ward or UP meetings; (ii) a meeting of the block grant coordinating committees, which will be held bi-annually in different UPs on a rotational basis; and (iii) a complaint to UNO, Upazila Development Coordination Committee (UDCC), or LGD. The UPs need support to undertake participatory strategic planning, design and manage project implementation, and monitor and evaluate project outcomes. The project focuses primarily on informing the general public and UPs about the conditions of access to and use of block grants, and also supplies quality core training around the local Public Expenditure Management cycle. In addition, peer-learning activities across participating local governments are implemented. Finally, the LGSP also directly strengthens the capacity of government institutions--such as LGD, the Comptroller & Auditor General, National Institute of Local Governments (NILG), and UNOs--to implement the reform process. Demand for Good Governance Project (DGGP) in Cambodia: Conceptually, DFGG refers to the extent and ability of citizens, civil society organizations, and other non-state actors to hold the state accountable and to make it responsive to their needs. And in return, this DFGG enhances the capacity of the state to become transparent, accountable, and participatory in order to respond to these demands. Given this operational definition, the proposed DGGP approaches strengthening of DFGG as a process, with four key elements: 1. Promotion of Demand: The ability of citizens, civil society, and other nonstate actors to demand better governance depends on their access to information and the degree to which they can act effectively on this information. Thus, DFGG depends on (a) disclosure of information: the level of government transparency (regarding budgets, expenditures, programs, and so on), (b) demystification of information: strengthening the level of citizen awareness and understanding (on laws, rights, budgets, policies, and so on), and (c) information dissemination: spreading information (related to governance issues, processes, finances, laws, and so on) to the ordinary public. In promoting demand, the three Ds should be followed by a C (collective action)--mobilizing broader action and advocacy around this information. Accordingly, initiatives such as freedom of information, awareness campaigns, rights education, and media programs that promotedemand are the first component of DFGG project support. 70 2. Mediation of Demand: While mobilizing demand through advocacy and information dissemination are important, these pressures only become effectivethrough mediation and institutionalized feedback to the state. Thus, strengthening DFGG also involves: (a) creation and strengthening of feedback avenues for citizens and civil society to public officials; (b) encouraging consultation of these actors in decision making and public actions of executive agencies; and, (c) formal and informal dispute resolution mechanisms through initiatives such as interface meetings between citizens and public officials, grievance redress mechanisms, and ombudsman's offices. 3. Response to Demand: After mediation comes response, without which demand alone would remain powerless and not credible. Thus, institutionalizing DFGG involves developing programs and initiatives within the executive that respond to demand either through (a) innovations in service delivery, (b) response-based performance incentives, or (c) participatory action planning. 4. Monitoring to Inform Demand: Finally, the last element of increasing DFGG is the process of public sector monitoring and oversight by nonexecutive actors such as the parliament, media, and civil society. This could take the form of: (a) participatory monitoring (using citizen feedback surveys of government performance, social audits, media investigations, and so on); (b) independent budget and policy analysis; and, (c) formal oversight mechanisms (parliamentary committees, vigilance commissions, and so on). These generate key information for the executive, civil society, and ordinary citizens that feeds back into the process of promoting DFGG--thus completing the cycle. It is important to note that the above definition of DFGG has a key implication--it means that the institutions that strengthen DFGG can be both government and nonstate institutions. A state-run broadcasting corporation involved with disseminating information about public programs and their budgets, and providing feedback of citizens to public officials is as much a demand- side actor as civil society and the private media promoting demand. Likewise, ombudsman offices and vigilance commissions, or ministries running programs for legal and rights awareness education are also demand-side actors. Moreover, the DFGG response function is something that is most often only provided through the state executive. What matters for strengthening DFGG under this project is therefore what an institution does, rather than where it is situated. Sources: World Bank. 2006b. Bangladesh Local Governance Support Project Appraisal Document; World Bank 2005; World Bank. 2007b. Cambodia Demand for Good Governance Project Information Document. 4.3 FURTHER CRITICAL ISSUES TO ADDRESS FOR ACHIEVING GOOD LOCAL GOVERNANCE Achieving good local governance requires major effort both from the country and the World Bank. Success of these efforts relies on several other critical external factors that include: Political Will In many countries, decentralization reforms are poorly designed and implemented not due to ignorance or lack of experience but due to a lack of political will. Politicians at the higher levels of government are inherently unwilling to devolve power. This kind of unwillingness often results in discrepancies between de jure versus de facto decentralization. Such discrepancy implies that there is a huge gap in implementation and that the practice of decentralization faces resistance, resulting in gaps between what the laws and regulations say and what the practice is on the ground. Local Context Decentralization reforms are ultimately shaped by the country contexts, which widely differ from one country to another. The historical and social context in which decentralization decisions are made, demographic structure of the society, and macro and local economic conditions are all important determinants of the outcome of decentralization reforms. Given differing country-specific social and economic settings, outcomes of decentralization reforms are also shaped by existing power relationships among actors both within and outside public sector. In this regard, understanding the country context and identifying the potential impacts of decentralization reforms on each relevant stakeholder within a country are crucial for achieving the expected outcomes through these reforms. 71 Donors In most decentralizing countries, donors play an important role in setting the policy agenda and reform implementation. In this supply-driven process, prioritization and sequencing are done according to what donors think is important. More importantly, in many cases donors establish parallel structures to bypass formal public sector institutions due to various reasons related to the desire for quick results and a lack of strong commitment to the country's long-term institutional development. Capacity The capacity of both central and local governments has an important bearing on decentralization outcomes. Therefore, an effective decentralization strategy should a component for building the capacities of both central and local governments. The issue of lack of capacity is equally important for demand-side initiatives. More often than not, citizens and all associated organizations that focus on citizen action lack capacity and resources for a long-term commitment to improving local governance. Coordination Decentralization brings a shift in the assignment of responsibilities and therefore a shift in power sharing that implies fundamental changes in the administrative system. Such major shifts create uncertainties in the transition process. This can lead to different stakeholders resisting change and not following the legal requirements, thus slowing down the transition process. Establishing an effective coordination mechanism between stakeholders is an important component of decentralization. Comprehension Many public services are credence goods. It is very difficult for citizens to make a judgment about the quality of these services. Increased reliance on only popular opinion may actually result in reduced expenditures on certain key services that may not be perceived as important by ordinary citizens but are crucial for social economic development in the long run. It is also important to note that citizens uniformly cannot possibly have the same level of comprehension about the local development issues they are facing. This kind of asymmetric comprehension among citizens about their role vis-à-vis government may end up increasing the influence of certain groups, i.e., of those educated, over local government decisions. Sustainability A key aspect of any decentralization reform is its sustainability. This is closely related to the challenges mentioned above, such as the existence of political will, developing a local governance strategy that represents differing views of local stakeholders, and dealing with capacity problems of both state and nonstate actors. Having a thorough analysis of whether or not institutions of decentralization reforms can be sustained requires galvanizing enough support from local stakeholders, building their capacities, and committing a long-term and coordinated donor support. 4.4 CONCLUSION The concept of governance has long been associated with central government. However, in the last two decades, there have been dramatic changes in public sector management and thereby in the definition of the concept of governance. With the acceleration of efforts to move decision-making power towards lower levels, decentralization and local governance have become an integral component of the broader good governance framework. This study intends to provide an analytical framework and highlights operational challenges in establishing a good local governance system in the decentralization context. It aims at first analyzing the devolution of power to lower levels in terms of its impact on local discretion--from the perspectives of 72 decentralization's political, administrative, and financial dimensions. It also provides a discussion on strengthening downward accountability of local decisionmakers by bridging approaches that have traditionally grown in parallel sometimes-conflicting tracks-- public sector and social accountability. In the public-sector approach, the emphasis is on strengthening supply-side institutions; there is an indirect reliance on pressures from citizens. On the other hand, social accountability approaches directly stress the importance of demand-side institutions--i.e., citizen involvement and participation/civic engagement. This study argues that these two complementary approaches can and should work together. Yet when it comes to practice, the challenges and risks are not minor. There are limited examples of decentralization reforms that successfully link discretion and accountability and bridge supply- and demand-side approaches. The challenges and risks are associated with adverse local country contexts and power relationships, lack of political will and ownership, weak capacity of both state and nonstate actors, and sustainability problems. This study's goal is to at least open the discussion on both the analytical and operational aspects of local discretion and accountability in the hope that this discussion will contribute to efforts to develop local governance strategies both in client countries and within the bank. 73 References ACIR (Advisory Commission on Intergovernmental Relations). 1974. "Governmental Functions and Processes: Local and Areawide." ACIR, A-45, Washington DC. Ackermann, John. 2003. "Co-governance for Accountability: Beyond `Exit' and `Voice'." World Development 23 (3): 447­63. ADB [Asian Development Bank]/DfID [Deprtment for International Development]/WB [World Bank]. 2004. Devolution in Pakistan: An Assessment and Recommendations for Action. http://siteresources.worldbank.org/EXTSAREGTOPDECENTRALIZATION/Resources/496899- 1095189822590/521764-1096384315479/PakDevolutionMain.pdf. Agrawal, A., and J. Ribot. 1999. "Accountability in Decentralization: A Framework with South Asian and African Cases." Journal of Developing Areas 33 (Summer): 473­502. Ahmad, Ehtisham, Maria Albino-War, and Raju Singh. 2006. "Sub-national Public Financial Management: Institutions and Macroeconomic Considerations" in Handbook of Fiscal Federalism, ed. Ehtisham Ahmad and Giorgio Brosio. Massachusetts: Edward Elgar. Ahmad, Junaid, Shantayanan Devarajan, Stuti Khemani, and Shekhar Shah. 2005. "Decentralization and Service Delivery." World Bank Policy Research Working Paper 3603, World Bank, Washington, DC. Alonzo, R. 2003. "The Political Economy of Decentralization in the Philippines." Washington, DC: World Bank, East Asia and Pacific Region. Andrews, Matthew and Shah, Anwar. 2005. "Citizen-Centered Governance: A New Approach to Public Sector Reform" in Anwar Shah (ed) Public Expenditure Analysis, World Bank: Washington, D.C. Arrow, K. J. 1985. "The Economics of Agency" in J. W. Pratt and R. J. Zeckhauser (eds.) Principals and Agents: the Structure of Business, Harvard Business School Press: Boston. Aucoin, P., and R. Heintzman. 2000. "The Dialectics of Accountability for Performance in Public Management Reform." International Review of Administrative Sciences 66: 45­55. Azfar, Omar, Satu Kähkönen, and Patrick Meagher. 2001. "Conditions for Effective Decentralized Governance: A Synthesis of Research Findings." Working Paper No. 256, Center for Institutional Reform and the Informal Sector, University of Maryland, College Park, Maryland. Bahl, R., and J. Martinez-Vazquez. 2006. "Sequencing Fiscal Decentralization." World Bank Policy Research Working Paper No. 3914, World Bank, Washington, DC. Bahl, Roy W. 1999. "Fiscal Decentralization as Development Policy." Public Budgeting & Finance, 19: 59­75. http://ssrn.com/abstract=468271. Bahl, Roy W. and J. F. Linn. 1992. Urban Public Finance in Developing Countries. New York: Oxford University Press. Bahl, Roy, and L. Schroeder. 1983. "Intergovernmental Fiscal Relations" in R. Bahl and B. Miller (eds.) Local Government Finance in the Third World: A Case Study of the Phillipines. New York: Praeger. Baltaci, Mustafa, and Serdar Yilmaz. 2006. "Keeping an Eye on Subnational Governments. Internal Control and Audit at Local Levels." World Bank Institute Working Paper Series No. 37257, The World Bank, Washington, D.C. 74 Bardhan, Pranab. 2002. "Decentralization of Governance and Development." Journal of Economic Perspectives, 16(4): 185­205. Behn, Robert, 2001. Rethinking Democratic Accountability. Brookings Institution Press: Washington, DC Bird, Richard. 2000. "Intergovernmental Fiscal Relations in Latin America: Policy Design and Policy Outcomes." Inter-American Development Bank, Washington, DC. http://www.iadb.org/sds/doc/Intergovernmentaltotal.pdf. Bird, R. 2000. "Intergovernmental Fiscal Relations: Universal Principles, Local Applications" Working Paper 00-02, Andrew Young School of Policy Studies, Georgia State University, Atlanta. Bird, Richard, and F. Vaillancourt. 1998. "Fiscal Decentralization in Developing Countries: An Overview." In Fiscal Decentralization in Developing Countries, ed. R. Bird and F. Vaillancourt. Cambridge: Cambridge University Press. Blair, H. 2000. "Participation and Accountability at the Periphery: Democratic Local Governance in Six Countries." World Development 28(1): 21­39. Bland, Gary. 2006. "Election and the Development of Democratic Local Governance." RTI International, Washington, DC. Bossert, Thomas J., and Joel C. Beauvais. 2002. "Decentralization of Health Systems in Ghana, Zambia, Uganda, and the Philippines: A Comparative Analysis of Decision Space." Health Policy and Planning 17 (1): 14­31. Bovens, Mark. 2006. "Analyzing and Assessing Public Accountability: A Conceptual Framework" European Governance Papers (EUROGOV) No. C-06-01. http://www.connex- network.org/eurogiv/pdf/egp-connex-C-06-01.pdf. Bovens, Mark. 2005. "Public Accountability." In The Oxford Handbook of Public Management, ed. Ewan Ferlie, Laurence E. Lynn, and Christopher Pollitt. New York: Oxford University Press. Brennan, Paul J., and Donald C. Miller. (n.d.). "Recommendations for Local Government Procurement Reform in New York State." New York State Association of Municipal Purchasing Officials, Inc., SAMPO NIGP Chapter, New York. http://www.thepartnership.org/files/pdfs/ECSP/nysampo.pdf Brinkerhoff, Derick W. 2004. "Accountability and Health Systems: Toward Conceptual Clarity and Policy Relevance" Health Policy and Planning 19 (6): 371­79. Campos, J. E., and J. S. Hellman. 2005. "Governance Gone Local: Does Decentralization Improve Accountability?" In East Asia Decentralizes: Making Local Government Work, 237­52. Washington, DC: World Bank. Crook, R., and J. Manor. 1998. Democracy and Decentralization in South-East Asia and West Africa: Participation, Accountability, and Performance. Massachusetts: Cambridge University Press. de Rezio, Paolo, Vitus Aseem and Veivek Ramkumar (2006). Case Study on the Uganda Debt Network . International Budget Project, Washington, D.C. Devas, Nick, and Ursula Grant. 2003. "Local Government Decision-Making--Citizen Participation and Local Accountability: Some Evidence from Kenya and Uganda." Public Administration and Development. 23 (4): 307­16. Eaton, Kent. 2001. "Political Obstacles to Decentralization: Evidence from Argentina and Philippines." Development and Change. Oxford: Blackwell Publishers. 75 Emboltura, A. 1994. "Recentralization or Suspension: A Step in the Right Direction?." Legislative Alert (July-September):23-27. Estrella, Marisol and Nina Iszatt. 2004. Beyond Good Governance, Participatory Democracy in the Philippines. Institute for Popular Democracy, Quezon City, Philippines. Available at http://www.ipd.ph/BBGC/frontCover.html Evans, Anne. 2004. "A Framework for Decentralizing Civil Servants." PREM Workshop on Decentralizing Civil Servants, World Bank, Washington, DC. Francis, Paul, and Robert James. 2003. "Balancing Rural Poverty Reduction and Citizen Participation: The Contradictions of Uganda's Decentralization Program." World Development 31(2): 325­37. Fung, Archon. 2004. Empowered Participation. Reinventing Urban Democracy. New Jersey: Princeton University. Fung, Archon. 2003. "Deliberative Democracy, Chicago Style: Grassroots Governance in Policing and Public Education." In Archon Fung and Eric Olin Wright (eds.) Deepening Democracy Institutional Innovations in Empowered Participatory Democracy. Verso Books, New York, pp. 3-31. Fung, Archon, and Erik Olin Wright. 2001. "Deepening Democracy: Innovations in Empowered Participatory Governance." Politics & Society 29 (1): 5­41. Gaventa, John. 2006. "Triumph, Deficit of Contestation? Deepening the `Deepening Democracy' Debate." IDS Working Paper 264, Sussex, UK. Goetz, Anne Marie, and Rob Jenkins. 2005. Reinventing Accountability: Making Democracy Work for Human Development. London: Palgrave Macmillan. Goetz, Anne Marie, Anuradha Joshi, and Mick Moore. 2004. "Diversity, Accountabilities and Service Delivery." Paper presented at the meeting on Politics and Service Delivery organized by DFID, London, March 30, 2004. Grun, S. (2000). Evaluacion del Componente Desarrollo Municipal del FISE. Managua. Guggenheim, S. E. 2006. "Crisis and Contradictions: Understanding the Origins of a Community Development Project in Indonesia." In The Search for Empowerment. Social Capital as Idea and Practice at the World Bank, ed. Bebbington et al. Bloomfield, Indiana: Kumarian Press. Guiza, E. 2002. "Transforming a Ravaged City into a Model in Local Revenue Generation: The Case of San Fernando, Pampanga. Philippines." Asian Institute of Management, Philippines. Herzog, Andre. 2007. "Urban Governance Portfolio Review Dissemination Paper." Mimeo. Helling, Louis, Rodrigo Serrano, and David Warren. 2005. "Linking Community Empowerment, Decentralized Governance, and Public Service Provision through a Local Development Framework." Social Protection Discussion Paper 535, World Bank, Washington, DC. Hiskey, Jonathan T. 2006. "Principals, Agents, and Decentralized Democratic Development: A Conceptual Framework for Democratic Local Governance." USAID (United States Agency for International Development) and the Urban Institute, Washington, DC. www.ddlg-at- work.com/UI%20Concept%20Paper%20Hiskey_Final2%2010%2027.pdf. IDEA (International Institute for Democracy and Electoral Assistance). 2004. "Leadership for Local Democracy: Local Democracy Assessment Guide." http://www.idea.int/dll/upload/Local_Dem_Assessment_Guide.pdf. Jensen, M.C. and Meckling, W. H. 1976. "Theory of the Firm: Managerial Behaviour, Agency Costs and Ownership Structure" Journal of Financial Economics, vol. 3. pp. 305-360. 76 Jeppsson, Anders. 2001. "Financial Priorities under Decentralization in Uganda." Health Policy and Planning 16 (2): 187­92. Kampala City Council and Uganda Management Institute (2005). The First Kampala Citizens' Report Card: Measuring Citizens' Satisfaction with Key Public Services. Kampala. Khemani, Stuti. 2007. "Does Delegation of Fiscal Policy to an Independent Agency Make a Difference? Evidence from Intergovernmental Transfers in India." Journal of Development Economics 82(2): 464­84. Khemani, Stuti. 2004. "Local Government Accountability for Service Delivery in Nigeria." Background paper for the World Development Report 2004. World Bank, Washington, DC. http://www.worldbank.org/research/wdr/WDR2004/papers/stuti_nigeria.pdf King, R. et al. 2003. Tracking of the District Assemblies Common Fund: A Pilot Study of Four District Assemblies. Accra: ISODEC. http://www.isodec.org.gh/Papers/monitoringDACF.pdf Kolehmainen-Aitken, Riitta-Liisa. 2004. "Decentralization's Impact on the Health Workforce: Perspectives of Managers, Workers and National Leaders." Human Resources for Health 2004 2:5. http://www.human-resources-health.com/content/2/1/5. Laffont, J. J. 1989. The Economics of Uncertainty and Information. MIT Press: Cambridge. Levy, Brian. 2006. "Governance, Corruption and Social Accountability" November 28, 2006. Presentation to the TFESSD Consultations, World Bank, Washington, DC. Livingstone, Ian, and Roger Charlton. 2001. "Financing Decentralized Development in a Low-Income Country: Raising Revenues for Local Government in Uganda." Development and Change 32: 77­ 100. Malena, Carmen, Reiner Forster, and Janmejay Singh. 2004. "Social Acccountability, An Introduction to he Concept and Emerging Practice." Social Development Paper No 76, World Bank, Washington, DC. Manin, Bernard, Adam Przeworski, and Susan Stokes. 1999. "Elections and Representation." In Democracy, Accountability, and Representation, ed. Bernard Manin, Adam Przeworski, and Susan Stokes. Massachusetts: Cambridge University Press. Mawhood, P. 1983. Local Government in the Third World. Chichester, UK: John Wiley. Mandondo, A. 2000. "Situating Zimbabwe's Natural Resource Governance Systems in History." CIFOR (Center for International Forestry Research) Occasional Paper No. 32, December 2000, CIFOR, Jakarta. Matthew, George. 2003. "The Crisis of Governance in South Asia: Assessing Decentralization's Promise in India." South Asia Partnership Canada Seminar Paper, Ottawa. http://action.web.ca/home/sap/south_asia_resources.shtml?x=48519&AA_EX_Session=29c8f438695 80c02e3e868b529e8ffed. McCubbins, Mathew D. and Thomas Schwartz. 1984. "Congressional Oversight Overlooked: Police Patrols versus Fire Alarms." American Journal of Political Science 28, no. 1: 165-179. McLean, Keith, Louis Helling, Jana Orac, and Rodrigo Serrano. 2006. "Exploring Partnerships between Communities and Local Governments in Community Driven Development: A Framework." Social Development Paper No. 96, World Bank, Washington, DC. 77 Mora, Mônica, and R. Varsano. 2001. "Fiscal Decentralization and Subnational Fiscal Autonomy in Brazil: Some Facts of the Nineties." Working Paper, Institute of Applied Economic Research, Rio de Janeiro, Brazil. http://www.ipea.gov.br/pub/td/td_2001/td_0854.pdf Mulgan, Richard. 2000. "Accountability: An Ever-Expanding Concept?" Public Administration 78 (3): 555­73. Musgrave, R. 1959. The Theory of Public Finance. New York: McGraw Hill. Narayan, Deepa, ed. 2002. Empowerment and Poverty Reduction: A Sourcebook. Washington, DC: World Bank. Oates, W. 1972. Fiscal Federalism. New York: Harcourt, Brace and Jovanovic. Olowu, D. 2001. "Decentralization Policies and Practices Under Structural Adjustment and Democratization in Africa." Democracy, Governance and Human Rights Programme Paper No. 4. United Nations Research Institute for Social Development, Geneva. Paul, Samuel. 2002. Holding the State to Account: Citizen Monitoring in Action. Bangalore: Books for Change. Paul, Samuel. 1992. "Accountability in Public Services: Exit, Voice and Control." World Development 20 (7): 1047­1060. PEFA (Public Expenditure and Financial Accountability). 2005. Public Financial Management Performance Measurement Framework. PEFA Secretariat. Available at http://www.pefa.org/PEFA%20Website%20--%20CURRENT%204-9- 03/www.pefa.org%20WEBSITE/about_test.htm. Piven, Frances Fox, and Richard Cloward. 1979. Poor People's Movements. New York: Vintage Press. Pollitt, C. 2003. The Essential Public Manager. London: Open University Press/McGraw-Hill. Rakodi C. 2002. Influence and Accountability, One World Action, London, UK. Rees, R. 1985. "The Theory of Principal and Agent" Bulletin of Economic Research, pp. 3-26. Reinikka, Ritva and Jakob Svensson. 2003. "The Power of Information: Evidence from a Newspaper Campaign To Reduce Capture." The World Bank Policy Research Working Paper Series No. 3239. World Bank, Washington, D.C. Ribot, J. C. 2004. Waiting for Democracy: The Politics of Choice in Natural Resource Decentralization. Washington, DC: World Resources Institute. Ribot, J. C. 2003. "Democratic Decentralization of Natural Resources: Institutional Choice and Discretionary Power Transfers in Sub-Saharan Africa." Public Administration and Development 23 (1): 53­65. Ribot, J. C. 1996. "Participation without Representation: Chiefs, Councils and Forestry Law in the West African Sahel." Cultural Survival Quarterly Fall 1996: 40­44. Robinson, Mark. 2006. "Budget Analysis and Policy Advocacy: The Role of Non-Governmental Public Action." IDS Working Paper 279, Sussex, UK. Rodden, J. 2002. "The Dilemma of Fiscal Federalism: Intergovernmental Grants and Fiscal Performance around the World." American Journal of Political Science 46 (3): 670­87. Romeo, L. 1996. "Local Development Funds: Promoting Decentralized Planning and Financing of Rural Development." United Nations Capital Development Fund, Policy Series. United Nations, New York. 78 Sahgal, Vinod, and Deepa Chakrapani. 2000 "Clean Government and Public Financial Accountability." OED Working Paper Series No.17, World Bank, Washington, D.C. Saito, Fumihiko. 2000. "Decentralization in Uganda: Towards a Positive-Sum Solution," Ryukoku University, Japan. Available at www.world.ryokoku.ac.jp/~fumisait/en/publication.html Schaeffer, M. 2005. "Public Finance, Transparency and Accountability: A Path toward Reducing Corruption." Mimeo. Schneider, Aaron and Marcelo Baquero. 2006. "Get What You Want, Give What You Can: Embedded Public finance in Porto Alegre." International Development Studies Working Paper Series No. 266, Sussex. Schroeder, Larry. 2004. "Mechanisms for Strengthening Local Accountability." Paper prepared for the Decentralization and Intergovernmental Relations Course, World Bank, March 29-31, 2004. http://www1.worldbank.org/publicsector/decentralization/March2004Course/Schroeder.doc. Serrano, Rodrigo. 2006. "Social Accountability and Local Governments." Paper prepared for Social Accountability Sourcebook (forthcoming). World Bank, Washington, DC. Shah, A., and F. Shah. 2007. "Citizen-centered Local Governance: Strategies to Combat Democratic Deficits" Development 2007, 50 (1): 72­80. Shah, A., and T. Thompson. 2004. "Implementing Decentralized Local Governance: A Treacherous Road with Potholes, Detours and Road Closures." World Bank Policy Research Working Paper No. 3353, World Bank, Washington, DC. Smoke, P. 2000. Fiscal Decentralization in Developing Countries: A Review of Current Concepts and Practice. Geneva: United Nations Research Institute for Social Development. Smulovitz, Catalina, and Enrique Peruzzotti. 2000. "Societal Accountability in Latin America." Journal of Democracy 11 (4): 147­58. Spahn, Paul B. 2005. "Strategies for Decentralizing Government While Maintaining Stability" Working Paper. www.worldbank.org/mdf/mdf2/papers/gov/spahn.pdf. Steffensen, Jesper, Per Tidemand, and Emmanual Ssewankambo. 2004. "A Comparative Analysis of Decentralization in Kenya, Tanzania and Uganda: Country Study Uganda." World Bank, Washington D.C. Stigler, G. 1957. "The Tenable Range of Functions of Local Government." In Federal Expenditure Policy for Economic Growth and Stability, ed. Congress Joint Economic Committee. Washington, DC: Government Printing Office. Teskey, Graham. 2005. "Capacity Development and State Building: Issues, Evidence and Implications for DFID." Draft, Governance and Social Development Group, DFID (Department for International Development), October. Transparency International (2001). The Corruption Fighters' Toolkit: Civil Society Experiences and Emerging Strategies. Berlin: Transparency International. UNCDF (United Nations Capital Development Fund). 2005. "Building Local Government Capacity to Achieve the Millennium Goals." UNCDF. New York USAID (U.S. Agency for International Development). 1999. "The Governance and Local Democracy Project (GOLD): Summary Assessment." USAID, Manila. 79 Vidaurre, O. E. (2003). Social Audit Program for the Local Road Construction Project, San Antonio del Monte Municipality, Sonsonate, El Salvador. http://info.worldbank.org/etools/docs/library/19102/CaminoVecinalElSalvadorEnglish.pdf. World Bank. 2007a. "Strengthening World Bank Group Engagement on Governance and Anticorruption." World Bank, Washington, DC. www.worldbank.org/html/extdr/comments/governancefeedback/gacpaper-03212007.pdf. World Bank. 2007b. Cambodia Demand for Good Governance Project Information Document. World Bank, Washington, D.C. World Bank. 2006a. "Towards Designing a New Intergovernmental Fiscal System in West Bank and Gaza." Policy Note. Washington, DC: World Bank World Bank. 2006b. Bangladesh Local Governance Support Project Appraisal Document, World Bank, Washington, D.C. World Bank. 2005. Social Accountability Stocktaking Exercise for South and East Asia. Washington, DC: World Bank Institute. World Bank. 2004a. "Fiscal Decentralization to Rural Governments in India." World Bank, Washington, DC. http://siteresources.worldbank.org/EXTSAREGTOPDECENTRALIZATION/ Resources/496899-1095189822590/521764-1095190669879/VolumeI.pdf)World Bank. World Bank. 2004b. World Development Report. Washington, DC: World Bank. World Bank. 2003. Voice, Eyes and Ears. Social Accountability in Latin America. Washington: The World Bank. World Bank. 2000. World Development Report 1999/2000. New York: Oxford University Press. Wittenberg, Martin. 2003. "Decentralization in South Africa." Working Paper. University of the Witswatersrand, Johannesburg. Zarei, Mohammad H. 2000. "Democratic Process and Accountability in Public Administration" Asian Review of Public Administration, XII (2): 40­55. 80 Annex 1: Study on Uganda Local Government Discretion and Accountability ABBREVIATIONS AND ACRONYMS ACAO Assistant Chief Administration Officer CAO Chief Administration Officer CSO Civil Society Organizations DEC District Executive Committee DSC District Service Commission LC Local Council LLG Lower Levels of Local Government LG Local Government LGDP Local Government Development Project LGIFA Local Government Integrated Fiduciary Assessment LGMSD Local Government Management and Service Delivery LGTB Local Government Tender Board MFPED Ministry of Finance, Planning and Economic Development NGO Non Government Organization NRM National Resistance Movement OAG Office of the Auditor General PAC Public Accounts Committee PAF Poverty Action Fund PEAP Poverty Eradication Action Plan PRSC Poverty Reduction Support Credit RDC Resident District Commissioners UDN Uganda Debt Network ULGA Uganda Local Government Association 81 Note: This case study is a product of desk review of literature on Uganda decentralization and local governance. Recognizing the limitations of preparing a case study solely on desk review, the goal is to present a broad outline of the kind of information needed to apply our analytical framework at the country level. The next step is to conduct a fieldwork and a more detailed case study and analyze the local governance conditions based on more specific and primary data using the analytical content of this study. 1.1 BACKGROUND Decentralization of the government of Uganda dates back to the war fought by the National Resistance Army between 1981 and 1986. During this war, the army's political wing, the National Resistance Movement (NRM), mobilized and politicized the people in the areas under its control. Decentralization was emphasized with new leaders highlighting the importance of proximity to policymakers and the necessity of engaging communities in the formulation of policies, which affected them in their daily activities. Museveni's National Resistance Movement saw decentralization and the 39 devolution of power as key means of introducing popular democracy and fostering local governance (Steffensen et al 2004, Francis and James 2003). It was meant to promote capacity building at the local level and ensure that there are local inputs in the decision-making process, especially with respect to service delivery. These factors in turn were to help foster a local sense of ownership of development programmes (Golola 2001). "The failure of the previous centralized systems, the wish to ensure improvements of services, and the political belief in decentralization as a way to involve and get political support from the people particularly in the rural areas and as means to ensure more efficient service delivery ­ are factors, which explain the initial speed of the reforms in Uganda" (Steffensen et al 2004). The people were given the opportunity to elect leaders themselves, setting seeds for democratization process and a civil society in a country that had witnessed years of civil strife. The NRM introduced a system of elected councils of governance at various levels. These councils came to be known as resistance councils and had a high degree of political importance, laying the foundation for a new local government system that came into place when the NRM attained power nationwide in 1986 (Steffensen et al 2004). With the 1995 Constitution (Government of Uganda 1995), the local government system was consolidated further, and the Resistance Councils were renamed Local Councils (LCs). The consolidation process continued with the adoption of the Local Governments Act of 1997 (Jeppsson 2001). The 1995 Constitution of Uganda provides a general framework for decentralization, which is spelled out in more detail in the 1997 Local Governments Act. The decentralization reforms in Uganda included three components: political, administrative, and financial. Political decentralization was based on the Resistance Councils and was implemented throughout the country immediately after the NRM government was formed in 1986. Administrative decentralization was introduced after the adoption of the Local Government Statute, and comprised new administrative structures with a non-subordinated, comprehensive and judicially answerable local administration. Financial decentralization was carried out in phases with the introduction of an unconditional block grant to the districts and through the introduction of locally decided budgets (Jeppsson 2001). The decentralization policy in Uganda is based on a hierarchy of councils and committees including the village (LC 1), parish (LC 2), sub-county (LC 3), county (LC 4) and district level (LC 5) (Refer to figure 1). Of these, only the district and sub-county levels have both political authority and significant resources. 39President of Uganda since 1986 82 In rural areas, the local government system comprises district and sub-county councils, with the latter constituting lower level governments and with administrative units comprising county, parish, and village councils. In urban areas the local government consists of city, municipal or town councils, according to size. Each level is headed by a chairperson, while district levels council meetings are chaired by speakers. Local governments are said to have "autonomy," i.e., legislative and executive authority within their listed areas of jurisdiction. These political structures were legalized by the 1993 Resistance Council Statute, which was in turn enshrined in the Uganda Constitution of 1995, with amendments made the Local Government Act, 1997. Local Government system enshrined in the Constitution protects the local governments against constitutional amendments that may substantially alter the system without their consent. 1.1.2 Challenges to Decentralization and Coordination Central Government Capacity- The establishment of new local governments has sometimes not been based on a thorough analysis of the economic sustainability of these new areas, the potential to generate sufficient revenues and ensure administrative and political capacity to run the council affairs efficiently (Steffensen etal 2004). In fact, the central government is said to use creation of new districts as a form of patronage. The government created 24 new districts between 2004 and 2006 and as a result there are now 80 districts. There are concerns that more districts will be created in the near future, increasing the resource requirements for administrative structures and making it difficult for the central government to monitor, inspect, and supervise local governments as required by law. It is estimated that about Sh40 1.2 billion per year is required to pay the wages and salaries alone for the smallest local government (Uganda PRSC 2007:39). At the same time Ahikire (2002:14) argues that creation of districts allows for ethnic communities to have a native home with full political rights and some element of territorial autonomy. Central Government Interference- At the local government level, the Central Government maintains its oversight through the Resident District Commissioners (RDCs). The RDC is appointed by the President and he or she is responsible for coordinating the administration of government services in a district, advising the district Chairperson on matters of a national nature, and carrying out such other functions as may be assigned by the President or prescribed by Parliament (Constitution 1995, article 203, Local Government Act section 71). Specifically, he/she is responsible for overseeing the proper implementation within the district of the functions that are not yet decentralized, especially those related to defense, security and maintenance of law and order, as well as supervision of projects implementation. RDCs exercise formal duties of monitoring, coordination, and advice with respect to local government units. This could take on the aspect of directing and overruling local government units' decisions in those areas of Uganda with the lowest levels of political mobilization and media exposure. Though the role of the RDCs is supposed to be diminishing, there are a few cases where conflicts have been experienced between the local government leaders and the RDCs, especially in local governments where the local leaders are perceived not to be ardent supporters of the Movement system (Steffensen et al 2004). Civil Society Organizations (CSOs) are relatively undeveloped in Uganda, but have within the last decade developed rapidly. Although the better off and better educated still dominate the decision-making at all levels, ordinary people are increasingly willing to speak out and challenge those in authority (Devas and Grant 2003). But the central government has not always been forthcoming. The National NGO Forum is a newly formed umbrella group of some 600 organizations. Its stated aim is to provide a common platform for all domestic, foreign, and international NGOs active in the country and to enhance dialogue between the nonprofit community and the government. But, the government has not yet recognized the forum­­it exercises significant control over NGO activities through the Non-Governmental Organizations 40One US Dollar~ 1650 Uganda Shilling 83 Registration Statute, which requires that all such groups operating in Uganda be registered. According to a recent Human Rights Watch (1999) report, the government sometimes uses this statute to muzzle its critics (Brinkerhoff and Goldsmith 2003). 84 Figure 1 Local Government Units in Uganda Level RURAL Local Government Units41 URBAN 5 DISTRICT COUNCIL CITY COUNCIL Chairperson: UAS, Secret Ballot Councilors: -do-, Secret Ballot 4 COUNTY COUNCIL MUNICIPAL COUNCIL CITY DIVISION (Administrative Unit) Chairperson EC Secret Chairperson: UAS, Secret Ballot Ballot Councilors: -do-, Secret Ballot Deputy C/P Women: E C, Secret Ballot 3 SUB-COUNTY TOWN MUNICIPAL COUNCIL DIVISION Chairperson: UAS, Secret Ballot Councilors: -do-, Secret Ballot Women: E C, Secret Ballot PARISH WARD WARD (Administrative Unit) (Administrative Unit (Administrative Unit 2 Chairperson & Executives EC Secret Ballot 1 VILLAGE CELL/ZONE CELL/ZONE C/Person ­ Secret Ballot C/Person ­ Secret Ballot C/Person ­ Secret Ballot UAS- Universal Adult Suffrage, EC- Electoral Commission, C/P- Chairperson 41Levels 1, 2 and County Council (at level 4) are Administrative Units, the rest are LGs. 85 Local Government Capacity- Decentralization has created a new set of rules, procedures, and incentives for the allocation of development resources which create characteristic lines of competition, conflict, and alliance. One axis of conflict is between the various levels of local administration. There is resentment at the village and parish at the failure of higher levels to include them in decision- making processes in more than a token way, or to release the funding which they are due. Likewise, the sub-county level, although now the lowest level of government, is chronically under-resourced and tends to blame the district for this (e.g., for its failure to remit the graduated tax of public employees). Higher levels justify their reluctance to devolve resources by the lack of capacity and accountability mechanisms at lower levels. Indeed, while sub-county books are audited, sanctions are rarely brought in cases of irregular spending (Francis and James 2003). The persistence of insurgency in the northern part of the country does not allow local governments (political and administrative structures) to function efficiently and deliver the mandated functions within this area. 1.2 POLITICAL SETTING AND ACCOUNTABILITY IN DECENTRALIZATION 1.2.1 Discretion Separation of Powers- The local government civil service in each district is headed by a Chief Administration Officer (CAO) in juxtaposition with an elected political authority, the district council, led by its Chairperson (LC5), also elected by popular vote at the district level. The CAO (Town Clerk in the city) is responsible to, and subject to, the general directions of the Chairperson and the district council (Refer to Figure 2). In addition, there is a Deputy CAO and a number of Assistant CAOs and other officers and employees of the district as council may determine (LG Act section 67). In a number of districts this has produced instances of conflict over decision-making responsibilities and expenditure priorities between technical and political personnel (Livingstone and Charlton 2001). The district council has a speaker elected from the council. The District Executive Committee members are nominated from the council by the chairman and approved by the members of the council. The Executive Committees of the lower local government (sub-county, parish and village) function as local courts. 86 Figure 2 Administrative Structure at the District Level District Council CAO District Technical Planning Committee Managemen Finance & Education Works and Directorate Directorate Community t Support Planning and Sports Technical of of Health Based Services Services Production Services Services Electoral System- The principles governing the Movement were participatory democracy, accountability and transparency, accessibility to all positions of leadership by all citizens, and individual merit as a basis for election to political offices. LC1s, LC3s, and LC5s are elected through 5-yearly elections with universal adult suffrage; the chairperson of each of these councils is also directly elected by universal adult suffrage (refer figure 1). LC2s and LC4s, however are indirectly elected by councilors on LC1s and LC3s respectively. This system allows unpopular councilors (at district, sub-county and village level at least)to be voted out at elections; in practice however, this rarely happens and the councils have a relatively stable membership Until 2005 the elections were held under the Movement political system, which meant that every candidate had to swear allegiance to the movement system in order to run for elections. Party System- The government until 2005 followed a "no party" system because the NRM leadership held that multi-partyism would revive ethnic and religious cleavages. Blair (2000) states that the existence of competitive political parties is a necessary requirement for encouraging a culture of local accountability. Azfar et al (2001) argue that the "no party" system affects the local level more than the national level since the state and the local councils have greater authority in the provinces than in the capital and other large cities. Observers cite evidence that both the local councils and the Resident District Commissioners (RDCs) did much of the Movement's work in the interior, providing platforms for candidates, spreading the Movement ideology, administering political training and other functions(Azfar etal 2001, 34). Thus, the absence of parties effectively neutralizes, delegitimizes, and disallows divergent tendencies (Brock etal 2002:37). 1.2.2 Supply Side Approaches to Improve Downward Accountability Electoral Accountability- The local council elections are organized, conducted and supervised by an independent Electoral Commission. The Electoral Commission appoints a Returning Officer for each district (Section 103 of the LGA 1997). Uganda has introduced several measures to improve the quality of the electoral system. As mentioned earlier, Councils and their Chairpersons are directly elected through a competitive system; provision has been made for minimum quotas of women, youth and disabled in councils (Ribot 2004). Independent candidates are admitted in local elections (Ribot 2004). Authorities chosen through competitive local 87 elections that admit independent candidates appear more likely to be locally accountable; in addition, they may be more sustainable and scaleable across national territories (Ribot 2004). The electorate has the power to recall local government members for neglect of their duties. Apart from provision for minimum quotas of women, there are also quotas for youth and disabled in councils (Ribot 2004, Golola 2001:7). But the reservation of seats for women, youth, and disabled does not seem to have had much impact on outcomes, partly because of the limited skills and effectiveness of those occupying such seats (Rakodi, 2002). Female councilors acknowledge that while men have become more aware of women and children issues as a result of statutory requirements for women to be represented on all local councils, this does not mean that their decisions are more gender sensitive. In addition, female councillors are found to be more hesitant and less vocal higher up the local political system: whereas discussions at the village or ward level (LC1) are effectively with neighbors, at district level (LC5) the issues are more complex and female councilors are intimidated about speaking out (Saito 2000). Electoral representation is challenged because membership to local councils is almost exclusively drawn from richer households; this is in part because of the `goodwill gestures' such as distribution of salt and soap in election campaigns and partly because of levels of literacy and conception of status (Gaventa 2002: 23) Currently, the elections of the President, members of Parliament, the chairpersons and councilors of the respective local governments are organized on separate days/occasions. It has been argued that the election of one category of leaders influences the election of the other leaders. The Electoral Commission has therefore made a proposal to merge some of the elections. For example, there is a proposal to merge the Presidential Elections with Parliamentary elections, district/city Chairpersons/Mayor with the district city councilors, local government chairpersons of the sub-county/town council/municipal division with the local government councilors and village chairpersons. Whether this will increase downward accountability of the local government is yet to be seen. Local Council Oversight- Uganda has a separate executive headed by the CAO and the council headed by the chairperson at the local government level. The local council oversees the administrative units in a number of ways such as approving budgets, staffing etc (for details refer to figure 3). Democratic control of local government affairs is enhanced through the committee system, which ensures the joint participation of elected councilors and civil servants in service delivery and management decisions. Also the local councils are responsible for nominating and approving the statutory boards and commissions such as the Local Government Tender Board and local government public accounts committee as well as District Service Commissions (DSC)42. These bodies are responsible for keeping account of different aspects of the local government (for details refer to figure 3). At times, since the body members are appointed by the local government council, they become highly dependant on the council thus affecting their integrity. 42The DSC is appointed by the district council on the recommendation of the District Executive Committee with the approval of the Public Service Commission. The DSC is constituted of a Chairperson and such other members as the district council determines. The district council has to ensure that at least one third of the seats are reserved for women and at least one member of the Commission is a person with disability (LG Amendment Section 55 2a). The DSC has a secretariat manned by the Secretary to the DSC at the rank of, or equivalent to or above, Principal Personnel Officer. The Secretary to the DSC is responsible for implementing the day-to-day decisions of the Commission. The functions of the DSC include appointments/recruitment, confirmation of appointments, exercising disciplinary actions including termination of service. 88 These clear levels of oversight between the council and administrative bodies do not always work well. There are large capacity constraints at different levels. Some elected officials do not have the capacity to comprehend and appreciate accountabilities submitted from line departments because of inadequate education of such politicians (Onyach-Olaa 2003; Golola 2001: 9). In a number of instances the civil servants are better educated than the political leadership of the district, yet their salaries and allowances are considerably lower. For example, the salary of a graduate assistant CAO is in the region of Sh. 210,000 per month, while a non-graduate chairperson earns Sh. 1,300,000 (Francis and James 2003). The capacity of LC 5 authorities to provide oversight of district level accounts needs to be strengthened. At present sub-county accounts come to the National Parliament (the Local Government Public Accounts Committee) for oversight, which is more than they can realistically cover. Strengthening District Public Accounts Committees (PAC) so they can start to handle oversight of sub-country accounts would help to strengthen accountability systems at the local level. At the lowest levels, much of the work done by councilors is voluntary. Some local council members even complain about `poor facilitation' and the fact that they have to abandon their income generating activities to attend to council work without compensation. Such people might be difficult to hold accountable to the local communities (Golola 2001). Also the role of RDCs, as discussed earlier, can sometimes undermine the role of the local government and blur the lines of accountability. The role of the Resident District Commissioner (RDCs) has led some to question whether the RDC, the District Chair, or the Executive Committee is really "in charge" at district level. Whoever wins this contest has a considerable say at the sub-county level, despite the latter's formal autonomy. The Local Governments Act spells out dispute-resolution processes to be used in cases of conflict of authority, but these apparently have been little used (Azfar et al 2001). The possible centralization43 of some elements of personnel management, in particular the appointment of the Chief Administration Officer (CAO), poses a fundamental threat for the future of devolution in Uganda ­ because the central government appointment of the CAO would cut the most crucial accountability link between the elected local councils and the local government staff. Also, such an approach to solving administrative problems in local governments (through centralization) could result in a snowball effect, whereby local accountability mechanisms become undermined (Steffensen et al 2004). 43 The Constitution Review Commission has suggested centralization of the appointment of CAOs who head the civil service and are currently appointed by the DSC. The arguments is that a central government appointed CAO will be more protected from local political interference and pressures and that the CAO in view of the local government dependence on central government financial transfers should be held centrally accountable (Steffensen et al 2004).. 89 Figure 3 Local Council Role and Oversight over Administration 90 Steffensen et al, 2004:48,49 1.2.3 Demand side Approaches to Improve Downward Accountability There are now NGO-led attempts to develop the skills of local representatives (Rakodi, 2002). The media, particularly local radio stations have helped to build local accountability. Local radio stations- of which there are many- hold phone-in programs in which mayors and chairpersons of councils are invited to respond to people's questions and complaints (Devas and Grant 2003). In spite of regular local elections, there is failure/omission by the councilors to report to the electorate the general decisions to the Council and the actions it has taken to solve problems raised by the people in the electoral area; electorates' reluctance to demand for explanation on issues they feel concern them; reluctance of the electorate to recall a councilor for neglect of his/her or duties as councilor or for having committed acts incompatible with his/her position as a council member; and reluctance of the electorate to sign a petition against their leader (ULGA 2004) . 91 1.3 ADMINISTRATIVE DISCRETION AND ACCOUNTABILITY IN DECENTRALIZATION 1.3.1 Discretion Ability to make, change, and enforce laws and regulations- The functions and services, which are the responsibility of central government, and those of local governments at various levels, are clearly distinguished to minimize overlap and conflict. The constitution provides districts the responsibilities for all the functions and services not specified in the sixth schedule of the constitution. They are thus charged with virtually all functions except policy issues and standard and inspection related to the delivery of basic social services (Steffensen etal 2004:9). In the past few years, local councils have assumed a number of functions which were previously performed by line ministries. These include political administration, judicial services involving minor cases, monitoring and supervision of development projects at the local level and maintenance of community roads (Golola 2001: 8). Local governments may enact local legislation to backup policy implementation, e.g. to control the use of local natural resources, to sanction urban pollution, etc. Also each district is to produce its own district development plan. As mentioned earlier, the lower courts possess judicial authority as well. This expansion of the role of local councils in political dimension of decentralization has had its risks and problems. In local disputes, village councils have the authority to settle civil disputes, including those related to land ownership, domestic conflict and payment of debts. Yet, general concern has been raised over the fact that, local councils often exceed their authority by hearing criminal cases including murder and rape/defilement (Human Rights Report, 1993). There are even other under-currents that the socially powerful groups are given an upper hand in these supposedly local and participative processes of dispute settlement. Poverty assessments carried out under the auspices of the Ministry of Finance, Planning and Economic Development (MFPED) across the country have established that local councils promote development but are in some cases endangering development and human rights. In Kapchorwa (eastern Uganda) it was found out that land was getting concentrated in the hands of the rich. Poor people complained that judgments in land wrangles were usually in favor of the rich due to corruption and bribery. They also complained that local councils over-assess them for graduated tax (which was later abolished) (Muhereza et al., 2001:13 in Ahikire 2002: 19. Discretion over civil service and employment policies- Local governments are empowered to establish their own appropriate staffing structures. Though the salaries of pensionable staff is generally determined centrally, local governments exercise autonomy in establishing various benefits such as staff loan schemes, training, as well as housing. In reality their autonomy is curtailed by the financing modalities, whereby a large number of staff is paid through specific conditional grants (teachers, health staff, water officers, agriculture staff etc.) Health staff, teachers, agriculture extension workers, staff of the district water office, as well as community development assistants are paid by earmarked transfers. These constitute the majority of local government employees. The few offices not benefiting from central government funding (e.g. human resource unit/personnel office) are often understaffed and de-motivated as they even lack operational budgets (Steffensen etal 2004:138). The Ministry of Public Service is currently leading a restructuring exercise where the main recommendation is to establish standardized staff structures in local governments (Steffensen etal, 2004:xvii) Since 1993, local governments ­ through their DSC­ have begun to take on direct personnel management responsibility for all local civil servants deployed at the district level. The District Council (consisting of elected politicians) on the advice of its District Executive, appoints its members who are unelected citizens (Francis and James 2003:333). The DSC appoints disciplines and has the power to remove all 92 district staff. This streamlining of the local governments' employment process has helped reduce instances of irregular appointments44 (Golola 2001). At the same time, civil servants feel that the power exercised by the DSC makes them vulnerable to undue pressure or even victimization should they go against the wishes of local politicians who at times exert pressure on DSC. The center is said to have political influence over the DSC as well. This has at times resulted in unjustified promotions, appointments, and dismissals (Uganda Local Government Association 2004). The DSC of one district, under pressure from the District Executive, had dismissed a Deputy CAO after he questioned the use of the unconditional grant for politicians' allowances (Francis and James 2003). Conflict is one response to this situation, collusion is another. In what is sometimes termed the ``sons of the soil'' phenomenon (Francis and James 2003:334, ULGA 2004:6), local applicants are favored for administrative appointments over candidates from other parts of the country. Politicians are said to exert pressure on the DSC to employ indigenes, who are likely to be more malleable, or easily enmeshed in local structures of patronage. This practice not only further weakens technical capacity, but also undermines the national character of public administration (Francis and James 2003). Discretion over procurement process- The local government is not directly involved in the procurement process. Instead it is responsible for hiring the District Tender Board staff. The District Tender Board, until recently, was responsible for awarding contracts and tax collection rights. The Local Government Tender Boards (district and urban) were formed in each district and were supposed to provide services to the district councils, sub-county councils and administrative units under a district. The District Tender Board consists of seven members, including the Chairperson, all of whom are appointed by the district council on the recommendation of the District Executive Committee. In addition, there is a Secretary to the District Tender Board, at the rank of Assistant CAO or above. All expenses of a District Tender Board, including salaries and emoluments of the members, are funded through transfers from the central government (Consolidated Fund). 1.3.2 Supply Side Approaches to Improve Downward Accountability The local governments are corporate, autonomous bodies, which can thus sue and be sued, manage funds, enter into contracts, employ staff, and be held legally accountable (Steffensen et al 2004). Regulatory decisions- Apart from the lower level local councils that have taken on the judicial role as well, there are magistrates' courts at higher local government levels. These courts can handle certain criminal cases and civil cases within a certain monetary value (http://www.judicature.go.ug/magistrates.php). When conflicts arise in local government decisions, people have resorted to the court system. For example, local governments that have dismissed CAOs without due course have experienced excessive costs implications in the form of compensation and court fees and they therefore now much more carefully follow the law (Steffensen etal 2004).. Apart from courts, communities wishing to take up issues of (for example) non-delivery by contractors can bring it up with a range of actors (such as the CAO, LC3 chairmen, or the member of parliament). In the case of major problems the Member of Parliament has the option of taking issues up with the Local Government PAC in parliament. Systems of redress clearly exist and can function. Deeper investigation would be valuable to find the limits of these, and the significant conditions which affect the likelihood of communities getting results from this type of action (Andy's BTOR). 44But on the whole, the recruitment of personnel has been easier in the richer districts, to the south of the country than the poorer ones to the north, where lack of opportunities for generating extra incomes has made it difficult to attract competent staff (Golola 2001). 93 Service delivery and procurement- Until recently, the District Tender Board was responsible for awarding contracts and tax collection rights. By virtue of being a separate, statutory body, it was supposed to minimize direct council interference in its processes. But this objective was not fulfilled. As a result of public outcry, the Local Governments Act was recently amended to abolish the local government tender boards and replace them with contract committees to depoliticize procurement. In compliance with the amendment to the Local Governments Act, 2006, local governments are required to establish procurement and disposal units, intended to professionalize the procurement function (PRSC 7 2007:39). This amendment was passed because local community members and administrators alleged that the patronage of the District Executive Committee over membership of the District Tender Board, with its privileges including generous sitting fees, enabled them to influence the decisions of the board unduly. Tenders (e.g., for tax and construction contracts) were supposed to be allocated on the basis of a points system that took account of a range of criteria including price, experience, and record of tax payments. In practice though, a letter of recommendation from a politician was believed to be a crucial prerequisite, and it was widely believed that successful tenderers were friends, relatives or protégés of the political class, or proxy companies operating on their behalf (Francis and James 2003). 1.3.3 Demand Side Approaches to Improve Downward Accountability The city of Kampala conducted its first citizen report card in early 2005. It provided the City Council and other basic public service providers with feedback on water and sanitation, health, education, roads and public transport, solid waste management, public toilets, the management of the city environment, maintenance of law and order, and management of city infrastructure (Kampala City Council and Uganda Management Institute 2005). There are currently two initiatives which seek to enhance accountability of public institutions for delivery of services and other benefits at the local level. The World Bank is supporting through an IDF grant an initiative coordinated through Margaret Kakande in the Ministry of Finance to monitor service delivery through a network of national NGOs. A key objective is to encourage the range of civil society groups involved to develop a more coherent analysis and advocacy platform in order to have more impact on public policy and service delivery at both national and local levels. They hope to complete fieldwork by the late Spring, with results to follow later in the year. On a smaller scale, but with a longer time-frame, the Uganda Debt Network (UDN)45 is currently supporting community monitoring of government projects and delivery of performance of contracts by local firms in 64 communities in seven Districts (Andy's BTOR, 2007). 1.4 FISCAL DISCRETION AND ACCOUNTABILITY IN DECENTRALIZATION Beginning in the 1993-94 financial year, there was a phased fiscal decentralization, which began with 13 districts, with another 14 districts decentralized in the 1994-95 financial year. In 1995-96 financial year another 35 districts also entered the process. Local government expenditure which peaked in 2002-2003 at 36 percent of the national budget has since declined, amounting to 32 percent of budget allocations in 2005-2006. Local government spending remains high at nearly 7 percent of GDP (Williamson et al 2005). 45 The UDN is an advocacy and lobbying coalition of NGOs, institutions and individuals formed in 1996 to campaign for debt relief. It has been engaged in budget advocacy since 1998, at both national and local levels, and is a member of the `International Budget Project' network coordinated from Washington DC. 94 At the center, Local Government Finance Commissions (LGFC) play a key role in providing technical assistance to the central and local governments in the areas of revenue sharing, local government revenue mobilization and reforms of the transfer system. They also play a strong role in championing fiscal decentralization (For further information visit http://www.lgfc.go.ug/index.php). 1.4.1 Discretion Expenditure Assignment- The five sectors of primary education, primary health care, water and sanitation, rural roads, and agriculture extension, were devolved to local governments (Onyach-Olaa 2003). The District Council list includes primary and secondary education, a range of primary health services (including certain hospitals and health centers, maternal-child health, communicable disease and vector control, and health education), and basic services in the areas of water provision, roads, planning, and licensing. A number of the listed areas, including primary education, community-based health services, hygiene, and low-level health units, are to be devolved by the district to lower-level councils (Azfar et al 2001). The formal system of planning in Uganda is supposed to proceed in an integrated bottom-up manner. Each village produces a Community Action Plan based on local needs and priorities. Parish councils incorporate these plans into parish plans, which are in turn synthesized into sub-county plans. The District Technical Planning Committee is then supposed to produce an integrated plan based on sub-county plans and the deliberations of a budget conference of key stakeholders, and this plan is ratified by the District Council. In practice, however, the system scarcely articulates at all between levels. First, priorities from LC1 and LC2, if they reach higher levels, are rarely actually incorporated into sub-county plans. Also, given the limited capacity and experience of politicians and civil servants, at the sub-county level plans of any quality or realism rarely materialize. When they do appear, plans are generally based on over- optimistic estimates of revenue (Francis and James 2003:331). Although the recurrent expenditure budget has been decentralized (Jeppsson 2001, Livingston and Charlton 2001), there is a shortage of funds to cover recurrent expenditure resulting in a failure to deliver services, as revealed in the uniformly run-down, even deteriorating condition of urban infrastructure and, in rural areas, in poorly maintained access roads (Livingstone and Charlton 2001). The development budget has not been decentralized and the district administrations have little or no capacity for funding development projects, including social projects such as the construction of schools (Livingstone and Charlton 2001). The district governments' taxing potential is not commensurate with the responsibilities being devolved. Specifically, while the local revenue sources and instruments available to local governments have remain unchanged from those operational in the pre-National Resistance Movement era, local expenditure responsibilities have expanded exponentially (Livingstone and Charlton 2001). Thus while local inputs have made an important contribution to local government activities in some parts of the country in others there has been a more than disproportionate dependency on central government funding (Jeppsson 2001). Revenue Assignment- Local governments are empowered to levy certain prescribed taxes and, furthermore, the lowest local government level ­ LC3 or Sub-county ­ is empowered to retain 65 percent of all revenue collected. Local governments generally have certain autonomy to adjust their taxes, charge fees and other local revenue to the local priorities and to improve on local government revenue raising, provided that the local conditions, political and administrative circumstances are conducive for this. Contrary to the situation in many countries, there is no interference from the Ministry of Local Government on specific tax rates and levels provided that the local governments are within the legal requirements (Steffensen etal 2004). 95 Apart from transfers (discussed in the next section), local authorities in Uganda are dependant on a limited number of revenue sources: graduated tax46 (suspended in FY 2004-2005), market dues, licenses and fees, and in the case of municipalities, property tax and ground rent. When the central government abolished the graduated tax, they took away one of the major sources of revenue of up to Sh 85 billion per annum. Although the central government promised to compensate local governments for this revenue loss, only Sh 35 billion was compensated during FY2005/06, while only Sh 25 billion is budgeted for in FY2006/07. As a result, local governments are finding it difficult to finance their operations (Uganda PRSC 2007:39). The PRSC 7 states that the biggest problem facing local governments today and which may undermine the decentralization policy is the declining local revenue. This results in local governments to continue to be upwardly dependant and thus, upwardly accountable. The local government may adopt additional taxes, but only with the approval of the Ministry of Local Government47. This essentially limits local governments to minor variations from the list, since the law provides no standard by which the Minister approves or disapproves of proposed new revenue sources (Azfar et al 2001). At the same time, as mentioned earlier, the central government does not interfere in setting of tax rates and levels by the local government. Some local governments are not viable ­ they are too small in population and local revenue potential (Steffensen et al 2004, Livingstone and Charlton 2001) (though Onyach-Olaa writes that the size of Uganda local governments- districts and sub-counties- are appropriate for the infrastructure service delivery role). Intergovernmental Transfer- In terms of revenue, the local governments are increasingly reliant on central transfers thus making them upwardly accountable. Besides block grants, two other grants are the conditional grant and the equalization grant. Conditional grants were to be based on certain conditions, to encourage districts to undertake projects in areas such as poverty reduction, which would have been ignored otherwise. However, the conditions have been difficult to define and these types of grants have been given to districts without the accompanying conditions, thus becoming a form of block grants. The equalization grant is to be based on assessment of resource endowments, with poorer districts compensated accordingly. Approximately 88 percent of the local government expenditures are financed by transfers as earmarked 48 conditional grants. Conditional grant terms contain not only affirmative duties but also list of things that grants cannot be used for. This leaves very limited autonomy for the local governments to decide on the priorities and adversely undermines the local government efforts to raise revenues (Steffensen et al, Francis and James 2003, Jeppsson 2001, Onyach-Olaa 2003). The conditional grant reporting requirements are said to be onerous (Azfar et al 2001). The center's tight control over the use of intergovernmental transfers may have impeded, in some ways, the development of local accountability as officials and elected representatives have focused their attention on meeting grant conditions (Devas and 46Graduated tax was approximately 80 percent of the total tax revenue in Uganda (Steffensen et al 2004). It was payable annually by all adult males and theoretically salaried women as well. The graduated tax was highly unpopular and during elections, politicians at all levels encouraged default (Francis and James 2003) It was unpopular not only because it was not means based, but also because its collection, especially at the lower levels, involved considerable coercion, including the deployment on armed guards at times (Golola 2001). 47The Ministry of Local Government is currently exploring new revenue measures for the local government. These include taxes on local services, hotels, and vehicles (PRSC 7: 14) 48Figure is unconfirmed. PRSC 7 states that 94% of the local government revenue was from central transfers in 2002-2003. 96 Grant 2003). Another reason for loss of accountability is that conditional grant funding has blurred the links between tax payment and the service provision and led to the diminishing of incentives to collect own revenues. The lower tiers of government lack the ability to manage public finances and maintain proper accounting procedures. Since these were a requirement for transferring money to the lower tiers, they actually received less money than before decentralization (Livingstone and Charlton 2001). Some of these issues are being addressed through realization of the grant system, more appropriate grant conditions and improved performance monitoring and inspection as well as training of officers and elected representatives (Government of Uganda 2001). Also funds for local government operations are not routed through line ministries. The Minister of Finance remits funds directly to the accounting officer of each district. An innovation is that the amounts disbursed are published in local newspapers. This enables the local populations to know what to expect in terms of service provision and acts as a deterrent to financial abuse (Golola 2001). While much of the resources transferred is in the form of specified grants over which there is relatively little local choice, some transfers--notably the Local Government Development Programme (LGDP)-- offer real choices from within a menu of types of local infrastructure which match national priority areas (education, health, roads, water, production). This process applies at all levels, with each tier having a specified amount of LGDP resources to allocate for projects appropriate to their scale of operation. Thus, citizens are able to participate at various levels in real choices about real resources. Although at LC1 and LC2 the amounts involved are quite small, the programme has produced a great deal of new infrastructure serving the poor. Projects are implemented by LC3, LC4 or LC5, depending on the nature of the project, with local Project Management Committees drawn from LC1 and LC2 monitoring the implementation of their projects (Devas and Grant 2003). Local Government Borrowing- The legislative framework in Uganda in unclear on the local government's discretion to borrow. In practice, local government borrowing is very limited and typically restricted, but short term arrears and overdrafts are significant (Steffensen etal 2004:111) Donor Funding- Foreign aid continues to account for a significant share of government expenditure, equivalent to 52% of the government budget in 2004 (Robinson 2005). All donor funding, the most significant source of capital spending, passes through the sector ministries, and the increasingly dominant Ministry of Finance, Planning and Economic Development (Francis and James 2003). The center is said to use its political influence in choosing the districts for pilot projects (Uganda Local Government Association 2004). 1.4.2 Supply Side Approaches to Improve Downward Accountability Public Financial Management- The decentralization of activities to the local government level over the last 10 years means that 40 percent of the total budgeted expenditure for spending agencies is now spent at that level. Seventy-five percent of the Poverty Action Fund (PAF) resources also are channeled through local governments. The Local Government Integrated Fiduciary Assessment (LGIFA) benchmarking exercise of a sample of districts and municipalities, together with the results of a financial management study commissioned in 2003, made an assessment of the state of public financial management at the local government level. LGIFA highlights the considerable progress made over the last decade in providing services at the local level; from this base, however, it notes with concern that the budgeting and planning processes at local governments are poor at articulating specific local needs within overall national objectives and policies. The assessment also raises concerns over the ability, desire, and willingness of local residents and politicians to hold their administrations to account for their performance (PRSC 7: 54). 97 In every district, city, municipal and town council there is an established internal audit department which reports directly to the council. The local government Public Accounts Committees (PAC)49, composed of people outside the administration, review the internal and external audit reports. Both the internal audit systems and local government PACs are so far quite weak and are not taken seriously by the councils (Steffensen etal 2004:119). 1.4.3 Demand Side Approaches to Improve Downward Accountability Civil Society Organizations are now actively engaged in a number of policy issues such as the development of the Poverty Reduction Strategy, the debate on the debt relief process, the Public Expenditure Tracking mentioned above, and monitoring of the PAF. The engagement of CSOs has mainly been through a few strong organizations based in Kampala, but similar engagement is also slowly emerging at the local government level. For example, in Tororo a network of local NGOs has been created that engages with the local government in a dialogue on budget issues and tries to monitor and report on aspects of corruption. This engagement is often donor funded (Steffensen et al 2004). Citizen's participation in budgeting and expenditure Public Expenditure Tracking- 1. To find out where money was being lost, through corruption or mismanagement, the Ugandan government, with the support of the World Bank, tracked public expenditure in a survey of 250 schools. The data on grants per schoolchild showed only a shocking 2% of the allocation reached schools in 1991. Most, particularly poor schools, received nothing at all. On the positive side, the survey revealed children enrolled in schools had in fact risen 60%. Both schools and districts had benefited financially from underreporting enrollments (World Bank,na). The government acted immediately to improve the flow of information and make budget allocations transparent. Key steps it took were: Publishing on a monthly basis in newspapers and radio broadcasting the exact amounts transferred; Requiring schools to maintain public notice boards, posting funds received; Legislation protecting accountability and information dissemination; Requiring districts to deposit all grants to schools directly into school bank accounts; and Delegating authority for procurement to schools (World Bank,na). Some of the impacts of these steps were: Grants per schoolchild rose dramatically, from 2% in 1991, to 26% in 1995, to more than 90% in 1999. 15,000 `ghost employees' were eliminated from payroll by 1993. These non-existent salaries represented 20% of the total number of teachers; Overall non- wage funds leaking from the system dropped, from 97% lost in 1991 to 18% lost in 2001; Schools with access to newspapers increased their funds by 10% more than schools that lacked newspapers (World Bank,na). 49 The Local Government Public Accounts Committee (PAC) consisting of four members appointed by the district council on the recommendation of the executive committee. The local government PAC examines the reports of the Auditor General, Chief Internal Auditor and any reports of commissions of inquiry and submits its report to the council and the minister responsible for Local Government, who is supposed to lay the report before Parliament. The Chairperson of the council and the CAO or Town Clerk are responsible for implementing the recommendations of the local government PAC. 98 UDN has been monitoring the implementation of a national School Facility Grant and had discovered a widespread inefficiency in fund utilization. Its subsequent national campaign, led to the changes discussed above. 2. The Uganda Debt Network established local monitoring committees in 2002 in order to track local public expenditures (de Renzio etal 2006). UDN set up community based monitoring and evaluation system in eight districts. Village and district level committees monitored local expenditures and promoted dialogue around problems that could be solved locally. For example, if community monitors found that certain teachers where not doing their jobs properly or where absent from work, the problem was discussed with the village authorities and the individual could be disciplined or transferred. On the other hand if monitors noted that building material for construction of schools did not conform to the relevant specifications, it was brought to the attention of the district authorities. Often the approach to demand corrections is non confrontational, building upon the committee's capacity to negotiate and the openness of government officials and the respective public accountability processes. At the same time UDN backs up its claims through the use of medias, like radio programs or regular anti-corruption campaigns. UDN's work on at district and sub-district level on budget implementation, service delivery and anti-corruption allows it to generate inputs and policy recommendation for the national level with a local perspective. Participatory Budgeting- There are examples of local councils which have attempted to make the budget conference into an inclusive process with proper discussion of budgetary choices. For example, to augment transparency and accountability in decentralized management of resources, advertisements are placed in the press indicating amounts disbursed to each district by sector. In the education sector, budget allocations for schools are posted on school notice boards (Mugambe and Robb 2000 in Brinkerhoff and Goldsmith 2003). In one municipality, Entebbe, the mayor undertook a major "budget outreach" exercise every year, visiting each village to discuss priorities with residents, accompanied by municipal officials, councilors and civil society organizations (Devas and Grant 2003). Ex-post Audits- An important organization in policy dialogue on decentralization issues is the Uganda Local Government Association (ULGA)50. ULGA has developed its capacity substantially and now engages itself proactively in policy dialogue with the central government and donors, and it is now capable of carrying out its own independent analysis ­ such as the annual analysis of the central government budget from a local government perspective (Steffensen et al 2004). Impact of local government accountability on Tax Compliance- There is general perception among many citizens that people's taxes are being used largely to cover the salaries of the local administration- for non productive and non-statutory uses (Francis and James 2003, Livingstone and Charlton 2001, Steffensen et al 2004). This general distrust in the relationship between the politicians and citizens is attributed to lack of information, transparency, communication, and dialogue. This has generated a widespread unwillingness to pay taxes, exacerbating the problem of poor service delivery and generating a `vicious circle' of non ­compliance. Increased legitimacy through participation and the improved tax collection go hand in hand and can be a mutually reinforcing process. 50In Uganda the two Local Government Associations are the Urban Authorities Association of Uganda (UAAU) and Uganda Local Authorities Association (ULAA). Both associations are registered as non-profit making organizations composed of local governments in Uganda. They advocate and negotiate with central government and other stakeholders on behalf of the local governments. 99 In spite of the work of the above mentioned CSOs, there are concerns that the level of participation is much less than is suggested by the legislation. Meetings at village level do not happen as frequently as they should; budget conferences are attended by relatively few, and the language and style of these effectively exclude many; and voter turnout at local elections appears to be declining (Devas and Grant 2003). 100 REFERENCES Ahikire, Josephine. 2002. Decentralization in Uganda Today: Institutions and Possible Outcomes in the Context of Human Rights. International Council on Human Rights Policy. Azfar, Omar, Satu Kähkönen, and Patrick Meagher. 2001. Condition for Effective Decentralized Governance: A Synthesis of Research Findings. Working Paper No. 256. Center for Institutional Reform and the Informal Sector. University of Maryland. Benson, Catherine. 2006. The Role of Local Institutions in Ensuring Poverty Reduction and Environmental Sustainability in Uganda's Natural Resource Decentralization. Presented at Survival of the Commons: Mounting Challenges and New Realities, the Eleventh Conference of the International Association for the Study of Common Property, Bali, Indonesia, June 19-23, 2006. Björkman, Martina, Ritva Reinikka, Jakob Svensson, and Gibva Kajubi. 2007. Promoting Local Government Accountability: A Pilot Project in Uganda's Health Sector. Working Paper. World Bank. Washington D.C. Blair, Harry. 2000. Participation and Accountability at the Periphery: Democratic Local Governance in Six Countries. World Development. Vol. 28, No. 1. pp 21-39, U.K. Bossert, Thomas J., and Joel C. Beauvais. 2002. " Decentralization of Health Systems in Ghana, Zambia, Uganda, and the Philippines: a Comparative Analysis of Decision Space." Health Policy and Planning.17(1):14-31. Oxford University Press, New York. Brinkerhoff, Derick, and Arthur A. Goldsmith. 2003. "How Citizens Participate in Macroeconomic Policy: International Experience and Implications for Poverty Reduction." World Development. Vol. 31, No. 4, pp. 685-701, Great Britain. Brock, K., McGee, R. and Ssewakiryanga, R. 2002. Poverty Knowledge and Policy Processes: A Case Study of Ugandan National Poverty Reduction Policy. Research Report 53. Institute for Development Studies: Brighton. de Rezio, Paolo, Vitus Aseem and Vivek Ramkumar (2006). Case Study on the Uganda Debt Network prepared for International Budget Project, IBP. Devas, Nick and Ursula Grant. 2003. "Local Government Decision-Making- Citizen Participation and Local Accountability: Some Evidence from Kenya and Uganda." Public Administration and Development. Vol. 23, Issue 4. 307-316. Department for International Development, UK Francis, Paul and Robert James. 2003. Balancing Rural Poverty Reduction and Citizen Participation: The Contradictions of Uganda's Decentralization Program. World Development, Vol. 31. No. 2, pp. 325-337, Great Britain. Gaventa, Jonathan. 2002. Legal and Policy Frameworks for Citizen Participation in Local Governance in East Africa: a regional report. LogoLink. Institute of Development Studies, University of Sussex,UK. Golola, Moses L. 2001. "Decentralization, Local Bureaucracies and Service Delivery in Uganda." Discussion Paper No. 2001/115. United Nations University/World Institute for Development Economics Research. Helsinki. 101 Jeppsson, Anders. 2001. Financial Priorities under Decentralization in Uganda. Health Policy and Planning. 16 (2): 187-192. Oxford University Press. Livingstone, Ian and Roger Charlton. 2001. "Financing Decentralized Development in a Low-Income Country: Raising Revenues for Local Government in Uganda." Development and Change. Vol. 32, 77-100. Institute of Social Studies, Oxford, UK. Onyach-Olaa, Martin. 2003. Lessons for Experiences in Decentralizing Infrastructure and Service Delivery in Rural Areas. Uganda Case Study. UNCDF-Sponsored Country Case Studies for Africities Conference. Ribot, Jesse. 2004. Waiting for Democracy: The Politics of Choice in Natural Resource Decentralization. Washington D.C.: World Resource Institute. Robinson, Mark. 2005. The Political Economy of Turnaround in Uganda. Paper prepared for the Low Income Countries under Stress (LICUS) Initiative, The World Bank, Washington, D.C. Saito, Fumihiko. 2000. Decentralization in Uganda. Towards a Positive-Sum Solution. Ryukoku University, Japan. Also available at www.world.ryokoku.ac.jp/~fumisait/en/publication.html Steffensen, Jesper, Per Tidemand, and Emmanual Ssewankambo. 2004. "A Comparative Analysis of Decentralization in Kenya, Tanzania and Uganda: Country Study Uganda." Washington D.C.: World Bank. Uganda Local Government Association. 2004. Discussion Paper 6. Accountability in Local Governments. Joint Annual Review of Decentralization 2004. Ministry of Local Government. Williamson, Tim, Charles Mugerwa, Gregory Smith, Peter Mubaale Kafuko, and Wilson Baryabanoha. 2005. The Republic of Uganda: Local Government Public Finance Management. Draft Report. Center for Aid and Public Expenditure Overseas Development Institute. U.K. World Bank. From Shouting to Counting. A New Frontier in Social Development, Washington D.C. 102 Annex 2: Study on Kerala, India Local Government Discretion and Accountability ABBREVIATIONS AND ACRONYMS DDC District Development Council DPC District Planning Committee NGO Non Government Organization PCDP People's Campaign for Decentralized Planning PRI Panchayati Raj Institutions VTC Volunteer Technical Corps Note: This case study is a product of desk review of literature on India decentralization and local governance. Recognizing the limitations of preparing a case study solely on desk review, the goal is to present a broad outline of the kind of information needed to apply our analytical framework at the country level. The next step is to conduct a fieldwork and a more detailed case study and analyze the local governance conditions based on more specific and primary data using the analytical content of this study. 1.1 BACKGROUND India was under British rule until 1947, when it became independent after about 150 years; it declared itself a republic in 1950. The country's political and administrative structure is largely inherited by the British who introduced local government constituting municipal councils and municipal corporations in the last quarter of the nineteenth century. The British also provided for a federal structure to the country with the Government of India Act in 1935. Self-governing village communities existed in India from the earliest times. These village councils of five persons were known as panchayats and were elevated to a sacred position of authority through custom and religion. These self-contained village communities were replaced by the British by formally constituted village administrations (Mathew and Mathew 2003). Following a struggle of four decades, the government of India passed a series of constitutional reforms designed to democratize and empower India's rural local political bodies- the panchayats in 1993. The 73rd (Panchayats) and 74th (Municipalities) amendments to the Constitution meant a major break-through for decentralization and empowerment at the local level as they recognized a third tier of government at the sub-state level, thereby creating legal conditions for local self-rule (Johnson 2003). These amendments made it mandatory for the state governments to organize local bodies which hold regular elections, and give these local bodies appropriate finances. The Indian government structure today consists of three tiers: the Union, States and Local governments (see figure 1 below). 103 Figure A2.1 Indian Government Structure Source: Mathew,2003 There are currently two distinct types of local government systems in India, the rural local government system and the urban local government system. The rural local government bodies are organized in three tiers, the village panchayat at the village level; the block for about 100 villages and a district council (zilla panchayat) comprising about 1000 villages, with elected bodies at all levels. At the village level, the gram sabha serves as citizens' assembly that constitutes all eligible voters within a gram panchayat area. It meets at least once in every three months. The urban level has similar counterparts called the nagar panchayat at the lowest level followed by the municipal council and the municipal corporation at the highest level. In line with the central government requirement, in 1993, the Kerala authorities introduced legislation to form two layers of local self-government in urban areas, and three layers in rural areas. In the urban areas, the municipal corporations were subdivided into municipalities. In the rural areas, districts were divided into blocks, which in turn were divided into villages. Each of these rural sub-divisions was governed by a panchayat. Since 1995, Kerala has had direct elections to local boards of the gram or village panchayats, the block panchayats which are groups of 2 to 15 villages with development officers and certain federal projects, and the district panchayats for each of the 14 districts in the state. Urban areas fall under the rubric of municipality or urban corporation and represent parallel structures to the gram and block panchayats. In 1996, a coalition of leftist parties led by the Marxist Communist Party of India returned to power and immediately fulfilled one of the most important campaign pledges by launching the `People's Campaign for Decentralized Planning' (PCDP)51. All 1,214 local governments in Kerala, municipalities and the three 511996 decentralization reform in Kerala began with the identification of gaps in local development at the gram sabha level. Development seminars were organized in each village for elected representatives, officials, and others. Panchayat development reports were drafted, and task forces were set up to address specific sectors and provide training. The village panchayat plans were expected to aggregate into the block and then district plans (Mukundan etal, 2004: 230). 104 rural tiers of district, blocks, and gram panchayats, were given new functions and powers over decision-making. The campaign also sought to maximise citizen's direct involvement in planning and budgeting. Going beyond the traditional focus on electoral participation, increased attention was given to direct participation. Initially involving a large campaign to mobilize citizens all over the state, the campaign has moved beyond its mobilization phase and now depends more on active citizens, local politicians, and local officials. The congress- led United Democratic Front government that came into power in 2001 showed far less enthusiasm for decentralization than its predecessor (Mathew 2003). Although in congress, government officially embraced the new institutions and did not alter any of the legislative architecture of the campaign, the campaign's mobilization mode, such as mass training programmes, voluntary technical committees and the movement like State Planning Board all came to an end. Matthew (2003) suggests that the overwhelming electoral victory of the Left Democratic Front in local government elections in 2005 and its return to power at the state level in May 2006 may yet infuse the process of local government democratization with new political vitality. What distinguished Kerala from the other states in India is the "big bang" approach that state took where the strategy was to implement a significant fiscal decentralization program and then "learn by doing." Currently Kerala has has 991 Gram Panchayats, 152 Block Panchayats, 14 District Panchayats, 53 Municipalities and 5 Corporations (http://www.kerala.gov.in/government/localself.htm). 1.1.2 Challenges to Decentralization In Kerala, by and large, the state government that initiated the program has resisted arbitrary and partisan interventions. Also Kerala has high political consciousness and a multitude of active political parties. Compared to other states in India, clientelism and corruption is largely absent. These characteristics have to be further institutionalized to ensure sustainability. Overall, the process of decentralization in Kerala is too recent to predict the main impediments and challenges. 1.2 POLITICAL SETTING AND ACCOUNTABILITY IN DECENTRALIZATION 1.2.1 Discretion Separation of power- In Kerala, the administrative body comprises the Panchayat Secretary, Junior Superintendent, and a Head Clerk with clerks under him/her. The political body comprises the president and a vice president and several committees (Refer to figure 2 for details). The finance and planning committee deals with finance planning etc; development standing committee deals with agriculture, animal husbandry and dairy development, minor irrigation, and small scale industries; welfare standing committee deals with education, fishing, social welfare, welfare of scheduled castes and schedule tribes, poverty eradication, rural development, and public distribution system; public works standing committee deals with housing, water supply, and public works. The president in practice discharges considerable executive responsibility. The panchayat secretary is responsible for financial transactions but does not have any financial powers (further discussed under `local council oversight). A World Bank study (2004) concluded that while in most of India decentralization is characterized by overlapping functions and proliferation of schemes, which lead to reduced lines of accountability to voters and inflated administrative costs, this is not the case in Kerala, where legislation provided for clear demarcation of functions devolved to the local government level, giving less scope for overlap and duplication of efforts and resources and overlap (World Bank 2004). 105 Figure 2 Organization Structure of the Gram Panchayat, Kerala Gram Panchayat Gram Panchayat Executive System Political System Panchayat Secretary* President Junior Superintendent** Vice President Head Clerk Development Finance Welfare Public Works Standing Standing Standing Committee Upper Lower Sanitation Committee Committee Committee Chairman and Division Division Staff and Members Chairman and Chairman and Members Clerks Clerks Members Members *Could be Special Grade-I or Grade-II or Note: Gram panchayats with a population Grade-III Secretaries. Some officers of the up to 10,000 have 10 elected members, one transferred departments have been additional member for every 4,500 of the designated ex-officio secretaries. population with a maximum of 20 members. ** only in Special Grade-I Panchayats World Bank, 2004: 148 Electoral System- The constitution provides for direct elections at the local level: Panchayat members at all levels are elected through direct elections for a period of five years. Affirmative action guarantees one third of seats for women; reservations for Scheduled Castes and Scheduled Tribes are proportional to their populations. Reservations are also made for chairpersons of the local government bodies, following the same guidelines. In addition, state legislature may provide reservations for disadvantaged groups. Local elections for panchayats and urban bodies allow for independent candidates. Party Systems- Parties are allowed except for panchayat elections. Nevertheless, independent candidates for panchayat elections tend to a have a party-affiliation; as a result political parties have played an active role (World Bank 2004). In general, local elections have been described as dynamic multi-party processes (Mathew and Mathew 2003). Strong links between national, state, and local level parties have led to the fact that decentralisation and panchayat issues play an important role in state and national level elections (Mathew and Mathew 2003). At the same time, this has also influenced local level party structures and elections campaigns. Local election campaign in Kerala in 2005 mirrored power struggles between dominant political formations at the state level and left less room for content-related discussions (Krishnakumar 2005). 106 1.2.2 Supply Side Approaches to Improve Downward Accountability Electoral accountability- Every state in India has a State Election Commission, which is in charge of organizing, supervising and overseeing panchayat elections at all levels. Kerala's election rules also stipulate that candidates for positions in panchayats/municipal local governments may be disqualified from contesting elections and members lose their membership if convicted for corruption. Property statements are required of elected members of the panchayats/ municipalities (http://www.kerala.gov.in/government/localself.htm). Citizens in Kerala have the right to recall the gram sabha president/chair person through a "no- confidence motion". One of the key elements of the 1992-93 constitutional amendments has been the affirmative action taken in favour of women and scheduled castes provided for reservation of a third of seats in local government bodies (including chair persons) for women. There is also a quota for members of scheduled castes and scheduled tribes in proportion of their percentage of the population. As an effect of these quotas, the number of women and members of schedules castes and tribes increased radically. The Kerala experience reaffirms the correctness of affirmative action in elections and is supportive of the extension of this principle to higher levels of government. The elected representatives are better educationally qualified than the male representatives (Isaac, 2000:31). Nevertheless, it has been observed that the issue of achieving representation of women's interests and accountability are more complex. As such, many women lacked capacities and political experience to be able to have strong political influence. A culture of intimidation and violence against women play an important role. Political influence is also related to membership in political parties, which partly denied access to women. While the constitution generally provides for direct elections at the local level, mayors in Municipal Corporations in Kerala are chosen through indirect elections by the councillors from among themselves for a term of one year, which is renewable (Fahim 2006). Fahim (2006) concludes that the indirect election of the mayor and the short tenure undermines his/her authority and "renders the role little more than a figure head". The gram sabha, the assemblies of voters in every gram panchayat provide an ideal starting block for the electoral system at the local level. A number of steps have been initiated to strengthen the gram sabhas. The minimum number of legally required meetings has been raised to 4. The quorum has also been raised from 50 to 100 (or 10% of voters). An official coordinator for each gram sabha is to be appointed who would be responsible for keeping the records. In terms of gram sabha's role in electoral accountability, if direct democracy is to be made operational it is important to move towards much smaller assemblies of voters that could be convened with less travel time and where participants know each other well (Isaac, 2000:27). Local Council Oversight- Kerala has a separate administrative and political body at the local level (refer to figure 2), each with separate assigned powers and responsibilities. The authority to incur expenditure, select contractors or suppliers, and issue administrative sanctions rests with the gram panchayat's elected members collectively, and the secretary implements their decisions. But the president, though not involved in the administrative or financial matters, has the responsibility to oversee the efficient functioning of the panchayat bureaucracy, and the secretary reports to the president. Thus the president has executive functions, but is not directly responsible, except as a member of the collective body of the panchayat, for any financial or administrative irregularity. The vice president, who is also the chairperson of the finance standing committee, has the statutory duty to examine the monthly accounts of the panchayat and report to he committee. He is also responsible for the preparation of the budget and for overseeing the function of budgetary control. Local governments do not have the power to hire, fire, and determine the compensation level of their employees (World Bank 2004).The administrative officials are under the supervision and disciplinary 107 authority of the local bodies during their tenure with them but the central government continues to pay the salary, allowances and other dues to the employees and officers transferred to the local bodies (Narayana and Kurup, 2000:9). The district, block, and gram panchayat are given power to write confidential reports on the secretaries. In addition, the administrative staff are required to prepare their monthly work schedule and make a copy available to the elected officials. The elected officials have powers to impose minor punishment such as requesting for the transfer of an official and recommending the suspension (Tharakan, 2002). Thus the system envisages dual control over the staff by the center and the local government 1.2.3 Demand Side Approaches to Improve Downward Accountability Village assemblies of citizens-gram sabhas- are the main mechanisms at the local level for direct citizen participation. In this regard, Kerala has gone further in terms of citizen participation than any other state in India: Gram sabhas, a village assembly of all voters, equivalent to gram sabhas, and at a smaller unit, neighbourhood groups have become active players in the state's panchayati raj system and are now the principle mechanisms for accountability at the local level. There has been a conscious attempt to endow the gram sabha with the character of an institution. The gram sabha has to be convened at least four times a year with a specific quorum of citizens of a panchayat/ municipality/ corporation participating. While citizen participation was high in the beginning of the Kerala campaign, citizens seemed to lose interest in direct participation and very few people attended meetings. Since the mechanism relies on strong citizens' participation, this created serious problems with regard to accountability, leaving space for manipulations of decisions by elected representatives, officials and contractors (Krishnakumar 2004). Also, several cases were reported where panchayat members forget signatures of citizens to ensure that the quorum of the gram sabhas was reached (Krishnakumar 2004). 1.3 ADMINISTRATIVE DISCRETION AND ACCOUNTABILITY IN DECENTRALIZATION 1.3.1 Discretion Ability to make, change, and enforce laws and regulations India's constitution limits legislative power to the union and states. Local Governments in India are subjects of the states. Elected local government bodies in India have no legislative power in the strict sense of the term. While panchayats can adopt resolutions, in Kerala, the state has the right to cancel/suspend panchayat resolutions, dissolve panchayat institutions and call for records and inspection (Mathew and Mathew 2003). In practice, these regulations restrict local government's autonomy, especially if they are vested with the state bureaucracy, and have placed local government institutions in a disadvantageous position, even with regard to mid-level officials of the state government (Mathew and Mathew 2003). In spite of the above constraints, the local bodies have full power to prepare development projects based on the needs of the people with the resources available to them. In 1996, several institutions and posts relating to line departments were transferred to panchayat institutions. These included institutions and posts in the field of agriculture, animal husbandry, primary education, primary health and the welfare of women, children, schedule castes and schedule tribes. The District Planning Committee (DPC) is a constitutional body consisting of 11-15 elected members from the district panchayat, one member from the municipality, and heads of line ministries that approves the plans prepared by panchayats at different levels. The DPC can send back a plan prepared by a local body on the grounds that norms and procedures are not followed. At the same time, the power of DPC is usurped by the District Development Council (DDC), a parallel body created by the government in each 108 district. DDC comprises of the state member of the legislative assembly, heads of the department, and collectors (civil servants). The district panchayat president is not a member. The DDC is supposed to meet every month to review the state-sponsored programs (Tharakan, 2002) Control over Civil Service and Employment Policies While democratic decentralization has gone far in India, the administrative dimension of decentralization is still lagging behind (World Bank 2004): although most state acts have spelled out functions to be devolved to local governments, states have mostly not been able to transfer administrative and technical control over locally administered programs. Some states, including Kerala, took measures to transfer staff but faced strong resistance from unions. Even in cases where transfers have actually taken place, the most important aspects, e.g. power to hire, fire, promote and transfer are still vested with line agencies and panchayats are reduced to merely paying salaries (World Bank 2004). The administrative officials are under the supervision and disciplinary authority of the local bodies during their tenure with them but the central government continues to pay the salary, allowances and other dues to the employees and officers transferred to the local bodies. For example, government doctors, engineers, and agricultural officers at the local level have their compensation determined by the state government but the panchayat committee has some power to initiate disciplinary action (World Bank 2004). Thus the system envisages dual control over the staff by the centre and the local government (Narayana and Kurup, 2000:9). In the health sector, Naranya and Kurup (2000) observed a gap between rhetoric and reality in Kerala: while a government order from 1995 transferred responsibilities for Primary Health Centres and Government Dispensaries to the Village Panchayats and Government Dispensaries to the Villages Panchayats, Blocks and Community Health Centres, it established dual control over staff, shared between the state government and panchayats. Although officials are under the supervision and disciplinary authority of local governments, their cadre conditions remain undisturbed. The government also continues to pay their salary, allowances and other dues to the employees and officials transferred to the local bodies from government (Narayana and Kurup 2000). As a result, reporting lines are blurred and discretion of local government bodies limited. Discretion over Procurement- Instead of implementing public works through contractors, the local bodies give preference to beneficiary committees. When a development work is to be undertaken, those persons and households who will be benefited by the work completion are called for a gram sabha meeting and are explained the work. They are encouraged to form a 10-15 member committee to implement the work. Thus procurement is largely handled by the beneficiaries themselves. Steps have been taken to create a new environment in which a genuine beneficiary committee can effectively function. The idea of the beneficiary committee is that those who will benefit from a project form a committee and execute the work on the basis of design and estimates prepared by the panchayat and approved by an expert committee (explained later). Thus an organization comprising beneficiaries independent of the local government has been set up to be responsible for procurement. The nexus between contractor, engineer, and politician has been broken by this in a large number of local bodies. Delays in the process and lack of expertise led to creating a provision for the non-official engineering experts of the Voluntary Technical Corps (VTC)52 to take up certain functions as well. 52Volunteer Technical Corps (VTCs) consist of retired technical experts and professionals who enrol themselves as volunteers to appraise projects and plans of local bodies. 109 1.3.2 Supply Side Approaches to Improve Downward Accountability Regulatory Decisions- As part the process of transferring responsibilities to the local level, Kerala revised public work rules in order to ensure transparency, make rules more favourable to understand and prevent the misuse of public resources and ultimately ensure administrative accountability. Clear rules aim at ensuring transparency and preventing an increased room for manoeuvre that may lead to corruption and waste. Each item of work is subject to scrutiny by the village assembly (http://www.kerala.gov.in/). All plan documents including those related to beneficiary selection, bills and voucher of works are public documents which any citizen can access. In Kerala, the Ombudsman institution was established to conduct investigations and inquiries in respect of any action involving corruption or maladministration or irregularities in the discharge of administrative functions by or related to the Local Self Government Institutions (See http://www.kerala.gov.in/grievanceredressal/ombudsman.htm for more details). There is also the Appellate Tribunal- a semi judicial body manned by a district judge with powers of hearing appeals from citizens aggrieved by decisions of local governments in exercise of their regulatory power (as different from their development power where the recourse is the ombudsman) like licensing and environmental legislation (Vijayanand, 2005: 5). In addition, there is a kind of internal audit system to assist the local governments in maintaining their accounts and to oversee adherence to the due process. It is not autonomous but has freedom to carry out its activities. It ensures that accountability systems are properly kept and guidelines observed (Vijayanand, 2005: 5). Service Delivery and Procurement- The following steps were taken to ensure accountability of the beneficiary committees described earlier. The first step is the adoption of local market rates for estimation of cost of works so that the beneficiary committees can execute work in a transparent manner, maintaining actual records of purchase and payments. The committee can fix different regional rates within a district on the basis of actual prevalent rates. Secondly, under the old procedure, technical sanction had been issued by department officials on the basis of their delegated powers. Under the new procedures adopted for plans of the local bodies, the technical sanctioning process has been made a function of the block/municipal and district level expert committees. The department officials are the convenors of the subject committee and continued to formally grant technical sanctions. However, now they have to act as members of a committee of peers rather than official in the department hierarchy. Thirdly, the work rules permit only a limited advance to be paid to the beneficiary committee. The rest of the payment is to be claimed as part of bill payments and final bill settlements as the work is completed. For claiming part bill payment, the work executed has to be measured, check- measured, and the bill prepared by official engineers. The delays in these procedures, often deliberately created, make the functioning of beneficiary committees extremely difficult. To overcome these problems, provision has been made for the non-official engineering experts of VTC to undertake the measuring and billing functions, if requested by the local bodies (Isaac, 2000:21). 110 It is true that a number of beneficiary committees are controlled by vested interests53 but also genuine committees face a number of challenges such as non-cooperation of government engineers especially when VTC or elected representatives support is absent (Isaac, 2000:29). 1.3.3 Demand Side Approaches to Improve Downward Accountability As a result of strong citizen movement in Kerala, the quality of services and development, especially in building roads, housing for the poor, and child services is rated higher than most other states. Less improvement has been achieved in areas such as economic development (employment, agricultural support, and irrigation) (Heller, Harilal, Chadhuri 2007). Yet, with an erosion of citizens' interest and participation in gram sabha, the oversight system became less effective, as many decisions, including the designation of beneficiaries and priorities on project priorities, were effectively made by politicians and government officials. They also undertake scrutiny of their own projects. In addition, the public display of cost estimates of projects have become an exception (Krishankumar 2004). Right to information- Kerala ensures comprehensive rights to citizen information, allowing for a strong role of direct citizens participation in administrative oversight: in Kerala, the village assembly, gram sabha, has the right to obtain information concerning (i) the action plan of schemes for the next three months, (ii) the detailed estimates of the proposed works, (iii) the detailed item-wise accounts for every expenditure incurred within gram sabha area, (iv) the rationale of every decision of the panchayat concerning that area, (v) the services the officials will render and the work they are to execute in the next three months, (vi) priorities for preparation of the Five-Year and Annual Plans. Failure to convene the ward sabhas in Kerala is a violation of the provisions of the Kerala Panchayat Raj Act and calls for penal sanctions, including loss of membership of the convenor in case of two consecutive instances of non- compliance. Volunteer Technical Corps- Kerala has VTCs consisting of retired technical experts and professionals who enrol themselves as volunteers to appraise projects and plans of local bodies. Expert Committees are formed by drawing volunteers from the VTC at the district. Each expert committee has a non-official as its chairperson and the block panchayat secretary or officer from the town planning department as convenor. The expert committees are, on the one hand, advisory arms of the District Planning Committees, helping the latter to appraise the plans and projects; and on the other hand, bodies of experts to render technical assistance to the local planners. The committee's tasks are clearly confined to technical and financial appraisal of the projects and providing suggestions for modifications to make them more viable. The District Planning Committees approves the plans on the recommendations of the expert committees. Over time, the expert committees were also given the power of approval of technical sanctions and tender excesses within certain limits. They were also given a role in plan implementation such as approval of revised estimates, settling disputes in measurement, inspection of all works and verification and approval of performance and completion reports of works within certain financial limits (Isaac, 2000:19). Social Audit Committees- Kerala has also set up social audit committees, consisting of respected citizens and professionals at all levels, to conduct independent performance evaluations. In order to conduct social audits, they are required to follow specific rules and procedures. These include approval by and reporting to the gram sabha (Mathew and Mathew 2003). Although it is not formally designed or legally enforced, the social audit appears to mean a view of the administrative system from the perspective of the vast 53 An overall assessment by Pillai etal (2000) found that only about 25% of beneficiary committees function effectively. 111 majority of people (World Bank, 2004: 113). In addition, Kerala also introduced special women's watchdog committees at the gram panchayat and municipal level. These include two nominees from each gram sabha or ward committee, one being a member of a Scheduled Caste or Scheduled Tribe. Such committees have the same rights as social audit committees and may scrutinize costs, estimates, the quantity and quality of materials used in works, adherence to norms in selection, etc. (World Bank 2004). Beneficiary Selection- A major change introduced by the people's campaign related to selection of beneficiaries for the development projects. Most of these initial projects involved distribution of assets (a cow or goat), or financial assistance (a home improvement loan or subsidy), and therefore selecting component of the implementation. Earlier, the dominant tradition was to utilize beneficiary schemes for patronage by local political leaders. Now the gram panchayats are required to provide maximum publicity to the criteria for eligibility and prioritize each project in which beneficiaries are to be selected. Notices listing the projects and the criteria have to be prominently displayed in public places and also printed and circulated. Application forms have to be printed in Malayalam and other languages as appropriate to the area. The list of applicants and the application forms has to be presented to the gram sabha and the eligibility criteria and prioritization criteria has be explained to the members of the gram sabha. In addition to the criteria set by the panchayat, the gram sabha can include sub-criteria for prioritization according to local precepts The panchayat as a whole consolidates the priority list of beneficiaries received from each gram sabha. The priority list has to be created on the basis of clearly stated norms. In no case can the relative priorities from each ward be overturned during the process of consolidation. Members of the public and local press can have access to the proceedings of this final selection. The draft list shall be exhibited prominently and the objections considered and reasons for rejection stated. The selection of beneficiaries for block and district panchayats also is to be made from the gram sabhas using gram panchayats as agents. In the municipalities, ward committees undertake the functions of gram sabhas. In the ward committee meeting for the selection of beneficiaries, all applicants must be invited and prioritization must be done in their presence (Isaac 2000:21). 1.4 FISCAL DISCRETION AND ACCOUNTABILITY IN DECENTRALIZATION 1.4.1 Discretion Expenditure Assignment- In terms of fiscal decentralization, state governments in India are responsible for implementing a higher proportion of the general government budget than in most developing countries (Purfield 2004). Kerala has the greatest degree of local expenditure autonomy and is the most fiscally decentralized state in India (Purfield 2004, World Bank 2004). About 18% of the state expenditure in Kerala takes place at the local government level with 20% spent at the Zilla Parishad level, 18% at the block level and 62% at the gram panchayat level (World Bank 2004). The discretionary proportion of the village panchayats budget increased dramatically, rising from Rs 1,000 million in 1996-97 (the year 54 before the people's campaign) to 4,204 million in 1997-98, and over 5,000 million in each of the three years to follow (Heller, Harilal, Chadhuri 2007). Functions for expenditure discretion have been divided according to the various tiers and in most cases broken into activities to avoid overlapping. Responsibilities are therefore clearer. Gram panchayats functions are broken into three mandatory functions with 27 items; general functions with 14 items; and 5441 Indian Rupees~ 1US Dollar 112 19 sectoral functions subdivided into 74 items. In 1995 only 32% of the expenditures for functions assigned to gram panchayats took place at the discretion of gram panchayats. Over a five-year period, this allocation has changed and the gram panchayat now has control over 68% of the expenditure (World Bank 2004). In terms of coming up with an expenditure plan, for the first time in India the gram panchayats or urban municipalities actually prepare operation plans. Gram sabhas are regarded as an ideal starting block for people's planning. Through discussions in these assemblies, people identify local development problems, analyze the factors responsible and put forward suggestions for possible solutions. After needs identification, an objective assessment is made of the natural and human resources in the locality and a development report is written for each panchayat/municipality. A seminar is organized with delegates from the gram sabha, to discuss the report. Task forces of around ten persons prepare project proposals on the basis of recommendations at the seminar. The chairpersons of the task forces are usually elected representatives and the convenor is usually an officer from a concerned line department. By this time the grant-in-aid for the local bodies is usually announced; so the project prepared by the task forces is prioritised and incorporated into the plan document. Block and district panchayats are supposed to start preparation of their annual plans only after gram panchayats have drafted their plans. The DPCs along with the expert committees of the VTCs evaluate the plan and offer technical input. Finally the funds are allotted to local bodies in specific instalments (Isaac 2000). With regard to service provision, the division of responsibilities between central, state, and local level varies between different sectors. In the education sector, for example, responsibilities gradually became more centralized since 1950, when control over education was primarily vested within the state governments rather than in the federal government. In 1976, constitutional amendments placed education on a concurrent list, making it a shared responsibility of both state and federal governments. However, while giving supremacy over education to the federal governments in the case of conflict, it also allowed for a role for Panchayat Raj institutions at the local level. The constitutional amendments of 1992-93 detailed requirements to devolve planning and management responsibilities to the local level and ensure greater community participation (Mukundan and Bray, 2004). However, Mukundan and Bray (2004) conclude that there is a gap between the rhetoric of decentralization in the educational sector in Kerala, which is not matched with reality, resulting mostly from a lack of capacity and specific technical expertise of planners to with regard to education projects. The volume of expenditures the panchayats incur or manage keeps expanding, as functional devolution expands. Thus the fiscal pressure on the gram panchayats is expanding with the need to increase its revenue generating capacity discussed in the next section. Revenue Assignment- In Kerala, the village panchayats and municipalities and municipal corporations alone has revenue raising powers, and interact by statute with the gram sabha. The other levels of government- block panchayats and district panchayats- which came into existence in 1996- do not have any revenue raising powers and depend almost entirely on grants from the central and state governments. Apart from grants mentioned in the next section, the major sources of revenue in Kerala are own source (tax revenue, non-tax revenue, and debt heads), donations, and borrowed funds (refer to figure 3 for details). Tax efforts comprise a large part of the panchayat's resource mobilization, property tax based on rental value being the largest source. The second largest source of tax revenue for village panchayats is the Profession Tax collected from those employed in vacious professions; in the case of municipalities Entertainment Tax is the second largest source, which in the case of village panchayats is the third largest source. Also the village panchayats and urban local bodies are empowered to levy surcharges on Property 113 Tax (in the case of village panchayats), on any tax other than Professional Tax (in case of urban local bodies). This has not been fully exploited. Sources of non tax revenue include user charges, beneficiary contribution, and donation in cash or in kind (land, material, or labor). In the case of village panchayats nearly one-third of this source is from river sand with other major items being rent from buildings, market fees and license fees. In case of urban local bodies about 30% of non-tax revenue comes from rent. The share of non-tax revenue in the total income of village panchayats is 24%, and in the case of municipalities 37%, and corporations 20%. User charges constitute an insignificant share. Though Kerala has given the gram panchayats, municipalities, and municipal corporations independent revenue raising power, it has not been effectively used. Overall own source revenues are negligible- less than one percent of Gross State Domestic Product. A number of reasons have been suggested for this, including, an unwillingness of the local officials to enforce the tax laws; the limited capacity of the local government officials to effectively administer a revenue system; weak administration procedures such as tax rolls not maintained etc (World Bank 2004:8). Thus in spite of financial discretion provided at the local level, the local governments have depended on transfers as the main source of revenue. At the same time the per capita own source revenue has been increasing since 1994 in spite of large transfers (discussed in the next section).Within own source revenues, non tax revenues have been much more buoyant than tax revenues. Per capita amounts have increased from Rs. 7 in 1994 to Rs. 29 in 1999 (World Bank 2004). Overall, the progress in revenue decentralization does not match the advance in expenditure decentralization. Some policy measures such as reviewing tax assignments, imposing or increasing user charges, and stepping up measures to augment tax efforts- have to be addressed (World Bank 2004: 156). Intergovernmental Transfers- The Kerala local bodies receive funding from the center and state through various channels (refer figure 3). The Kerala government grants to local bodies consist of non-plan and plan categories. Non-plan grants are broadly divided into statutory and nonstatutory grants. Statutory grants are the largest of these two components and consist of assigned taxes and vehicle tax compensation. The assigned tax consists of a basic tax on land and a surcharge (4 percent) on stamp duty collected under the provisions of the Kerala Stamp Act, 1959, on the value of the property transacted. The entire amount is meant to be transferred to panchayats after deducting collection charges. Besides these taxes, 20 percent of the Motor Vehicles Tax's net collection, also called Vehicle Tax Compensation, is to be distributed to gram panchayats on the basis of the length and types of roads in a panchayat. Statutory grants accounted for about 13 percent of all transfers to gram panchayats in 1999, and there is local government discretion as to the expenditure of these funds. Plan grants are the most important and were the preferred instrument to devolve more resources to panchayats in the "big-bang" decentralization. They are the single largest component (about 57-63%) of the total expenditure of a gram panchayat (World Bank 2004: 120). The per capita state plan grant has increased from Rs. 12 in 1993-94 to Rs. 207 in 1998- 99, an increase of about 15 times in a period of six years. This corresponds to an increase from 3 percent to about 12 percent of state revenue. 55 Apart from plan and non-plan grants in the Kerala system, the local bodies receive grants through sponsored schemes where the local bodies act as agents of the central and state governments (see figure 3). For the sponsored schemes broad guidelines are issued by the sponsoring government or department. 55This increase is probably overestimated though since panchayats have only been receiving a share of their entitlements. 114 Even so, the local bodies decide the location of the project and select the beneficiaries, besides taking up the responsibility for the implementation of the project. Figure 3 Source of Finances for Gram Panchayats in Kerala World Bank 2004:115 An important feature of financial devolution in Kerala is the comprehensive area plan (as described earlier) prepared by each local body to claim grants in aid. The plan outlay is not confined to the grant in aid amount (though the plan grant is usually the largest source of its funding). The grant in aid is integrated with different state and centrally sponsored schemes, and own revenue surplus of the local bodies, loans from financial institutions, beneficiary contributions, voluntary labor and donations and other sources. Local Borrowing- State borrowing is facilitated by India's comparatively liberal borrowing regime (Purfield 2004). However, external borrowing is prohibited for sub-national governments. Approval for domestic borrowing by the central government is required. At the local level, in Kerala, a lack of access to credit has been identified as a major challenge for the success of the citizens' campaign (Heller 2000). 1.4.2 Supply Side Approaches to Improve Downward Accountability According to a World Bank study (2004), a consolidation of revenues and expenditures does not take place, despite of a requirement for panchayats, stipulated by the Kerala Panchayat Raj Act (1994), to prepare annual statements of all receipts and expenditures. In Kerala, the control mechanisms of the state government over financial management in the panchayats appear to be relaxed. The local government accounting system is in danger of collapse and annual audits have fallen in arrears. It is a cash-based accounting system. The registers and books of accounts are outdated and unable to account for the 115 increased and diversified flow of resources of the present decentralized system. Moreover, there are no uniform accounting codes or rules for panchayats. Gram panchayats still rely on the old panchayat accounting rules, while block and zilla parishads do not have designated accounting rules. As a result of this state of affairs, ongoing monitoring of financial management is difficult. Thus the panchayats do not capture fully the extent of the financial resources they handle. Budgeting seems to have been reduced to a compliance formality. There is little use of the budget for fiscal planning. There is no linkage between the annual plan and the budget, making monitoring and evaluation difficult. There is a lack of reliable information about the assets required to be maintained by the panchayats. It is difficult to determine unit costs of projects and various services like water supply, street lighting, sanitation, and so on. Also, the economic viability of proposed projects is unclear. State Finance Commissions have had to assess the financial requirements of the PRIs by informed guesswork more than from data. In Kerala, old systems are under stress and new ones have not yet emerged to take their place. Since 1996-97, panchayat funds have tripled and the state government's grants to gram panchayats have increased ninefold. Despite this substantial increase in the flow of resources, financial management is still led by accounting rules that were framed more than 40 years ago. In addition, the state government has complicated the situation by issuing numerous regulations for the use of these funds. The result is a complicated accounting system, and one that is yet to be computerized. Consequences in Kerala of these inadequacies are also serious. Although panchayats are required to prepare an annual statement of all receipts and expenditures, consolidation of accounts does not take place in any of the three tiers of rural local government. Gram panchayats prepare an Annual Financial Statement, but it does not include Plan funds or funds received for Centrally Sponsored Schemes. Block and zilla parishads do not prepare an Annual Financial Statement. In the absence of consolidated accounts and financial statements, it is difficult even for the panchayat committee (not to mention the general public) to know the extent of receipts and expenditures as well as the financial position of the local body. 1.4.3 Demand Side Approaches to Improve Downward Accountability At the village level, the gram sabha, which constitutes all eligible voters within a gram panchayat area, is meant to serve as a principle mechanism for transparency and accountability. Among its principle functions are to review the annual statement of accounts, review reports of the preceding year, review and submit views on development programmes for the following year, and participate in the identification of beneficiaries for some government schemes (Johnson 2003). The process of direct citizens' participation in the annual planning and budgeting cycle essentially consists of four stages: (1) Ward Sabhas (open meetings, held on holidays, in public buildings) are held at the ward level to identify local development problems, generate priorities, and form sub-sector Development Seminars in which specific proposals take shape. (2) Development Seminars develop integrated solutions for various problems identified by Ward Sabhas. The development Seminars are required to produce a planning document. (3) Task Forces are selected in Development Seminars and charged with converting the broad solutions of the seminars into project/scheme proposals to be integrated into the final panchayat plan. (4) Actual formulation of the panchayat or municipal budget, as the fourth step of the annual planning exercise (Heller, Harilal, Chadhuri 2007). As such, citizens participate in all major steps of policy-making, planning and budgeting, going beyond civic engagement in other states in the country. 116 In addition, the Freedom of Information Act ensures citizens' rights to information: on payment of a small fee, every citizen can demand and receive details of expenditure on the work done over the last five years in his/her village. Its also stipulates that holding back information regarding the administration of the local self-government bodies to citizens is punishable. The "Every Citizen has the Right to Information Campaign" was initially piloted by an NGO in Rajasthan in 1995. Implemented in practice only after social pressure, the initiative was later introduced in other states all over the country, such as Kerala, and finally incorporated into central legislation in 2003 (Mathew 2003). 117 REFERENCES Blair, Harry (2000): Participation and Accountability at the Periphery: Democratic Local Governance in Six Countries, World Development Vol. 28, No.1, pp. 21-39, 2000 Fahim, Mayraj (2006): Local government in India still carries characteristics of its colonial heritage, in CityMayors Government, (online source: http://www.citymayors.com/ government/india_government.html) Heller, Patrick (2001): Moving the state: The Politics on Democratic Decentralization in Kerala, South Africa and Porto Allegre, Politics & Society, Vol. 29 No. 1, March 2001 131-163 Heller, Patrick/ Harilal, K.N. and Chaudri, Shubham (2007): Building Local Democracy: Evaluating the Impact of Decentralization in Kerala, India, World Development ­Articles in Press doi:10.1016/j.worlddev.2006.07.001 Isaac, T.M. Thomas (2000): Campaign for Democratic Decentralization in Kerala. An Assessment from the Perspective of Empowered Deliberative Democracy, Working Paper, Centre for development Studies and Kerala State Planning Board, 12 January 2000 Jayal, Niraja Gopal (2007): Unity in Diversity: Learning from Each Other. The Indian Perspective, Conference Paper, The Fourth International Conference on Federalism, Pre-Conference Event, February 2-23, 2007 Johnson, Craig (2003): Decentralization in India: Poverty, Politics and Panchayati Raj, Overseas Development Institute, Working Paper 199 Khemani, Stuti (2001): Decentralization and Accountability. Are voters More Vigilant in Local than in National Elections?, Policy Research Working Paper 2557, The World Bank Development Research Group, Public Economics February 2001 Krishnakumar, R. (2004): Derailing Decentralisation, in: Frontline, Volume 21-Issue 26, Dec. 18 ­31, 2004 Krishnakumar, R. (2005): Another Round for Kerala, in: Frontline, Volume 22-Issue 20, Sep. 24-October 07, 2005 Mathew, George with Anand Mathew (2003): India: Decentralization and Local Governance. How Clientelism and Accountability work, in: Hadenius, Axel (2003), Decentralization and Democratic Governance. Experiences from India, Bolivia and South Africa, (online source: http://www.egdi.gov.se/pdf/study/study2003_3.pdf) Matthew, George (2003): The Crisis of Governance in South Asia: Assessing Decentralization's Promise in India, Seminar paper, (http://action.web.ca/home/sap/south_asia_resources.shtml?x=48519&AA_EX_Session=29c8f43 869580c02e3e868b529e8ffed) Mukundan, Mullikottu-Veetil /Bray, Mark (2004): The Decentralisation of Education in Kerala State, India: Rhetoric and Reality, International review of Education 50:223-243, 2004 Narayana, D./Hari Kurup, K.K. (2000): Decentralisation of the Health Care Sector in Kerala: Some Issues, Seminar presented on 3 December 1999, Centre for Development Studies Thiruvananthapuram, online source: http://www.cds.edu/download_files/wp298.pdf 118 Narayan, Jayaprakash (2003): Decentralization, Voting and Public Good, in: India Together, (http://indiatogether.org/2003/apr/gov-votedecnt.htm) Pillai, A.R.V, N. Jayaprakash, and Ajith Bhaskaran. 2000. Beneficiary Committees. An Experiment under People's Planning. Paper presented at the International Seminar on Democratic Decentralization. Thiruvananthapuram. May 23-27, 2000. Purfield, Catriona (2004): The Decentralization Dilemma in India, IMF Working Paper WP/04/32, Washinton, D.C. Singh, Nirvikar (2007): Fiscal Federalism and Decentralization in India, MPRA Papers No. 1447, (online source: http://mpra.ub.uni-muenchen.de) Tharakan, Michael, P.K. 2002. Political Decentralization and Accountability: Case of Kerala. World Bank Workshop Paper. Trivandrum, Kerala. Veron, Rene et al. (2006): Decentralized Corruption or Corrupt Decentralization? Community Monitoring of Poverty Alleviation Schemes in Eastern India, World Development Vol. 34, No.11. pp.1922- 1941, 2006 Vijayanand, S.M. (2005): Social Accountability and Participatory Planning-Lessons from the Kerala Experience, Note prepared for the Workshop on "Application of Social Accountability Mechanisms in Community-Driven and Decentralization Programmes in South Asia at Hyderabad, 16th to 19th May 2005, organised by World Bank and Centre for good Governance, online source: http://www.sasanet.org/documents/Case%20Studies/Participatory%20Planning%20Kerala%20Ca se%20Study.pdf World Bank Studies (2004): Fiscal Decentralization to Rural Governments in India, Washington, D.C., (online source: http://siteresources.worldbank.org/EXTSAREGTOPDECENTRALIZATION/ Resources/496899-1095189822590/521764-1095190669879/VolumeI.pdf) World Bank (2005): State Fiscal Reforms in India. Progress and Prospects. Washington, D.C. 119 Annex 3: Study on Philippines Local Government Discretion and Accountability ABBREVIATIONS AND ACRONYMS AIP Annual Investment Plan ARMM Autonomous Region of Muslim Mindanao BAC Bid and Award Committee BBGC Barangay-Bayan Governance Consortium BOT Build-Operate-Transfer CCAGG Concerned Citizens of Abra for Good Government CDD Community Driven Development CDF Countrywide Development Funds CIA Congressional Insertion Allocation COA Commission on Audit COMELEC Commission on Election CSO Civil Society Organization DepEd Department of Education DILG Department of the Interior and Local Government DOH Department of Health DPWH Department of Public Works and Highways GPRA Government Procurement Reform Act IRA Internal Revenue Allotment KALAHI-CIDSS The Kapitbisig Laban sa Kahirapan (Linking Arms against Poverty)-- Comprehensive and Integrated Delivery of Social Services LGC Local Government Code LGU Local Government Unit NEDA National Economic Development Authority NGO Non-Government Organization PET Public Expenditure Tracking PO People's Organization PCGC Presidential Commission on Graft and Corruption PRA Preparatory Recall Assembly OSR Own Source Revenues 120 Note: This case study is a product of desk review of literature on Philippines decentralization and local governance. Recognizing the limitations of preparing a case study solely on desk review, the goal is to present a broad outline of the kind of information needed to apply our analytical framework at the country level. The next step is to conduct a fieldwork and a more detailed case study and analyze the local governance conditions based on more specific and primary data using the analytical content of this study. 1.1 BACKGROUND Philippines has a unitary state system with complex multi-tiered structures. It is a presidential republic with a bi-cameral legislature, whose central government is divided into 15 administrative regions and one autonomous region, the Autonomous Region of Muslim Mindanao (ARMM). Philippines has a clearly- defined formal decentralization framework embodied in the 1987 Constitution where autonomy of local governments is enshrined as a state principle. The 1991 Local Government Code (LGC) provides a framework for increased local autonomy with its far-reaching provisions affecting assignment of functions across different levels of governance, revenue sharing between central and local governments, resource generation and utilization authorities of Local Government Units (LGUs)56, and participation of civil society in governance processes. The 1989 Organic Act for Muslim Mindanao provides for an expanded share and the automatic retention of national internal revenue collected in the Autonomous Region of Muslim Mindanao, along with significant discretion in development planning, primary role in delivering basic services and management of natural resources in the region. In general, local government in the Philippines is composed of three layers, each level of which is autonomous (although central government institutions exercise some degree of supervision in terms of budgeting and legislation): (1) provinces and independent cities; (2) municipalities and component cities; and (3) barangays. There are 79 provincial governments, with 117 cities and 1500 municipalities, with 41,939 barangays (Stevens 2006). 1.1.2 Challenges to Decentralization and Coordination The Philippines decentralized electoral system predates the enactment of the LGC and finds its roots more than a century ago, in the colonial American project of self-government that followed a strategy of benevolent assimilation (Hutchcroft 2003). The timing, phasing and structural design of colonial electoral politics in the Philippines left indelible imprints on the character of Philippine polity facilitating: "the emergence of local 'bosses' whose constituencies remained trapped in webs of dependence and insecurity, and whose discretion over state resources, personnel, and regulatory powers provided enormous opportunities for private capital accumulation." (Sidel and Hedman 2001:7) The concentration of power on strong local families thus has a deep-rooted historical basis and makes local elections less competitive. It also probably explains why the constitutional provision against political dynasties remains only as a state principle but not enacted into an implementable law. After the historic decentralization Act in 1991, there were partially successful attempts immediately to re- centralize authority- due to political incentives. The president's incentive was that traditionally presidents have formal authority to release pork barrel funds57 that allow them to buy support of legislators. 56 In this paper LGU refers to all levels of sub-national government- provinces, cities, municipalities, and barangays. 57 `Pork Barrel' refers to local infrastructure projects in legislators' home districts for which they are able to secure financing from the central government. Two major budgetary vehicles for pork barrel projects in the Philippines are Countrywide 121 Therefore, the effectiveness of president's control would have been seriously undermined with broad devolution of revenues and expenditures. The legislature's disincentive was that local executives with autonomous fiscal resources would be much more threatening as opponents, should legislators choose to contest local offices in the future. Also, local executives could use their enhanced powers to contest incumbent legislators for national office (Eaton 2001). The opposing reasons to decentralize were that President Aquino58 was committed to `no re-election clause'. She wanted decentralization to be the legacy of her government (Eaton 2001). Many national legislators maintained solid and strategic alliances with sub-national politicians. For example, when relatives of national legislators occupy local offices, as they often do in the Philippines, the devolution of revenue and expenditure authority does not substantially threaten the political interests of legislators (Eaton 2001). Reflecting the above ambivalence of many legislators, the Local Government Code sat in Congress for more than four years (1997-1991) and a compromise code-LGC was created. 1.2 POLITICAL SETTING AND ACCOUNTABILITY IN DECENTRALIZATION 1.2.1 Discretion Separation of Powers- Each LGU layer is headed by an elected chief executive (governor, mayor, barangay captain) and has an elected legislative body/Sanggunian (vice governor/vice-mayor and council members) (Manasan, 2004:1-3) (Refer to figure 1 and 2 for details of executive and legislative structures). All local government officials are elected every three years with three-term limits, except for barangay officials, who are elected every five years. There are constitutional safeguard against arbitrary dismissal of local government. Development Funds (CDF) and Congressional Insertion Allocations (CIA). CDF monies are distributed evenly among all legislators, who received the right to identify local projects eligible for the funds up to a certain amount. CIA funds are not distributed evenly among all legislators but instead are controlled by members of the Appropriations Committees of each chamber who, in budget deliberations, support higher budgets for bureaucrats in exchange for a say in how these extra funds are spent in their districts (Eaton 2001). 58 President of the Philippines from 1986 to 1992 122 Figure 1 Municipal/City Government Structure Sanggunian Bayan Municipal Mayor SB Secretary Administrative Staff Office of the Mayor Office of Municipal Office of the Municipal Office of the Municipal (Municipal Administrator) Planning and Treasurer Assessor D l t Economic Enterprises Civil Security Administrative Staff Administrative Staff Management Unit Permist and Public Affairs, Plan & Programs Accounting Licenses Information & Special Personnel Research, Evaluation Cash Services & Statistics General Services Interior Control Civil Registration Revenues 123 Figure 2 Provincial Government Structure Sangguniang Provincial Governor Panlalawigan Office of the Provincial Governor Provincial Secretary Civil Security Unit Administrative Staff Ordinance & Journals & General Personnel Internal Legal Resolution Minutes Service Division Control Services Provincial Budget Officer Provincial Public Affairs Special Provincial Administrative Staff Information Services Jail Office of the Provincial Planning & Provincial Municipal Budget Budgets Administrative Staff Office of the Provincial Treasurer Plans & Special Provincial Field Supervision Administrative Program Services Jail Accounting Cash Revenue Division Division Division Office of the Provincial Engineer Planning, Designing & Progressive Quality Control Construction & Motor Pool Administrative Division Maintenance Division Division Source: Adapted from The Governor's Handbook, League of Provinces in the Philippines (1994) 124 Electoral System- One of the strongest points of political decentralization in the Philippines is the extent of direct local elections. The current electoral system--described as "first past the post" where winners are those who garner the most number of votes--involves the direct elections of all elective local government officials. De jure The Local Government Code 1991 does not have quotas for women and minorities. As a result women hold only about 10 percent of elected local offices (Blair 2000, Iszatt 2002). The LGC mandates the president with the power to issue administrative suspension orders on governors and mayors of component-cities, based on loosely-worded grounds including "disloyalty to the constitution", "dishonesty, oppression, misconduct in office, gross negligence and dereliction of duty", "moral turpitude" and "abuse of authority". This power is open to question and abuse. For example, the Office of the President slapped preventive suspensions orders, a few weeks before the official start of the campaign period in the May 2007 elections, on a number of elected local government officials--all of whom either have been vocal of their opposition to the president or have local political opponents with auspicious ties to the Palace (Cuarteros, 2006) The discretionary powers of the president are mirrored in the powers held by the local chief executives. Governors and mayors likewise hold the power to suspend mayors and barangay captains, respectively, directly under their jurisdiction and line veto powers in the budget. Party System- One hallmark of Philippine political life, resulting partly from the early consolidation of local political power, is the nature of parties as "electoral machineries" rather than platform or ideology- based political organizations. The early consolidation of local political power pre-figures what Sidel and Hedman (2001:7) describe as "a pattern of political competition in which local, particularistic, patronage- based concerns and networks would serve as the building blocks of electoral competition". Hutchcroft (2001) argues that national electoral politics has been shaped into one that works through patron-client networks from barangay village leaders to municipal mayors, provincial governors, congressmen and president. In these networks, central-local relations are marked by the delivery of local votes during elections in exchange for clientelistic largesse once state power is secured by the elite. These networks took the place of political parties as electoral-organizational base and policy platforms as currency for political exchange. Says Hutchcroft (2003:7), "Local leaders delivered blocs of votes in exchange for benefits from allies in Manila, while `national' politics itself was often dominated by the need of congress persons to consolidate local bailiwicks (through such means as rampant pork-barrel spending)". This explains the lack of rootedness of party politics in the Philippines. While the Philippines can be nominally described as having a multi-party system--the term "party" takes on a meaning specific to Philippine politics. Hutchcroft and Rocamora (2003:262) note that a political party in the Philippines is "a very different animal" from the conventional Western definition of "an organized group of persons pursuing the same ideology, political ideas or platforms". They quote David (2001: 24-25 in Hutchcroft and Rocamora 2003:262) describing political parties as "nothing more than the tools used by the elites in a personalistic system of political contests." Political parties tend to be extensions of politicians who control them; with loyalties extended to individual personalities. Party switching is rampant, driven by calculations of personal gain and regional advantage. 1.2.2 Supply Side Approaches to Improve Downward Accountability Electoral Accountability- The constitution's article VI, on the legislative department, specifies that legislative power is not solely vested on Congress, composed of a Senate and a House of Representatives, 125 but also to "the people" through the provision on initiative and referendum. There is provision for popular recall of local officials. The LGC provides for local sectoral representation in the local Sanggunian at all local levels of government, where three seats are reserved for representatives of marginal sectors: one for women, one for labor and the third from a sector to be determined by the Sanggunian. As mentioned earlier, no enabling law has been passed to enact this provision. The allowing of independent candidates in elections and the limit to LGU terms (3 years for locally elected municipal, city and provincial officials, 5 years for barangay officials) hopefully acts as additional accountability mechanisms. Local Council Oversight Popularly elected local legislative councils operate at the barangay (Sanggunian Barangay), municipal/city (Sanggunian Bayan) and provicial (Sanggunian Panlalawigan) performing a number of oversight and check and balance functions. In terms of budget, all the local councils perform budget oversight functions through the authority vested upon them by the LGC to review and approve annual and supplemental budgets and all fund appropriations for government programs, projects and activities implemented within their respective jurisdictions. The Sanggunian Bayan also reviews barangay budgets while Sanggunian Panlalawigan review municipal budgets. However, the capacity of these bodies to perform these functions has been open to questions and scrutiny. Local Councils are supposed to oversee monthly financial reports but it is unlikely that they exercise effective oversight of budgets and expenditures by their governments and subordinate LGUs, since few council members appear to understand the budgetary and expenditure figures they are provided (Azfar etal 2001: 49). In terms of procurement, the Sanggunian Bayan exercises oversight of procurement processes to the extent that the LGC requires its authorization for all municipal or city government contracts entered into. In terms of personnel management, the Sanggunian Bayan determines the powers and duties of all officials and employees of the municipality--including the positions, salaries and other pecuniary benefits of officials and employees--and authorizes the payment of compensation to qualified personnel taking up temporary vacancy positions. The Sanggunian determine the members of the personnel selection board, which the LGC mandates to be established in every province, city and municipality to assist the local chief executive in the judicious selection and promotion of personnel as well as in the formulation of policies designed to look after the local bureaucracy's welfare. 1.2.3 Demand side Approaches to Improve Downward Accountability Philippines has an active NGO community that has played a crucial role in elections. NGO advocacy for alternative politics, particularly issue-oriented electoral politics over personality, continues to gain momentum. Voters' education reached a high point in the 1998 elections when a coalition of 16 Cebu City NGOs spearheaded the political education of urban poor groups on the mechanisms of the party-list system. Although local elections in many countries, including high income ones, do not always elicit heavy turnouts, a recent review of participatory democracy in Cebu City, Philippines suggests that voter turnout can be substantial with 83.3 percent of the eligible voters participating in the 1998 mayoral elections and approximately two-thirds of the eligible voters in both 1988 and 1992 (Etemadi, 2000). The article suggests that this high level of participation was due primarily to the organizing efforts of local NGOs which helped organize the poor and get several of the mayoral candidates to agree to a program that would allow the poor to participate in local government decision-making. 126 Generic Legislation that Empowers Citizens to Demand Local Accountability- Chapter II, Sections 120 to 127 of the LGC enable direct legislation. As mentioned earlier, initiative may be exercised by citizens (i.e. 1000 registered voter in provinces and cities, 100 in municipalities, or 50 in barangays) by filing a petition with the local Sanggunian proposing adoption, enactment, repeal or amendment of an ordinance. Registered voters may also directly approve, amend or reject any ordinance enacted by the Sanggunian through a local referendum, which the LGC mandates shall be held under the control and direction of the Commission on Elections (COMELEC) within sixty days in case of provinces and cities, forty-five days in case of municipalities and thirty days in case of barangays. Chapter 5, Sections 69 to 75 provide the means for the people to terminate tenure by popular vote through two modes of recall: (i) recall initiated by registered voters through a petition signed by at least 25 percent of the voters and filed with COMELEC; and (ii) recall initiated through a Preparatory Recall Assembly (PRA), comprised of elective official in a given locality, with the majority of the PRA adopting a resolution for the recall of provincial, city or municipal officials. Romero (1997 in Iszatt 2004) finds that there have been no strides occurring at the local level in terms of referendum and recall. He finds the implementing procedures cumbersome. Iszatt (2004) observes that the recall provision has been largely unoperationalized by NGOs and POs and articulates apprehension that the provision may be abused by local particularistic interests because of the loose grounds on which a recall may be initiated and the relative ease of manipulating the process because of the provision related to PRA (which is not direct recall in the strictest sense). Specific Bodies and Processes for Citizen Oversight- The law on devolution in the Philippines provides for people's participation in Local Development Councils (LDCs). The LDCs are chaired by the local chief executive and the LGC requires that one-fourth of its membership is constituted by civil society representatives. They are to be established in all barangays (Barangay Development Council), provinces (Provincial Development Council), cities (City Development Council) and municipalities (Municipal Development Council). The primary tasks of these councils relate to the formulation of development plans and public investment programs, which are to be approved by the relevant Sanggunian and on which local budgets are based. The LDCs also formulate the Annual Investment Plan, which is the basis for allocating use of the local development fund (20 percent of the IRA). The development plans articulate the major development concerns and priorities, development vision, goals and strategies, projected revenues and expenditures, public investments and zoning plans for three to ten year of a given local government unit. Chua (2001:48) explains the importance of development plans in various phases of the budget cycle: "During budget preparation, it is used to determine expenditure and sectoral ceilings formulate the functions and project activities and determine cost estimates. During budget review, it is used to verify consistency of the budget with approved activities, goals and objectivesw. During the budget execution, the plas is the basis for determining activtieis to be undertaken during the period. In the budget accountability phase, it sets the standards against which the performance can be measured." The Department of the Interior and Local Government (DILG)59 provides technical support for the operationalization of LDCs. For example, it provides guidelines that LGUs can adopt to accredit civil society organizations participating in the LDCs and other local speci bodies. 59The DILG is a department of the executive branch that assists the president in the exercise of general supervision of local governments, rendering advise on the promulgation of policies, rules, regulations and other issuances related to local government supervision and public order and safety. Among its key functions include designing 127 Creating a Political Culture for Citizen Oversight through CDD operations- The Kapitbisig Laban sa Kahirapan (Linking Arms against Poverty)--Comprehensive and Integrated Delivery of Social Services (KALAHI-CIDSS) is the state's flagship antipoverty program and is supported by the World Bank (Arroyo-Sirker 2005). The project is creating a cadre of leaders who will demand projects from their local politicians. 1.3 ADMINISTRATIVE DISCRETION AND ACCOUNTABILITY IN DECENTRALIZATION 1.3.1 Discretion Ability to make, change, and enforce laws and regulations- The LGC devolves to local government units the responsibility for the delivery of various aspects of basic services that earlier were the responsibility of the national government. These basic services include the following: health (field health and hospital services and other tertiary services); social services (social welfare services); environment (community based forestry projects), agriculture (agricultural extension and on-site research); public works (funded by local funds); education (school building program); tourism (facilities, promotion and development); tele- communications services and housing projects (for provinces and cities); and other services such as investment support (Brillantes, Moscare 2002, Pimentel 1993). Under the LGC, the provinces administer tertiary health services (e.g., hospitals) and are involved in social welfare services and infrastructure provision. Whereas provinces are envisioned in the Code as "dynamic mechanism[s] for developmental processes and effective governance" within their component local governments, municipalities are expected to be the primary general-purpose units of government and the delivery points for most basic public services. Municipalities have responsibility for primary health care, disease control, purchase of supplies and equipment necessary for this, as well as municipal health facility and school buildings. Cities have essentially the equivalent of the combined authority of provinces and municipalities, and have only barangays as their component LGUs. The barangay, the lowest formal level of government, is described in the Code as the "primary planning and implementing unit of government policies..." In practice, the barangays have little policymaking or planning capacity, although they have significant fiscal resources in comparison to their responsibilities. The LGC also devolves to local governments the responsibility for the enforcement of certain regulatory powers, such as the reclassification of agricultural lands; enforcement of environmental laws; inspection of food products and quarantine; enforcement of national building code; operation of tricycles; processing and approval of subdivision plans; and establishment of cockpits and holding of cockfights. The LGC empowers the Sanggunian Bayan to grant franchises and issue permits or licenses for promoting the general welfare of inhabitants of a locality, including: fees and charges for all services rendered by the municipal government to private entitities; regulate business, occupation or practice of profession which does not require government examination; prescribe the terms and conditions under which public utilities owned by the municipality shall be operated by the municipal government or leased to private persons or entities, preferably cooperatives; regulate the display of and fix the license fees for signs, signboards, or billboards at the place or places where the profession or business advertised thereby is, in whole or in and implementing capacity-building programs that would strengthen the administrative, technical and fiscal capabilities of local government offices and personnel. It also organizes, trains and equips the Philippine police force. 128 part, conducted; and grant a franchise to any person, partnership, corporation, or cooperative to establish, construct, operate and maintain ferries, wharves, markets or slaughterhouses, or such other similar activities within the municipality, with preference for cooperatives. Discretion over civil service and employment policies- In the LGUs, the autonomy in pay scale determination is hindered by legal provisions that tie local wage scales to the Salary Standardization Law- -and in the case of health workers, the Magna Carta for Health Workers. Devolution of personnel gave rise to salary differentials between devolved and other LGU staff. The centrally mandated pay and grading structures for permanent LGU positions have created significant fiscal burdens for LGUs. This has forced LGUs to realign wage rates for other staff and encouraged them to circumvent salary and hiring constraints through the use of temporary personnel (World Bank 2004: iii) In some cases, the transfer of personnel has given rise to excess staff for the same position (e.g. each province maintains a provincial agriculturist, but also has absorbed personnel from the provincial agricultural office of the Department of Agriculture) (World Bank 2004: 32). Local governments have found ways to circumvent these policies. For example, in the case of police personnel and school principals, in practice the central department submits a list of names (candidates must be local) to the mayor, who expresses a preference. The favored candidate "pays respects" to the local politicians, and the governor then endorses the appointment. In the case of mandatory devolved positions, local governments appear to face no real consequences if they fail to make the requisite appointments. Opting not to fill such positions enables LGUs in essence to create fiscal surpluses to be deployed elsewhere. In some cases, local governments hire temporary "consultants" to fill civil service needs quickly and with minimal bureaucratic intervention (Azfar etal 2001: 54). Discretion over procurement process- The enactment of the Government Procurement Reform Act (GPRA) in 2002 substantially improved the legal framework for procurement. The LGC was passed at a time when no such unified framework was available and procurement was governed by 60 laws, executive orders, presidential decrees and different issuances from line agencies. This led to much confusion, as each autonomous local government unit adopted its own rules in the procurement of infrastructure projects. Important features of the GPRA and its implementing rules and guidelines are: (1) reorganizing and strengthening agency and LGU Bid and Award Committee (BAC)60 and procurement units to ensure accountability in procurement processes, including prohibiting the local chief executive to be the Chairman of the BAC; (2) minimizing anomalous procurement practices such as rigged bidding, fake advertising notices, misrepresentations in bid proposals, irregularities in bid evaluation, inspection and contract implementation; (3) strengthening system of rewards and punishments in the performance of the procurement function; (4) establishing an appropriate complaints mechanism; and (5) strengthening the involvement of civil society organizations in policy formulation and reform implementation (WB 2005:33-34). 1.3.2 Supply Side Approaches to Improve Downward Accountability 60 Local Pre-Qualification Bids and Awards Committee is the deciding authority on bids and awards for local infrastructure projects, although its jurisdiction does not include procurement of supplies, purchase of office equipment, renovations, and other local government unit disbursements. Two representatives of NGOs and POs seated in the LDC are allocated seats in the committee. Also a certified public accountant from the private sector designated by the local chapter of the Philippines Institute of Certified Public Accountants. 129 Regulatory decisions- The Presidential Commission on Graft and Corruption (PCGC) exposes cases of abuse publicly, but it is not independent, has no prosecutorial authority, and deals only with Presidential appointees (Assistant Regional Director or higher). Another such body is the Ombudsman, which does have independence and prosecutorial authority. However, a recent Supreme Court decision limited its prosecutorial jurisdiction to higher-level officials, thereby giving it authority solely for prosecuting lower officials to the Department of Justice, a cabinet ministry that does not have political independence (Azfar etal 2001:30). All cases against government officials are prosecuted and tried by the Ombudsman. The Supreme Court has exhibited relative independence and in practice been the "institution of last resort" for the interpretation of laws and constitutional provisions that protect civil and political rights (Martinez, 2007). The DILG, as mentioned earlier, performs the presidential function of supervising the operations of local governments. It currently does not perform any prosecutorial functions. The LGC empowers the president to undertake remedial measures when Governors and City Mayors act in error, a function delegated to DILG. At the same time, Also DILG and other central government bodies and officials continue to view sub-national governments as dependent administrative units rather than independent local governments (Trujillo 2000). The Regional Development Council is convened by the National Economic Development Authority (NEDA) as part of its national and regional planning mandate, and includes regional representatives of each major national department, the provincial governors and the city mayors in the region. The provincial and city LDCs need to relate directly with the Regional Development Councils to integrate provincial and city plans and programs within the regional plan, for approval of projects over PhP25061 million, and to implement Build-Operate-Transfer arrangements" (AusAid, 2005). Civil Service Practices. Sections 76-97 of the LGC indicates the assignment of responsibilities for human resource management and development in local governments. The LGC mandates that all local governments design and implement their own organizational structure and staffing pattern taking into consideration its service requirements and financial capability, subject to the minimum standards and guidelines prescribed by the Civil Service Commission. The chief executive of every local government unit is responsible for human resources and development in his unit in accordance with the Constitutional provisions on civil service, pertinent laws, and rules and regulations, including such policies, guidelines and standards as the Civil Service Commission may establish. The local chief executive has the power to employ emergency or casual employees for a period of six month paid on a daily wage or piecework basis and hired through job orders for local projects authorized by the Sanggunian concerned, without need of approval or attestation by the Civil Service Commission. The LGC does not allow the appointment in the career service of the local government if he is related within the fourth civil degree of consanguinity or affinity to the appointing or recommending authority. It mandates local chief executives to establish personnel selection board, tasked with assisting the local chief executive in the judicious selection or promotion of personnel. A Civil Service Commission representative sits in this board as an ex officio member. To regulate the wage bill, the LGC mandates that personnel expenditures should account for no more than 45-55 percent of local income. Service delivery and procurement- The passage of a clear procurement framework notwithstanding, implementation is still in its infancy and thus administration may still be weak. In particular, the 611 USD~ 46 Philippines Peso 130 following need to be monitored: (1) reports that congressional initiatives/pork barrel spending in district- level projects tend to be non-transparent as some officials are reported to pre-determine winning contractors; (2) dissemination of information regarding the implementation of the procurement act among stakeholders; (3) capacity building for BAC members as they lack experience and capability in dealing with procurement issues (World Bank 2005:35-36). 1.3.3 Demand Side Approaches to Improve Downward Accountability According to the LGC, accredited NGO representatives are empowered by the Code to sit in the pre- qualification, bids and awards committee; the school board; the health board, the local peace and order council, and the people's law enforcement board (see box 1 for additional details) Box 1 CSO Representation Local Health Board--Proposes budgetary allocations for health programs to the sanggunian, serves as an advisory committee, and sets technical and administrative standards for health. One seat in the board will be allocated to an NGO or private sector representative involve in the health sector Local Peace and Order Council--Assesses peace and order situation, formulates, monitors, implements plans and programs to improve peace and order and public safety in the locality. Three representatives of the private sector representing the academic, civic, religious, youth, labor, legal, business and media organizations will be appointed by the chairman of the board upon consultation with members of the POCs. Local School Board--Serves as an advisory committee on educational matters, determines supplementary budget for maintenance of schools, and authorizes disbursement of said budget. Three representatives from NGO or private sector (one from the local parents-teachers association, one from the local teachers organization and one from the organization of non-academic personnel of public schools) are to sit in the board. Local Pre-Qualification Bids and Awards Committee is the deciding authority on bids and awards for local infrastructure projects, although its jurisdiction does not include procurement of supplies, purchase of office equipment, renovations, and other local government unit disbursements. Two representatives of NGOs and POs seated in the LDC are allocated seats in the committee. Also a certified public accountant from the private sector designated by the local chapter of the Philippines Institute of Certified Public Accountants. (Arroyo, Sirker 2005). Source: Adapted from Iszatt (2004) and DILG (undated) Many civic groups sprouted across the country after the 1986 People Power Revolution. An association of election watchdogs made up the National Citizens Movement for Free Elections and volunteers in Abra province set up an NGO, the Concerned Citizens of Abra for Good Government (CCAGG), in response to the perceived need to monitor public expenditures. This network has been instrumental in the Textbook Count Project by mobilizing volunteers from its chapters nationwide. The Textbook Count Project began an effort to curb corruption in textbook procurement that resulted in a shortage of textbooks in many schools. To this end, civil society groups are helping the Department of Education monitor the delivery of the correct numbers of textbooks to district and high schools. Because of reforms in its procurement process in 2001, the Department of Education in the Philippines reports that its prices for textbooks have fallen by as much as 65 percent, allowing for the purchase of an additional 11 million to 16 million textbooks (Arroyo, Sirker 2005). KALAHI-CIDSS mentioned earlier includes as part of its program activities, simplified performance report cards through community billboards and reporting in public forums and independent grievance redress systems plus monitoring by civil society groups. It is creating a cadre of leaders who will total some 50,000 by the time the project ends and who will be equipped with the skills to prepare project proposals and to demand projects from their local politicians (Arroyo-Sirker 2005). 131 1.4 FISCAL DISCRETION AND ACCOUNTABILITY IN DECENTRALIZATION 1.4.1 Discretion Expenditure Assignment- Manasan (2004:5) submits that, except in education and health, the expenditure assignment is generally consistent with the decentralization theorem "each public service should be provided by the jurisdiction having control over the minimum geographic area that would internalize the benefits and costs of such provision".62 The Code transferred to LGUs expenditure functions formerly provided by the national departments of agriculture, education, culture and sports, environment and natural resources, health, public works and highways, transportation and communication, and social welfare (Pimmentel, 1993). The Code also permits LGUs to regroup into larger cooperative units when they deem appropriate. For example, there are cases of inter-LGU cooperation in coastal resource management, solid waste management, water supply development and distribution and construction of inter-municipal roads (World Bank 2004:4) Expenditure assignment is to be bottom-up as the LGC mandates each LGU to have a comprehensive multi-sectoral development plan formulated by its LDC and approved by its Sanggunian (Ramos, 2007). In this framework, the LDC assists the Sanggunian in setting the direction of economic and social development and in coordinating development efforts within its territorial jurisdiction. In more specific terms, the LDC is supposed to formulate long-term, medium-term, and annual socioeconomic development plans as well as the medium-term and annual public investment programs. Many cities and provinces report having long- and medium-term development plans as well as comprehensive land use plans. However, the only plan usually available in many of the smaller LGUs is the Annual Investment Plan (AIP). Most often, these plans consist of a simple listing of projects with corresponding budget estimates (World Bank 2004: 28). However, Manasan notes the lack of clarity in expenditure assignments still persists. In particular the LGC allows central government to finance devolved public works and infrastructure projects and other programs and services provided that these are "funded by the national government under the Annual General Appropriations Act, other special laws, pertinent executive orders, and those wholly or partially funded from foreign sources" and to augment the basic services and facilities provided by local government when these are inadequate or unavailable (Manasan 2004:6). These provisions are utilized by Congressmen to access pork barrel funds by "the simple act of inserting a special provision in the General Appropriations Acts which ordains that monies from such augmentation funds can only be released for 'projects that are identified by members of Congress" (Manasan 2004:7) These provisions thus have the effect of obfuscating the division of responsibilities between central and local governments. They provide the legal framework for both Congress and central agencies to retain funding for devolved services. Rough estimates of these funds are PhP 15 billion annually (World Bank 2004: 13). Such large funds at the discretion of the Congress diminish the expenditure discretion of the LGUs. Capuno et al (in Manasan 2004:7) also reveal that agencies like DepEd, DOH, and Department of Public Works and Highways (DPWH) continued to spend significant amounts on devolved activities between 1995-99. 62 In general, provinces were assigned responsibility for services whose catchment area cover more than one municipality while municipalities were made responsible for the delivery of frontline basic services. However, the national government--in particular the Department of Health (DOH)--retains budget for immunization, control of communicable diseases and provision of drugs and medicines to devolved facilities in hospitals in the National Capital Region. The Department of Education (DepEd) also continues to allocate budget for school buildings. (Manasan 2004:3-4) 132 The national government also tends to pass on to LGUs so-called unfunded mandates. For example, as mentioned earlier, LGUs are expected to provide budgetary support to many national agencies operating at the local level such as the police, fire protection, and local courts (World Bank 2004: 5). Thus, much local funding is tied in some way. An estimated 80% of local revenue is tied to specific centrally determined budget categories. In addition, the LGC requires local governments to set aside 20% of their IRA shares for a Development Fund to support local projects (Azfar etal 2001: 48). Revenue Assignment- The three main sources of revenue for the LGUs are: a. The Internal Revenue Allotment (IRA), the main fiscal transfer to local governments; b. Own Source Revenues (including borrowing) and smaller shared revenues; and c. expenditures by national government departments and agencies on devolved functions and activities (see table 1 for breakdown). The IRA revenues are discussed in further detail in the next section. In terms of own source revenues (OSR), in 2001, OSR accounted for 36 percent of total revenues of all local governments. 76 percent of OSR are from tax revenues of which 40 percent is from real property taxation. The World Bank (2005:19- 20) also finds that OSR has been declining by more than 20 percent since the 1980s while revenue from national government transfers have increased by the same amount. 133 Table 1: Composition of LGU Revenues and Funds for Devolved Functions: 2003 Size Billion Type of Revenue Pesos Size million $US Share Percentage I) LGU REVENUES 1.1. IRA 141.0 2,601.5 69.2 1.2. LGU shares in other taxes20 1.8 33.4 0.9 1.3. LGU Tax revenues 42.1 775.9 20.6 1.4. LGU Non-Tax revenues 13.1 241.2 6.4 1.5. Extraordinary receipts 21 1.8 33.2 0.9 1.6. Inter-local transfers 0.8 15.5 0.4 1.7. Borrowing 3.3 60.2 1.6 I) Total LGU Revenues (1.1 - 1.7) 203.9 3,761.1 100.0 II. NON-IRA FUNDING 2.1. Government Funded Programmes & Projects (GFPP) 4.1 75.7 16.2 2.2. Congressional Allocations to devolved functions (CA) 15.6 288.6 61.7 2.3. ODA Loans and Grants Funded transfers and spending on devolved functions (LGF) 5.4 100.0 21.3 2.4. Off Budget Funding of devolved functions (OBF) 0.2 3.7 0.8 II) Total Non-IRA funding of devolved functions (2.1-2.4) 25.4 468.0 100% III) Total funds used on devolved functions (I + II) 229.3 4,229.1 100% Part 1:Share of LGU revenues of total funds 89% Part 2: Share of Non-IRA Funding for devolved functions of total funds 11% Source: Soriano, et al, 2005 134 Non-tax revenues are mainly constituted by income of local public enterprises like public markets, slaughterhouses, public cemeteries. In terms of taxes, the LGC allows LGUs the option to levy additional taxes, fees, and charges, provided they are not explicitly prohibited in the Code. LGUs are also allowed to enter into international aid agreements. Some LGUs have used the authority to levy taxes and fees not specifically enumerated in the Code. For example, in the province of Bulacan, the governor and mayors jointly agreed with the legislators to levy garbage charges at the provincial level (World bank 2004: 18). At the same time, the LGC limits the power of local governments to set tax rates, in particular: (1) fixes the tax rate of some taxes like the community tax; (2) sets limits on the tax rates of the rest; and (3) mandates that tax rates can only be adjusted once in 5 years and by no more than 10 percent. (Manasan, 2004:20). The horizontal assignment of taxes across local government units also distorts collection incentives. For example, provinces retain only 35 percent of their real property tax collections (World Bank, 2005:18). Also local tax administration capacities are weak. Manasan (2004:21) notes that personnel assigned to the tax division tend to be not technically-equipped--very few units, for example, have certified public accountants thereby impairing audit and assessment capabilities. The World Bank (2005:22-25) finds weak administrative capacity for all major tax administration functions: registration, collection and compliance. The local government officials tend not to fully exercise tax powers assigned to them for fear of electoral reprisal. Manasan (2004:21) finds that provinces and cities have revised market values, the basis for real property tax collections, only once since 1991. This has been one of the main reasons for the steady decline in real property tax collection. Tax codes have also remained unrevised as local chief executives and members of the Sanggunian are reluctant to raise tax levels. Intergovernmental Transfer- The LGC increased local government's share of the national taxes, i.e., IRA, from a previously low of 11 percent to as much as 40 percent. The LGC provides local governments with a specific share from the national wealth exploited in their area (e.g., mining, fishery and forestry charges). In terms of revenue sharing, the code replaced the previous system of negotiated transfers, where local official lobbied national politicians, to automatic revenue-sharing. This enhanced local autonomy. At the same time, there is a high level of dependence of most localities in central transfers. High share of internal revenue allotment in local income (as seen in table 1) indicates that local expenditures are mostly financed by central transfers. IRA, the local government's guaranteed share in national wealth, is distributed across local government units based on the following shares: 23 percent to provinces, 23 percent to cities, 34 percent to municipalities and 20 percent to barangays. The respective shares of local governments, in turn, are determined on the basis of population (50 percent), land area (25 percent) and equal sharing (25 percent). Thus, provinces and municipalities combined receive 57% of revenue transfer but shoulder 92.5% of cost of devolved function. Highly urbanized cities and villages bear only 7.5% of the costs but receive 47% of the transfers (Embolura, 1994a, Manasan 2004:32). This skewed distribution reflects the reality that governors are seen as bigger threat than mayors (Eaton 2001). The IRA system has also exhibited instability in terms of timely and predictable releases and this has had the effect of further impinging on the fiscal discretion of localities. In periods of fiscal distress, central government has been known to flout the provision in the Code mandating the automatic release of IRA. Manasan (2004:27) tracks actual IRA fund releases between 1998 and 2004 and finds that in three of those seven years (viz., 1998, 1999, 2000)--notably years of fiscal crunch--central government appropriations did not match their obligations. 135 Categorical grants or non-IRA central transfers, while constituting a relatively small share of local revenue, account for a substantial share of local capital expenditures--providing further evidence of the depth of dependence on central transfers for local development financing. Based on shares in local government revenues, non-IRA transfers appear small. Table 1 decomposes local government revenues and shows that non-IRA funding constituted only 11 percent of total funds available for devolved functions in 2003. However, this figure remains significant because, unlike IRA transfers, a bigger portion of non-IRA transfers are allocated for capital expenditures (rather than personnel and overhead costs). Soriano et al (2005:16) find that as much as 89 percent of non-IRA transfers in 2003 were spent on capital outlay, while the same could be said of only 20 percent of IRA transfers. In absolute terms, capital expenditures from IRA and non-IRA amount to roughly the same: PhP 22.6B and PhP 28.2B, respectively. This would indicate that their developmental impact in terms of financing actual projects might be at par with one another. Non-IRA transfers also include the haphazard transfer of funds from the center to local governments as pork barrel funds. Allocation and utilization of pork barrel funds by legislators can be a source of corruption and commissions. Because many of the devolved national government agencies are accountable for outcomes in their respective areas, they deem it their responsibility to direct LGU activity, and most tend to make full use of Section 17(f) of the Code. Congressmen employ Section 17(f) to provide pork-barrel funds through special provisions in the General Appropriations Act which specify that such funds can only be released for "projects that are identified by members of Congress." This has resulted in a coordination failure between Congress and LGU officials (Loehr and Manasan 1999). Local Government Borrowing- The LGC lays the foundation for the development and evolution of more entrepreneurial-oriented local governments. For instance, it provides the foundations for local governments to enter into build-operate-transfer (BOT) arrangements with the private sector, float bonds, obtain loans from local private institutions, etc., all within the context of encouraging them to be "more business-like" and competitive in their operations in contradistinction to "traditional" government norms and operations. 1.4.2 Supply Side Approaches to Improve Downward Accountability Public Financial Management- Local public financial management systems can be substantively improved to enhance transparency and openness to the role of civil society actors in planning, monitoring and oversight functions (World Bank 2005). Financial management skills are low, leadership is lacking and there is weak oversight by national government agencies. Internal control over cash and property management needs major improvement to safeguard LGU assets and strengthen accountability over LGU resources. LGUS are currently not mandated to publish budget documents and proposed projects. Internal audit is a key weakness in LGU financial management, and its absence is one of the factors weakening LGU governance. Most LGUs do not have internal auditors. Apart from internal auditors, the Commission on Audit (COA) is a central agency that audits LGU spending. The COA's role in the financial management of LGUs requires re-examination to make it more effective from governance and accountability perspectives. 1.4.3 Demand Side Approaches to Improve Downward Accountability Citizen's participation in budgeting and expenditure- The provision of the Local Government Code of 1991 allows NGOs, POs, and the private sector to directly engage in governance by giving them a say, at the very least, on the use of official funds (Arroyo, Sirker 2005). The LGC also mandates the formation of local special bodies--described in (not limited to those in) Box 1 and tasked with formulating policy recommendations, devising developmental and sectoral plans and proposing measures that will guide the Sanggunian--with provisions for significant civil society representation. 136 Estralla and Iszatt (2004) edit a volume that synthesizes the experience of the NGO network Barangay- Bayan Governance Consortium (BBGC) in over five years of training communities in social technologies for participatory barangay development planning in over 2,500 villages in the Philippines. The book documents the experience of 11 of the 1,200 villages where the consortium initiated participatory village level planning. In the synthesis chapter, Rocamora (2004:335) reports that Consortium experience has led to substantive achievements in terms of generating new perspectives at the barangay level on how public monies are to be spent and has generated new capacities among political organization leaders, who have evolved from being experts in "expose and oppose" politics to being able to devise active propositions for solving commonly-perceived problems in engagements within state institutions. KALAHI-CIDSS, as part of its mandate, trains communities to formulate budgets for their own infrastructure projects. The program's activities include participatory prioritization and budgeting for community projects. In some villages, participation has reached 98 percent (Arroyo, Sirker 2005). But, Iszatt's (2002) review of the bigger picture--in terms of the extent implementation of the de jure framework--is quite bleak. She cites that the Department of Interior and Local Government's own study (Gabito, 2001) which finds representation of civil society in local special bodies of a "questionable" nature--33 percent of the civil society organizations were found to be either appointed or selected by the local chief executive. She also cites a study specific to participation in municipal development councils conducted by an NGO (PhilDHRRA, 2001), which finds LGUs resistant to the participation of independent NGOs for fear of being unable to control them and NGOs reluctant to work with LGUs for fear of being coopted. The inactivity of most LDCs is not puzzling if analyzed against the nature of the political incentives in a clientelistic political system. Local chief executives will expectedly be reluctant to yield their discretion over the allocation of fiscal resources; doing so runs them the risk of running afoul those who expect to have privileged access to public resources in exchange for their political support. Only 30-50 percent of local governments have LDCs in place (Azfar, et al, 2000 in WB, 2005:29) and less than one-third of the local governments have development plans with meaningful civil society participation (GOLD 2002 in WB, 2005:29). DILG officials interviewed indicate that of the boards listed in box 1, only local school boards tend to be operational at the municipal level. Local school boards are operational because they have an actual budget--they decide on the use of the Special Education Fund, which is comprised of the additional 1 percent tax on real property collected for the purpose of this fund. Meanwhile LDCs are convened--at the very least to pass the Annual Investment Plan, necessary to enact use of the IRA from central government. Public Hearings on budget information. There is lack of effective participation in the preparation and monitoring of the budget. In the LGUs visited by the World Bank study (World bank 2004), public was not typically invited in budget hearings conducted by the Sanggunian, although officially open to the public. Budget information was not widely disseminated. Public monitoring of budget execution is not possible since the budget document and other expenditure statements are not circulated. Civic monitoring of inter-governmental fiscal transfers. Azfar et al (2001) found that media (particularly radio and television) in Philippines was a principal source of information regarding local corruption. An online petition was set up by citizens to bring attention to the pork barrel funding and endorse its abolition. (http://www.petitiononline.com/fzf0205/petition.html) Impact of local governance and accountability on tax compliance. In the Philippines, the level of tax compliance is a core accountability signal that citizens can send to local governments regarding their performance. Citizens tend to show lower tax compliance towards corrupt and poorly performing local governments than towards more transparent and effective municipalities. The city of San 137 Fernando in the Philippines was able to increase its local revenue by around 60 per cent between 1996 and 201 after a reformist mayor institutionalized citizen participation, reduced bureaucratic expenses and implemented a range of highly visible projects. The city government gradually gained enough legitimacy to address tax issues successfully. It launched an information campaign on taxes, created incentives for citizens to pay taxes, overhauled the tax office, and created a one stop shop for various government services. Additionally it could address less popular measures like the creation a tax enforcement unit and a watchdog body that reported on tax avoidance of businesses (Guiza 2002). 138 REFERENCES Ahmad, Junaid, Shantayanan Devarajan, Stuti Khemani, Shekhar Shah. 2005. Decentralization and Service Delivery. World Bank Policy Research Working Paper 3603, World Bank, Washington, D.C. Arroyo, Dennis, and Karen Sirker. 2005. Stocktaking of Social Accountability Initiatives in the Asia and Pacific Region. Stock No. 37255. World Bank Institute. Washington D.C. Azfar, O, T. Gurgur, S. Kahkonen, A. Lanyi and P. Meagher. 2000. "Decentralization and Governance: An Empirical Investigation of Public Service Delivery in the Philppines. IRIS Center, University of Maryland: Maryland, USA. Azfar, Omar, Satu Kähkönen, and Patrick Meagher. 2001. Condition for Effective Decentralized Governance: A Synthesis of Research Findings. Working Paper No. 256. Center for Institutional Reform and the Informal Sector. University of Maryland. Behrman, Jere, Anil Deolalikar, and Lee-Ying Soon. 2002. "Promoting Effective Schooling through Education Decentralization in Bangladesh, Indonesia, and Philippines." Economic and Research Department Working Paper No. 23. Asian Development Bank. Philippines. Besley, Timothy, and Andrea Prat. 2005. "Handcuffs for the Grabbing Hand? Media Capture and Government Accountability", American Economic Review. Besley, Timothy, and Robin Burgess. 2002. "The Political Economy of Government Responsiveness: Theory and Evidence From India", The Quarterly Journal of Economics 177(4):1415-51. Blair, Harry. 2000. Participation and Accountability at the Periphery: Democratic Local Governance in Six Countries. World Development. Vol. 28, No. 1. pp 21-39, U.K. Bossert, Thomas J., and Joel C. Beauvais. 2002. " Decentralization of Health Systems in Ghana, Zambia, Uganda, and the Philippines: a Comparative Analysis of Decision Space." Health Policy and Planning.17(1):14-31. Oxford University Press, New York. Brillantes, Alex, and Donna Moscare. 2002. "Decentralization and Federalism in the Philippines: Lesson from Global Community." Discussion paper presented at the International Conference of the East West Center, July 1-5, Kuala Lumpur. Brillantes, Alex. 1999. "Decentralization, Devolution and Development in the Philippines." UMP- Asia Occasional Paper No. 44. Brinkerhoff, Derick, and Arthur A. Goldsmith. 2003. "How Citizens Participate in Macroeconomic Policy: International Experience and Implications for Poverty Reduction." World Development. Vol. 31, No. 4, pp. 685-701, Great Britain. Chua, Y. 2001. "Local Fiscal Administration". In C. Balgos, ed, Investigating Local Governments: A Manual for Reporters. Philippine Center for Investigative Journalism: Quezon City, Philippines. Cuarteros, G. 2006. "Preventive Suspensions, Preventing Re-Election". Accessed from http://ipd.org.ph/main/index.php?option=com_content&task=view&id=50&Itemid=1 on June 20, 2007. David, R. 2001. "Political Parties in the Philippines" in R. David, ed., Reflections on Sociology and Philippine Society. Manila: University of the Philippine Press, 2001. 139 Department of Interior and Local Government (DILG) with Urban Resources and the Evelio Javier Foundation.. 2002. A Study on People's Participation in the Local Development Councils. DILG: Quezon City, Philippines Eaton, Kent. 2001. "Political Obstacles to Decentralization: Evidence from Argentina and Philippines." Development and Change. Oxford: Blackwell Publishers. Emboltura, A. 1994. Searching for Equity: The IRA Debate. Legislative Features 2(1): 5-8. Etemadi, Felisa. 2000. "Civil Society Participation in City Governance in Cebu City." Environment and Urbanization. Volume 12, No. 1, Sage Publications. Gabito, M. 2001. "Status of Compliance with Department of Interior and Local Government Memorandum Circular No. 2001-89", Appendix D in documentation of 10.10.10 Celebration: CSO Conference on Participatory Local Governance, Traders Hotel, Manila, October 7-9. Government of Philippines. 1991 "General Primer- The Local Government Code." The Government of Philippines. Guiza, E. (2002). Transforming a Ravaged City into a Model in Local Revenue Generation: The case of San Fernando, Pampanga. Philippines. Asian Institute of Management, Philippines. Hedman, E. and J. Sidel. 2000. Philippine Politics and Society in the Twentieth Century: Colonial Legacies, Post-Colonial Trajectories, pp 1-12. Routledge: London, UK Hutchcroft, P. 2003. "Paradoxes of Decentralization: The Political Dynamics Behind the 1991 Local Government Code of the Philippines," in KPI Yearbook 2003, ed. Michael H. Nelson (manuscript) ___________. 2001. "Centralization and Decentralization in Administration and Politics: Assessing Territorial Dimensions of Authority and Power". Governance: An International Journal of Policy and Administration. Vol 14, No1, January 2001, pp 23-53. Hutchcroft, P. and J. Rocamora.(2003). "Strong Demands and Weak Institutions: The Origins and Evolution of the Democratic Deficit in the Philippines", Journal of East Asian Studies 3(2003), pp 259-292. Iszatt, N. 2002. "Legislating for Citizens' Participation in the Philippines". Logolink: Manila, Philippines. Iszatt, N. and M Estrella (eds). 2004. Beyond Good Governance: Participatory Democracy in the Philippines. Institute for Popular Democracy: Quezon City, Philippines. Manasan, R. 2004. "Local Public Finance in the Philippines: In Search of Autonomy with Accountability". PIDS Discussion Paper Series no. 2004-42. December 2004. Philippine Institute for Development Studies: Makati, Philippines. Martinez, A. 2007. "From the Heart of Manila to the Country's History Books". Sunday Inquirer Magazine. Pp 2, 13. June 24, 2007 Pimentel, Aquilino. 1993. The Local Government Code of 1991: The Key to National Development. Metro Manila: Cacho Publishing House, Inc. Ramiro, Laurie, Fatima A Castillo, Tessa Tan-Torres, Cristina E Torres, Josefina G Tayag, Rolando G Talampas, and Laura Hawken. 2001. Community Participation in Local Health Boards in a Decentralized Setting: Cases from the Philippines. Health Policy and Planning, Volume 16, Number 2, pp. 61-69(9) Oxford University Press. Ramos, Charmaine G. 2007. "Enabling Environment Assessment for Social Accountability in Local Governance in the Philippines." Mimeo. 140 Ribot, Jesse. 2004. Waiting for Democracy: The Politics of Choice in Natural Resource Decentralization. Washington D.C.: World Resource Institute. Romero, S.E. 1997. "Issues and Trends in NGO-PO Participation in Local Governance". In Working Group, Working Papers. NGO-PO Perspectives for the Local Government Code Review. Institute of Politics and Governance: Quezon City, Philippines Shah, Anwar. 2006. Local Governance in Developing Countries. The World Bank. Washington D.C. Soriano, M.C., J. Steffensen, E. Makayan and J. Nisperos. .2005. "Assessment of Non-IRA Transfers and Other Funds for Devolved Services in the Philippines: Final Main Report. Manuscript Stevens, Andrew (February 12 2006). www.citymayors.com Local government in The Philippines Government System Based on US and Spanish Models under Review. Trujillo, Horatio. 2000. "Local Government Revenue Strategies: Recognizing Pitfalls in the Intergovernmental Fiscal Reform. Experiences of China and the Philippines." Project Report. Center for Institutional Reform and the Informal Sector. University of Maryland at College Park. World Bank. 2004. "Decentralization in the Philippines: Strengthening Local Government Financing & Resource Management in the Short Term". World Bank Economic Report. Washington, D.C. World Bank. 2005. "Decentralization in the Philippines: Strengthening Local Government Financing and Resource management in the Short term". Report No. 26104-PH. Asian Development Bank Philippine Country office and East Asia and Pacific Region, The World Bank. 141